Talk:Book - Chemical embryology 1 (1900)

From Embryology




M. A., Ph.D.

Fellow ofGonville & Cains College, Cambridge, and

University Demonstrator in


Volume One

New York: The Macmillan Company Cambridge, England: At The University Press


Printed In Great Britain

JOSEPHO NEEDHAM in Univ. Aberdon. Anat. olim Professori


Schol. Undel. olim Praeposito


Coll. Gonv. et Caii olim Custodi


Artis chemicae ad animantia spectantis in Univ. Cantabrig. Professori


hanc suam disquisitionem


sacram voluit

For more, and abler operations are required to the Fabrick and erection of Living creatures, than to their dissolution, and plucking of them down : For those things that easily and nimbly perish, are slow and difficult in their rise and complement.

William Harvey, Anatomical Exercitations concerning the generation of living creatures, London, 1653, Ex. XLi, p. 206.

That discouraging Maxime, Nil dictum quod non dictum prius, hath little room in my estimation, nor can I tye up my belief to the Letter of Solomon ; I do not think, that all Science is Tautology; these last Ages have shown us, what Antiquity never saw; no, riot in a dream.

Joseph Glanville, Scepsis Scientifica, an essay of the vanity of dogmatizing, and confident opinion, London, 1 66 1, Chap. XXII.



Prolegomena page 2


The Theory of Chemical Embryology

Philosophy, Embryology, and Chemistry 7

The Historical Perspective 10

Obstacles to Chemical Embryology 1 3

The Stumbling-block of Hormism 1 4

Finalism as a Rock of Offence 1 6

Organicism as an Occasion of Falling 25

Organicism and Emergence 3^

Neo-Mechanism as a Theory for Chemical Embryology 32


The Origins of Chemical Embryology

Preliminary Note 41

Section i . Embryology in Antiquity 44

I • I . Non-Hellenic Antiquity 44

1-2. Hellenic Antiquity; the Pre-Socratics 50

1-3. Hippocrates; the Beginning of Observation 53

1-4. Aristotle ^g

1-5. The Hellenistic Age 77

1-6. Galen 85

Section 2. Embryology from Galen to the Renaissance gi

2-1. Patristic, Talmudic, and Arabian Writers gi

2-2. St Hildegard; the Lowest Depth g^

23. Albertus Magnus gy

2-4. The Scholastic Period 103

2-5. Leonardo da Vinci 1 07

2-6. The Sixteenth Century; the Macro-iconographers I lO

Section 3. Embryology in the Seventeenth and Eighteenth Centuries page 125

3-1. The Opening Years of the Seventeenth Century 1 25

32. Kenelm Digby and Nathaniel Highmore 1 29

3-3. Thomas Browne and the Beginning of Chemical Embryology 135

3-4. William Harvey 138

35, Gassendi and Descartes; Atomistic Embryology 1 56

36. Walter Needham and Robert Boyle 1 60 3-7. Marcello Malpighi; Micro-iconography and Preformationism 166 3-8. Robert Boyle and John Mayow 169 3-9. The Theories of Foetal Nutrition 176 3-10. Boerhaave, Hamberger, Mazin 1 82 3-11. Albrecht v. Haller and his Contemporaries 1 88 3-12. Ovism and Animalculism 1 99 3- 1 3. Preformation and Epigenesis 205 3-14. The Close of the Eighteenth Century 215 3-15. The Beginning of the Nineteenth Century 220


General Chemical Embryology

Preliminary Note

Section i. The UnfertiUsed Egg as a Physico-chemical System

1 . Introduction

2. General Characteristics of the Avian Egg

3. The Proportion of Parts in the Avian Egg

4. Chemical Constitution of the Avian Egg as a Whole

5. The Shell of the Avian Egg

6. The Avian Egg-white

7. The Avian Yolk

8. The Avian Yolk-proteins

9. The Fat and Carbohydrate of the Avian Yolk

10. The Ash of the Avian Egg

1 1 . General Characteristics of non-Avian Eggs

12. Egg-shells and Egg-membranes

13. Proteins and other Nitrogenous Compounds

14. Fats, Lipoids, and Sterols

15. Carbohydrates

16. Ash


Section 2. On Increase in Size and Weight

page 368 368

2-1. Introduction

2-2. The Existing Data 369

2-3. The General Nature of Embryonic Growth 383

2-4. The Empirical Formulae 389

2-5. Percentage Growth-rate and the Mitotic Index 399

2-6. Yolk-absorption Rate 405

2 '7. The Autocatakinetic Formulae 408

2-8. Instantaneous Percentage Growth-rate 420

2-9. Growth Constants 434

2-10. The Growth of Parts 440

2-1 1. Variability and Correlation 455

2-12. Explantation and the Growth-promoting Factor 460

2-13. Incubation Time and Gestation Time 470

2-14. The Effect of Heat on Embryonic Growth 498

2-15. Temperature Coefficients 503

2 • 1 6. Temperature Characteristics 515

2-17. The Effect of Light on Embryonic Growth 533

2-i8. The Effect of X-rays and Electricity on Embryonic Growth 536

2-19. The Effect of Hormones on Embryonic Growth 538

Section 3.

3-1 3-2

33 3-4 3-5 3-6

37 3-8


On Increase in Complexity and Organisation The Independence of Growth and Differentiation Differentiation-rate

Chemical Processes and Organic Form The Types of Morphogenetic Action Pluripotence and Totipotence Self-differentiation and Organiser Phenomena Functional Differentiation Axial Gradients Organised and Unorganised Growth

3-10. Chemical Embryology and Genetics

541 541 544 552 559 567 570 580 582 606 608


Section 4. The Respiration and Heat-production of the Embryo 615

4- 1 . Early Work on Embryonic Respiration 6 1 5

4-2. Respiration of Echinoderm Embryos in General 623

4-3. Rhythms in Respiratory Exchange 64 1



Section 4-4. Heat Production and Calorific Quotients of Echinoderm page 649 Embryos

4-5. Respiration of Annelid, Nematode, Rotifer, and Mollusc 659 Embryos

46. Respiration of Fish Embryos 665

47. , Respiration of Amphibian Embryos 67 1 4-8. Heat-production of Amphibian Embryos 682 49. Respiration of Insect Embryos 687 4- 1 o. Respiration of Reptile Embryos 692 4-11. Respiration of Avian Embryos in General 693 4-12. Heat-production of Avian Embryos 7^4 4-13. Later Work on the Chick's Respiratory Exchange 7^8 4- 1 4. The Air-space and the Shell 7 1 9 4-15. Respiration of Mammalian Embryos 726 4" 1 6. Heat-production of Mammalian Embryos 732 4-17. Anaerobiosis in Embryonic Life 742

^, 4" 1 8. Metabolic Rate in Embryonic Life 746

4-19. Respiratory Intensity of Embryonic Cells fn wVro 755

4-20. Embryonic Tissue-respiration and Glycolysis 758

4-21. The Genesis of Heat Regulation 772

4-22. Light-production in Embryonic Life 776

Section 5. Biophysical Phenomena in Ontogenesis 777

5-1. The Osmotic Pressure of Amphibian Eggs 777

5-2. The Genesis of Volume Regulation , 786

53. The Osmotic Pressure of Aquatic Arthropod Eggs 790

54. The Osmotic Pressure of Fish Eggs 793

5-5. Osmotic Pressure and Electrical Conductivity in Worm and 799 Echinoderm Eggs

5-6. The Osmotic Pressure of Terrestrial Eggs 8l2

5-7. Specific Gravity 820

5-8. Potential Differences, Electrical Resistance, Blaze Currents 825 and Cataphoresis

5-9. Refractive Index, Surface Tension and Viscosity 833

Section 6. General Metabolism of the Embryo 839

6-1. The j&H of Aquatic Eggs 839

6-2. The j&H of Terrestrial Eggs 855

63. rH in Embryonic Life 865

6-4. Water-metabolism of the Avian Egg 870


Section 6-5. Water-content and Growth-rate page 883

6-6. Water-absorption and the Evolution of the Terrestrial Egg 889

6-7. Water-metabolism in Aquatic Eggs 906

6-8. The Chemical Constitution of the Embryonic Body in Birds 911 and Mammals

69. Absorption-mechanisms and Absorption-intensity 917

6- 10. Storage and Combustion; the Plastic Efficiency Coefficient 934

6-1 1. Metabolism of the Avian Spare Yolk 939

6-12. Maternal Diet and Embryonic Constitution 943

Section 7. The Energetics and Energy-sources of Embryonic Development 946

7-1. The Energy Lost from the Egg during Development 946

7-2. Energy of Growth and Energy of Differentiation 956

7-3. The Relation between Energy Lost and Energy Stored 962

7-4. Real Energetic Efficiency • 969

7-5. Apparent Energetic Efficiency 972

7*6. Synthetic Energetic Efficiency 98 1

7-7. The Sources of the Energy Lost from the Egg 986

Sections. Carbohydrate Metabolism 1000

8-1. General Observations on the Avian Egg lOOO

8-2. Total Carbohydrate, Free Glucose, and Glycogen lOOl

8-3. Ovomucoid and Combined Glucose 1 007

8-4. Carbohydrate and Fat 1014

8-5. The Metabolism of Glycogen and the Transitory Liver 1018 8-6. Free Glucose, Glycogen, and Insulin in the Embryonic Body IO29 8-7. General Scheme of Carbohydrate Metabolism in the Avian Egg 1035

8-8. Embryonic Tissue Glycogen 1 036

8'9. Embryonic Blood Sugar 1039

8- 10. Carbohydrate Metabolism in Amphibian Development 1043

8*i I. Carbohydrate Metabolism of Invertebrate Eggs I047

8-12. Pentoses 1 05 1

8-13. Lactic Acid 1 051

8-14. Fructose 1054

Section 9. Protein Metabolism 1055

9* I. The Structure of the Avian Egg-proteins before and after 1055


9*2. Metabolism of the Individual Amino-Acids I059

9-3. The Relations between Protein and non-Protein Nitrogen 1065

9-4. The Accumulation of Nitrogenous Waste Products 10 76

Section 9-5. Protein Catabolism page

9-6. Nitrogen-excretion; Mesonephros, Allantois, and Amnios

9-7. The Origin of Protective Syntheses

9*8. Protein Metabolism of Reptilian Eggs

9-9. Protein Metabolism of Amphibian Eggs

9' 10. Protein Metabolism in Teleostean Ontogeny

9-11. Protein Metabolism in Selachian Ontogeny

9* 1 2. Protein Metabolism of Insect, Worm, and Echinoderm Eggs

9-13. Protein Utilisation in Mammalian Embryonic Life

9-14. Protein Utilisation of Explanted Embryonic Cells

9-15. Uricotelic Metabolism and the Evolution of the Terrestrial Egg

Section 10. The Metabolism of Nucleins and Nitrogenous Extractives

[O-i. Nuclein Metabolism of the Chick Embryo

IO-2. The Nucleoplasmatic Ratio

[0-3. Nuclein Synthesis in Developing Eggs

[0-4. Creatinine, Creatine, and Guanidine

Section 11. Fat Metabolism

I • I . Fat Metabolism of Avian Eggs

1-2. Fat Metabolism of Reptilian Eggs

I •2- Fat Metabolism of Amphibian Eggs

I -4. Fat Metabolism of Selachian Eggs

1-5. Fat Metabolism of Teleostean Eggs

1-6. Fat Metabolism of Mollusc, Worm, and Echinoderm Eggs

1-7. Fat Metabolism of Insect Eggs

1-8. Combustion and Synthesis of Fatty Acids in Relation to Metabolic Water

1 1 -9. Fat Metabolism of Mammalian Embryos

Section 12. The Metabolism ofLipoids, Sterols, Cycloses, Phosphorus and Sulphur

1 2- 1. Phosphorus Metabolism of the Avian Egg

12-2. Tissue Phosphorus Coefficients

I2'3. Choline in Avian Development

12-4, The Metabolism of Sterols during Avian Development

1 2*5. The Relation between Lipoids and Sterols; the Lipocytic Coefficient

12-6. Cycloses and Alcohols in Avian Development

12-7. Sulphur Metabolism of the Avian Egg

12-8. Phosphorus, Sulphur, Choline, and Cholesterol in Reptile Eggs

Section 12-9. Lipoids and Sterols in Amphibian Eggs page 1237

12-10. Lipoids, Sterols, and Cycloses in Fish Eggs 1239

i2'ii. Phosphorus, Lipoids and Sterols in Arthropod Eggs 1241

12-12. Phosphorus, Lipoids, and Sterols in Worm and Echinoderm 1243


12-13. Lipoids and Sterols in Mammalian Development 1252

Section 13. Inorganic Metabolism 1255

[ 3- 1 . Changes in the Distribution of Ash during Avian Development 1 255

32. Calcium Metabolism of the Avian Egg 1260

33. Inorganic Metabolism of other Eggs 1 268 3-4. The Absorption of Ash from Sea-water by Marine Eggs 1 27 1 3-5. The Ani on/Cation Ratio 1 2 74 3-6. Inorganic Metabolism of Mammalian Embryos 1277 3-7. Calcium Metabolism of Mammalian Embryos 1285

Section 14. Enzymes in Ontogenesis 1289

4-1. Introduction 1 289

4-2. Enzymes in Arthropod Eggs 1290

4-3. Enzymes in Mollusc, Worm, and Echinoderm Eggs 1293

4-4. Enzymes in Fish Eggs 1 295

4-5. Enzymes in Amphibian Eggs 1300

4-6. Enzymes in Sauropsid Eggs 1303

4-7. Changes in Enzymic Activity during Development 1307

4-8. Enzymes of the Embryonic Body 1 3^0

4-9. Enzymes in Mammalian Embryos 13 12

4-10. The Genesis of Nucleases 1326

4-11. Foetal Autolysis 1 329

Section 15. Hormones in Ontogenesis 1335

5-1. Introduction 1335

5-2. Adrenalin ^337

53. Insulin 1342

5-4. The Parathyroid Hormone 134^

5-5. The Hormones of the Pituitary 134^

5-6. Secretin 134^

5-7. Thyroxin 134^

5-8. Oestrin and other Sex Hormones • 1353



Section i6. Vitamins in Ontogenesis page 1359

[6-1. Vitamin A 1359

[6-2. Vitamin B 1360

[6-3. Vitamin C 1 360

[6-4. Vitamin D 1 360

[6-5. Vitamins in Mammalian Development 1 363

[6-6. Vitamin E 1 365

Section 17. Pigments in Ontogenesis 1368

[7-1. The Formation of Blood Pigments 1 368

[7-2. The Formation of Bile Pigments 137^

[7-3. The Formation of Tissue Pigments 1 375

[7-4. The Pigments of the Avian Egg-shell 137^

[7-5. The Pigments of the Avian Yolk 1378

[7-6. Egg-pigments of Aquatic Animals 1380/

[7-7. Melanins in Ontogenesis 13^^

Section 18. Resistance and Susceptibility in Embryonic Life 1383

•I. Introduction 1 3^3

•2. Standard Mortality Curves 1 3^3

[8-3. Resistance to Mechanical Injury ^3^5

$-4. Resistance to Thermal Injury 1 388

5-5. Resistance to Electrical Injury ^392

[8-6. Resistance to Injury caused by Abnormal j&H 1 397

5-7. Resistance to Injury caused by Abnormal Gas Concentrations 1 399

(non-Avian Embryos)

!-8. Critical Points in Development 1 409

!-g. Resistance to Injury caused by Abnormal Gas Concentrations 1 4 1 4 (Avian Embryos)

>-io. Resistance to Injury caused by Toxic Substances 1420

••I I. Resistance to Injury caused by X-rays, Radium Emanation, 1 43 1 and Ultra-violet Light

Section 19. Serology and Immunology in Embryonic Life 1444

ig-i. Antigenic Properties of Eggs and Embryos ^444

19-2. The Formation of Natural Antibodies 1446

19-3. The Natural Immunity of Egg-white ^447

19-4. Inheritance of Immunity in Oviparous Animals HS^

19-5. Serology and Pregnancy 1452

19-6. Resistance of the Avian Embryo to Foreign Neoplasms 1 454

Section 20. Biochemistry of the Placenta page 14.^6

20-1. Introduction 1 45"

20-2. General Metabolism of the Placenta 145^

20-3. Placental Respiration 1 46 1

20-4. Nitrogen Metabolism of the Placenta 1 462

20-5. Carbohydrate Metabolism of the Placenta 14^9

20-6. Fat and Lipoid Metabolism of the Placenta 1472

20-7. Placental Enzymes 1481

Section 21. Biochemistry of the Placental Barrier 1485

2 1 • I . The Autonomy of the Foetal Blood 1 4^5

21-2. Evolution of the Placenta -4^7

21-3. Histotrophe and Haemotrophe 149^

21-4. Mesonephros and Placenta 1 493

21-5. Colostrum and Placenta ^497

21-6. Placental Transmission and Molecular Size 1497

21-7. QuaHtative Experiments on Placental Permeability 1 505

21-8. The Passage of Hormones 15^^

2 1 -9. Factors Governing Placental Transmission 15^2

2I-IO. Quantitative Experiments on the Passage of Nitrogenous 15 14

Substances 2 1 -I I. Quantitative Experiments on the Passage of Phosphorus, Fats, 1520

and Sterols 2i'i2. Quantitative Experiments on the Passage of Carbohydrates 1525

2i*i3. Quantitative Experiments on the Passage of Ash 1 52 7

21-14. The Passage of Enzymes 15^9

2i*i5. The Unequal Balance of Blood Constituents 1530

Section 22. Biochemistry of the Amniotic and Allantoic Liquids 1534

22-1. Introduction 1 534

22-2. Evolution of the Liquids ^535

22-3. Avian Amniotic and Allantoic Liquids 1 537 22-4. Amount and Composition of Mammalian Amniotic and Allan- 1539

toic Liquids

22-5. Maternal Transudation and Foetal Secretion 154^

22-6. Interchange between Amniotic and Allantoic Liquids 15^2

22-7. Vernix Caseosa 1 5^4

Section 23. Blood and Tissue Chemistry of the Embryo 1565

23-1. Blood 1565

23-2. Lung 1 57 1

23-3. Muscle 1574

Section 23-4.



Nervous Tissue


Connective Tissue




Sense Organs


Intestinal Tract

Section 24.

Hatching and Birth




Hatching Enzymes


Osmotic Hatching




Hatching of the Avian Egg


MammaUan Birth

page 1577

1583 1592 1593 1594 1594

1595 1595 1595 1600

1602 1602 1605


The Two Problems of Embryology 1 6 1 3

The Cleidoic Egg and its Evolution 16 1 3

Chemical Synthesis as an Aspect of Ontogeny 1 623

Biochemistry and Morphogenesis 1 624

Transitory Functions in Embryonic Life 1627

The Theory of Recapitulation 1629

Recapitulation and Substitution ' 1632

Chemical Recapitulation 1638

Provisional Generalisations for Chemical Embryology 1 647 The Organisation of Development and the Development of Organisation 1659

The Future of Embryology 1 664



i. Normal Tables of Magnitudes in Embryonic Growth 1669

ii. A Chemical Account of the Maturation of the Egg-cell 1679

iii. The Chemical Changes during the Metamorphosis of Insects (by 1685

Dorothy Needham)

iv. The Development of the Plant Embryo from a Physico-chemical View- 1 7 1 1

point (by Muriel Robinson)


Bibliography and Author-Index


Index Animalium

1725 1971 2013



William Harvey frontispiece

I. Primitive methods of incubation : (A) Egyptian, (B) Chinese facing page 46 II. The oldest known drawing of the Uterus (gth century) . „ „ 82

III. Illustration from the Liber Scivias of St Hildegard (ca.

1150A.D.) jj J3 96

IV. A page from Leonardo da Vinci's Anatomical Notebooks

(ca. 1490 A.D.) ........„„ 108

V. Illustration from the De Formatione Ovi et Pulli of Fabricius

(1604) „ „ 116

VI. Illustration {rom Highmore's History of Generation {16^1) . „ „ 134

VII. Illustrations from Malpighi: i)e Or;o in^M^a/o (1672) . ,, „ 168

VIII. Reaumur's Illustration of his Incubators (1749) . . „ „ 198

IX. Microphotograph of the yolk of the hen's egg at the time

of laying, to show the vitelline globules . . . . ,, ,, 236

X. Microphotograph of the yolk of the hen's egg, not yet

liberated from the ovary, to show the stratification . . „ ,, 288


The frontispiece of William Harvey's Generation of Animals ( 1 65 1 ) ; Zeus liberating living beings from an egg . frontispiece

XI. Microphotograph of the yolk of the hen's egg at the eleventh day of incubation, showing its heterogeneous state .......... facing page 836

XII. Microphotograph of the yolk of the hen's egg at the second

day of incubation, showing the cholesterol esters . . „ ,,1218


An embryological investigation in the eighteenth century frontispiece


27. Ash of the avian egg ....

34. Distribution of amino-acids in egg-proteins

47. Ash content of egg .

195. Enzymes in the hen's egg

199. Enzymes in the human embryo

201. Enzymes in the pig embryo

220. Placental enzymes .

227, Passage of substances through the placenta Appendix 1, Table 3. Embryonic growth of the hen

facing page


»5 55


.J 55


55 55


55 55


55 55


55 55


55 5'


55 5?


Acknowledgements of Indebtedness

Those who have assisted me in the preparation of this work are so numerous that it is impossible to mention them all by name. Its original impetus was derived from a discussion with Professor Sir F. G. Hopkins in 1923 on the observation of Klein that inositol, though absent from the undeveloped hen's egg, was present in considerable quantity at hatching; and throughout the period of preparation his encouragement, help, and advice were never-failing. I have derived great benefit from the discussion of various points with Miss Marjory Stephenson, M. Louis Rapkine, Dr R. A. Fisher, and my wife. Professor J. T. Wilson has been repeatedly helpful to me on anatomical points, and in the Zoological Laboratory I was always sure of obtaining expert advice from Mr James Gray, Mr J. T. Saunders, Mr C. F. A. Pantin and Dr Eastham. I have relied much upon the kindness and wide biological knowledge of Dr D. Keilin and Dr F. H. A. Marshall. As regards the historical chapters, I am most grateful to Dr Charles Singer, who annotated them with valuable comments, and to Professor R. C. Punnett who placed unreservedly at my disposal his knowledge of the history of generation, and his library of old and rare biological books. To Dr Arthur Peck I am indebted for the correction of my Greek, and it was Professor A. B. Cook who guided me to the embryology of the ancients. Without the assiduous backing of Mr Powell, the Librarian of the Royal Society of Medicine, and his assistants, I should have dealt much more inadequately than I have with the papers which cannot be consulted in Cambridge. I have also to thank the administrators of the Thruston Fund of Gonville and Caius College for a grant which was devoted to incidental expenses. For the indexes I wish to thank Miss Helen Moyle, and for other services which have made the book possible, Mrs V. Townsend. My thanks are also due to the Editors of the following journals: Biochemical Journal, Journal


of Experimental Biology, Biological Reviews, Science Progress, and the Monist, for permission to reprint passages from papers. I must record my gratitude to the following friends, who very kindly read through and criticised the proofs of the various sections:

Part I

Professor A. E. Boycott Dr J. H. Woodger

Part II

Professor R. C. Punnett Dr Charles Singer Dr Reuben Levy Dr Arthur Peck Sir William Dampier Professor A. B. Cook The Rev. W. Elmslie Professor F. M. Cornford

Part III


1 Professor R. H. A. Plimmer Mr J. B. S. Haldane

2 Dr Samuel Brody Mr James Gray Dr E. N. Willmer

3 Mr G. R. de Beer

Mr C. H. Waddington Mr J. B. S. Haldane

4 Dr D. Keilin Professor Munro Fox

5 Mr T. R. Parsons Dr Malcolm Dixon

6 M. Louis Rapkine Mr C. Forster Cooper

7 Miss Marjory Stephenson M. Louis Rapkine

Dr D. Keilin

8 Dr Eric Holmes

Dr Bruce Anderson & Mrs Margaret Whetham Anderson




Dr Dorothy Jordan Lloyd

Professor J. Murray Luck

Mr C. Forster Cooper


Mile Eliane LeBreton


Professor J. B. Leathes


Dr Irvine Page


Dr Elsie Watchorn


Dr Barnet Woolf

MrJ. B. S. Haldane


Dr Howard Florey


Dr Leslie J. Harris

Dr A. L. Bacharach


Dr Howard Whittle


Mr C. F. A. Pantin

Professor A. R. Moore & Mrs Moore


Dr John Hammond


Dr St G. Huggett


Dr Arthur Walton


Dr Barbara Holmes


Dr F. H. A. Marshall


Professor L. G. M. B. Becking

Dr D. Keilin

Dr G. S. Carter

Professor Lancelot Hogben

Mr G. R. de Beer

Professor A. R. Moore & Mrs Moore

Appendix III Dr L. E. S. Eastham

I am indebted to the Master of Gonville and Caius College for permission to reproduce the portrait of William Harvey (attributed to Rembrandt) in the Senior Combination Room. Finally, I am glad to record here my gratitude to the StafTof the Cambridge University Press for the unremitting care which they gave to my book during the course of its preparation.

J. N.

Note: The use of the shortened and (&) indicates collaboration between two or more authors.


The Sciences, unlike the Graces or the Eumenides, are not limited in number. Once born, they are immortal, but, as knowledge increases, they are ever multiplying, and so great is now the dominion of the scientific mind that every few years sees a new one brought into the world. Some spring, fully armed, from the brains of one or two men of genius, but most of them, perhaps, come only gradually to their full development through the labours of very many obscure and accurate observers.

If the analogy may be permitted, physico-chemical embryology has so far been living an intra-uterine existence. Its facts have been buried in a wide range of scientific journals, and its theories have lain dormant or in potentia in reviews of modest scope. Physicochemical embryology has, indeed, arrived at the stage immediately priox to birth, and all it needs is a skilful obstetrician, for, when once it has reached the light of day and has passed for ever out of the foetal stage, it will be well able to take care of itself. This obstetrical task is that which I have chosen and obviously enough it divides into three principal heads: first, to collect together out of all the original papers on the subject the facts which are known about the physico-chemical basis of embryonic development; second, to relate these facts to each other and to the facts derived from the labours of investigators in morphological embryology and " Entwicklungsmechanik," and, third, to ascertain whether, from what is at present known, any generally valid principles emerge.

I may as well say at the outset that in order to do this certain arbitrary boundary-lines are inevitable. The following arrangement has been adopted. Chronologically speaking, the prelude to all embryonic development is the maturation of the egg-cell, but this is not strictly embryology, and so has been relegated to an appendix. The egg-cell as a physico-chemical system is dealt with at the opening of Part III, and thereafter the physico-chemical aspects of development follow in order. No mention will be made of fertilisation, for this has been treated exhaustively by other writers (Lillie, Dalcq) and, after all, embryology presupposes fertiUsation whether natural or artificial. Nor in later chapters will any complete treatment be given of the events going on in the maternal organism during pregnancy : for the present purpose the discussion will go as far into the mother as the placenta but no farther. Again, hatching or birth will put an end to the discourse as to the foetal state itself, save that, in the cases of animals which hatch before the yolk-sac is absorbed, their embryonic life is assumed to end when they first take food for themselves. Appendices are added dealing with the plant embryo and the insect pupa, which, in the later stages of metamorphosis, have points both of resemblance to and of difference from the growth of the embryo. It is natural to hope that the outcome of all this labour may be an increase of interest among biologists in this section of their domain, and a great accession to the number of those investigators who devote their energies to actual experiments in this new field.

For it must be confessed that it is a new field. It has been opened up in very gradual stages: fitful and sporadic experiments on the constitution of embryonic tissues in the seventeenth century, a gradual growth of knowledge about the chemical composition of eggs in the eighteenth, a big increase of activity in the early nineteenth; d'lTxiug which appear the first observations on the physico-chemical changes taking place in the embryo during its development, and then in our own time a mass of very widely scattered work bringing the subject up to the "obstetrical" stage. Such a work as this, in my opinion, should not be compared with laboratory experiments in a derogatory sense, for, while it is true that facts are the ultimate court of appeal in any scientific discussion, yet at the same time the number of investigators has grown to such extraordinary proportions in this century that some danger exists lest we should be so busily engaged in accumulating new facts as to be left with no time at all to devote any thought to those we have already. Classification, indexing, and maturer consideration about the facts we actually possess are at least as great a need at the present moment as the invention of new facts. "Everyone must realise", says Eugenio Rignano, "how much this theoretical elaboration, performed by means of analyses and comparisons, of generalisations and hypotheses controlled and verified by the correspondence of facts with the results of the reasoning, is useful and necessary if one wishes to reach a progressive systematisation and an ever more synthetic vision of the confused mass of facts which experimentalists pour daily in a continuous stream into the scientific market."


My predecessors in this work have been few in number. The volumes of Haller's, Buffon's, and Milne-Edwards' great treatises, in which they deal with the phenomena of generation, contain as much information as was available up to 1863, but this is purely of historical interest to us. In 1885, W. Preyer, Professor of Physiology at Jena, published his Spezielle Physiologie des Embryo, which still remains a most valuable review, and indeed, even to-day, is the only existing book specially devoted to embryonic physiology. The present century has produced only three books which even touch upon my subject, namely, T. B. Robertson's Chemical Basis of Growth and Senescence, F. H. A. Marshall's Physiology of Reproduction and E. Faure-Fremiet's La Cinetique du Developpement. The first of these was admittedly written to support a particular theory, and in any case says comparatively little about physico-chemical embryology. The second and the third deal with it only as a constituent part of a much wider field. In Marshall's case, the whole array of facts relating to oestrus and breeding, fertilisation and fertility, lactation and sex determination, have to be dealt with, and only three chapters out of sixteen are devoted to the subject of this book. The first of these is contributed by W. Cramer, and covers the biochemistry of the sexual organs, including the unfertiUsed egg ; the second, which deals with foetal nutrition and the placenta, is by J. Lochhead ; and the third, by these two investigators together, is concerned with changes in the maternal organism during pregnancy. Admirable as these chapters are, they are now rather out of date. Moreover, though one or two corners of the field I have before me were covered in Marshall's book, it was from a quite different standpoint.

Faure-Fremiet's work is exactly analogous; it deals with physicochemical embryology only, as it were, in passing. The relevant discussion takes up only two chapters out of seven ; the rest are occupied with tissue culture, growth of protozoal populations, and general cytology. His book covers, it might be said, the third and fourth corners : all the main expanse of the field remains.

Thus neither of these books deals with physico-chemical embryology in an exhaustive and comprehensive fashion, treating it as, in my view, it ought to be treated, with the thoroughness which is deserved by a new branch of natural knowledge. Inseparable, however, from thoroughness of treatment is the submergence of the parts of more general interest under a mass of detail, and it may be well.


therefore, to mention now what sections of the book could be said to be most valuable to any student of general biology. Part i comes in this class, and of Part iii, the middle portion of Section i, all of Sections 2, 3, and 5, thelatter half of Section 7, Sections 8, 9 (especially the end), 11, possibly 18, and finally the Epilegomena.

For my models in the preparation of this book, if it is permissible to name them, I have taken, Growth and Form by d'Arcy Thompson, surely the most scholarly work produced by a biologist in our time, and The Physiology of Reproduction by F. H. A. Marshall, already mentioned, which showed to all successors, in my opinion, how a colossal array of facts can be welded together into an absorbing and readable book, I am conscious that I shall not attain the level of these classics of modern biology, but then

.... Pauci, quos aequus amavit Jupiter, aut ardens evexit ad aethera virtus.

The progress of any branch of natural knowledge can be best described as a continual pilgrimage towards the quantitative. QuaUties can never be altogether left out of account and this is what makes it impossible for science to achieve its end with absolute finality. Yet an association with the probably unattainable is common to all the great types of man's activity. But "Fuyez toujours les a peu pres", as O. W. Holmes used to put it, is a proper maxim for the scientific mind, and whatever this book can do towards making embryology an exact science will be its final justification.


. . . .to measure all things that can be measured, and to make measurable what cannot yet be measured.


Philosophy, Embryology, and Chemistry

The penetration of physico-chemical concepts into embryology has not been entirely peaceful. "In experimental embryology", it has been said, "concepts borrowed from the physical sciences do not admit of calculations being made, and until they do they are not really playing the same role as they do in the sciences from which they have been borrowed and for which they were devised." "Nothing is more clear", says another writer, "in chemistry and physics than that identical results follow upon identical causes. Introduce a disturbing element, even a small one, into your experiment, and the experiment will fail. Such is not the case with the developing egg." W. McDougall, too, endows the egg with good intentions. "The embryo", he says, "seems to be resolved to acquire a certain form and structure, and to be capable of overcoming very great obstacles placed in its path. The development of the forms of organisms seems to be utterly refractory to explanation by mechanical or physicochemical principles." Finally, J. A. Thomson goes farther than them all, and does not hesitate to say, "It is a mere impious opinion that development will one day be described in terms of mechanics". Chapter iv of his Gifford Lectures illustrates the antagonistic attitude to physico-chemical embryology in its most acute form.

It can hardly be a coincidence that so many among the great embryologists of the past were men of strongly philosophic minds. It would be absurd to support this opinion by citing Aristotle, but it holds less obviously true of William Harvey, whose book on generation is full of thoughts about causation, and in the cases of Ernst von Baer, Ernst Haeckel, Wilhelm Roux, Hans Driesch, dArcy Thompson and J. W. Jenkinson, there is no doubt about it. It is not really surprising, for of all the strange things in biology surely the most striking of all is the transmutation inside the developing egg, when in three weeks the white and the yolk give place to che animal with its tissues and organs, its batteries of enzymes and its dehcately regulated endocrine system. This coming-to-be can hardly have failed


to lead, in the minds of those most intimately acquainted with it, to thoughts of a metaphysical character. Nor, it seemed, did those who worked on it do much to diminish its wonder. "Neither the schools of physicians", as Harvey said, "nor Aristotle's discerning brain, have disclosed the manner how the Cock and its seed, doth mint and coine, the chicken out of the Ggg,'^ Or, in the words of Erycius Puteanus, "I will neglect gold, and will praise what is more precious than any metal, I will despise feasts, and will set forth praises of something better than any food or drink. If you would know of what it is that I intend to speak, it is the egg; men marvel at the sun, at meteors flung from heaven, at stars swimming therein, but this is the greatest of all wonders". Here, however, there is one significant thing. It is that the very chapter of Harvey's book in which the preceding remark is found has as its heading "The Efficient Cause of the Chicken, is hard to be found out". It certainly was, but the right clue was in the heading to that exercitation.

This close association of embryology with philosophy, then, made it necessary to discuss at the outset of this book certain points in the more theoretical regions of biology, and, as it were, to defend from a theoretical angle the extension of the domain of physics and chemistry over embryology. I might have entitled this part of the book "The philosophy of embryology", but, in deference to those metaphysicians who rightly insist that the word philosophy should only be used of a definite system of experience which looks at the universe as a corporate whole, I adopted the present heading. Under it I propose to discuss the exact status of the chemical aspect of embryology. For many biologists, having perhaps insufficiently considered the nature of the scientific method, think it likely that the discoveries of modern times may allow of some other basis for biology than mathematical physics and that the scientific niethod may rightly be different in biology from what it is in chemistry. It is this factor in our present intellectual climate which makes it necessary to preface by a philosophical discussion a book in which the concepts of physics and chemistry are extended to a field of biology where they have never before received more than a conventional and formal reverence.

The aim of all studies in physico-chemical embryology must be that expressed by T. H. Huxley when he said, " Zoological Physiology is the doctrine of the functions or actions of animals. It regards


animal bodies as machines impelled by certain forces and performing an amount of work which can be measured and expressed in terms of the ordinary forces of nature. The final object of physiology is to deduce the facts of morphology on the one hand and those of oecology on the other hand from the laws of the molecular forces of matter". It may be regarded as very noteworthy that Huxley here puts morphology as secondary to physiology and as it were derivable from it; he does not place morphology and physiology on two high places, "neither afore or after other", as has so often been done, but he plainly states his view that the anatomical aspect of animals, their external and internal forms, could be deduced from the interplay of physico-chemical forces within them, if we only knew enough about those forces. This is the idea of the primacy of function. It seems always to have two meanings, firstly, the Epicurean-Lucretian one which Huxley adopts here and Roux so brilliantly developed, in which shape is regarded as the outward and visible sign of the properties of matter itself, and, secondly, the Aristotelian one emphasised by J. B. de Lamarck's writings in the eighteenth century, and in our time by E. S. Russell's great work Form and Function, in which psychical factors are introduced as the essential elements in the ultimate analysis of shape. In both these interpretations, function has the priority over form, but the meaning of function is the point of difference. Some biologists, however, seem to think that physiology and morphology are categorical, and the latter is emphatically not reducible to or derivable from the former. The two spheres of study represent, for them, correlative and immiscible disciplines, morphology aiming ultimately at solid geometry, physiology at causation, and "rerum cognoscere causas" is not the basic desire of the scientific mind. They object to the view which regards "the ovum as a kind of chemical device wound up and ready to go off on receipt of a stimulus, the task of the causal morphologist being to disentangle the complex of events which constitute the unwinding process" (Woodger), complaining that in this view no account is taken of the past history of the race, which is left to genetics, again a causal discipHne. To some extent these opinions spring from a conviction that the analytical method is inapplicable to a living being because it is an organism, and of that there is more to be said. But they also arise from a profound unwillingness to subsume biology under physics and a desire to uphold


"the autonomy of biology". This precludes the promise of an everincreasing homogeneity in the structure of science, and hence an ever-increasing simplicity.

The Historical Perspective

That the older embryologists awaited the extension of physicochemical conceptions to embryology is no mere matter of conjecture. Until the mechanical theory of the universe had been consolidated by the " corpuscularian philosophy" of the seventeenth century it would be useless to look for illustration of this, but by 1674 John Mayow was tracing the part played by the " nitro-aerial particles" in the development of the embryo, and in 1732 Hermann Boerhaave was discussing chemical problems with explicit reference to embryonic development. Many other examples of this point of view in the eighteenth century will be given later. Then, when the second decade of the nineteenth century had nearly gone, von Baer, perhaps the greatest of all embryologists, was careful to preface his Entwicklungsgeschichte by a careful account of all that was known about the chemical constitution of the Qgg, and that, although his philosophical inclinations were deeply vitalistic, and even his practical interests morphological. In Roux, of course, this future reference came out explicitly, and the extension of biochemistry into embryology was allowed for and foreseen. An early instance was the association between Wilhelm His and Hans Miescher. Miescher, writing to HoppeSeyler in 1872 said, "I am now collecting material from fishes, birds, and amphibia to lead to a chemical statics of development. With this end in view I shall do analyses of ash, nuclein, and lecithin".

Embryology before Harvey, however, was rigidly Aristotelian, a statement the meaning of which George Santayana has lucidly explained. "Aristotle", said he, "distinguished four principles in the understanding of Nature. The ignorant think that these are all, equally, forces producing change, and the cooperative sources of all natural things. Thus, if a chicken is hatched, they say that the Efficient Cause is the warmth of the brooding hen, yet this heat would not have hatched a chicken out of a stone, so that a second condition, which they call the Material Cause, must be invoked as well, namely, the nature of an egg; the essence of eggness being precisely a capacity to be hatched when warmed gently — because, as they wisely observe, boiling would drive away all potentiality of hatching. Yet, as they


further remark, gentle heat-in-general joined with the essence-ofeggness would produce only hatching-as-such and not the hatching of a chicken, so that a third influence, which they call the Final Cause, or the End-in-view, must operate as well, and this guiding influence is the divine idea of a perfect cock or a perfect hen presiding over the incubation and causing the mere eggness in the egg to assume the likeness of the animals from which it came. Nor, finally, do they find that these three influences are sufficient to produce here and now this particular chicken, but are compelled to add a fourth, a Formal Cause, namely, a particular yolk, a particular shell, and a particular farmyard, on which and in which the other three causes may work, and laboriously hatch an individual chicken, probably lame and ridiculous despite so many sponsors." The Aristotelian account of causation could not be better expressed. Santayana puts this description of it into the mouth of Avicenna in his imaginary dialogue, and makes him go on to say, "Thus these learned babblers would put nature together out of words, and would regard the four principles of interpretation as forces mutually supplementary combining to produce material things ; as if perfection could be one of the sources of imperfection or as if the form which things happen to have could be one of the causes of their having it. Far differently do these four principles clarify the world when discretion conceives them as four rays shed by the light of an observing spirit". In this last observation we may perhaps trace the germ of the Copernican revolution in philosophy effected by Kant, if we may take it to enclose the idea of the activity of the experient subject in all perception.

In science generally, however, the x\ristotelian conceptions went without serious contradiction, and thus formed the framework for all the embryological work that was done, as, for instance, by Albertus Magnus. Owing to its association with the idea of the plan of a divine being, the final cause tended in the Middle Ages to eclipse the others. In the seventeenth century this feeling is well shown in a remarkable passage, which occurs in the Religio Medici of Sir Thomas Browne: "There is but one first cause, and four second causes of all things; some are without Efficient, as God; others without Matter, as Angels; some without Form, as the first matter; but every Essence created or uncreated, hath its Final cause, and some positive End both of its Essence and Operation ; this is the cause I grope after in the works of Nature ; on this hangs the providence of God ; to raise so


beauteous a structure as the World and the Creatures thereof, was but his Art; but their sundry and divided operations, with their predestinated ends, are from the Treasure of his Wisdom. In the causes, nature, and affections of the EcHpses of the Sun and Moon there is most excellent speculation, but to profound farther, and to contemplate a reason why his providence hath so disposed and ordered their motions in that vast circle as to conjoyn and obscure each other, is a sweeter piece of Reason and a diviner point of Philosophy; therefore sometimes, and in some things, there appears to me as much Divinity in Galen his books De Usu Partium, as in Suarez' Metaphysicks: Had Aristotle been as curious in the enquiry of this cause as he was of the other, he had not left behind him an imperfect piece of Philosophy but an absolute tract of Divinity". This was written in Harvey's time, and in Harvey's thought the four causes were still supreme ; his De Generatione Animalium is deeply concerned with the unravelling of the causes which must collaborate in producing the finished embryo. But the end of their domination was at hand, and the exsuccous Lord Chancellor, whose writings Harvey thought so little of, was making an attack on one of Aristotle's causes which was destined to be peculiarly successful. There is no need to quote his immortal passages about the "impertinence", or irrelevance, of final causes in science, for they cannot but be familiar to all scientific men. Bacon demonstrated that from a scientific point of view the final cause was a useless conception; recourse to it as an explanation of any phenomenon might be of value in metaphysics, but was pernicious in science, since it closed the way at once for further experiments. To say that embryonic development took the course it did because the process was drawn on by a pulling force, by the idea of the perfect adult animal, might be an explanation of interest to the metaphysician, but as it could lead to no fresh experiments, it was nothing but a nuisance to the man of science. Later on, it became clear also that the final cause was irrelevant in science owing to its inexpressibility in terms of measurable entities. From these blows the final cause never recovered. In England the seventeenth century was the time of transition in these aflfairs, and in such books as Josfeph Glanville's Plus Ultra and Scepsis Scientifica, for instance, and Thomas Sprat's Defence of the Royal Society, the stormy conflict between the "new or experimental philosophy" and the Aristotelian "school-philosophy" can be easily followed. Francis


Gotch has given a delightful account of the evening of AristoteUanism, but it involved a stormy sunset, and the older ideas did not give way without a struggle. Harvey's work is perfectly representative of the period of transition, for, in his preface under the heading "Of the Method to be observed in the knowledge of Generation", he says, "Every inquisition is to be derived from its Causes, and chiefly from the Material and Efficient". As for the formal cause. Bacon expressly excluded it from Physic, and it quietly disappeared as men saw that scientific laws depended on the repeatableness of phenomena, and that anything unique or individual stood outside the scope of science. Thus in the case of the developing egg, the formal (the particular farmyard, etc.) and the final causes are scientifically meaningless, and if it were desired to express modern scientific explanation in Aristotelian terminology, the material and efficient causes would alone be spoken of, essence-of-eggness being a "chymical matter" as well as the heat of the brooding hen.

Obstacles to Chemical Embryology

The complexity of living systems, however, is such that many minds find it difficult to accept this physico-chemical account as the most truly scientific way of looking at it. This is doubtless due in part to an erroneous notion, which is yet very tenacious of existence, that the mechanical theory of the universe must, if accepted at all, be accepted as an ultimate ontological doctrine, and so involve its supporter in one of the classical varieties of metaphysical materialism. It cannot be too strongly asserted that this is not the case. To imagine that it is, is to take no account of the great space that separates us from the last century. "When the first mathematical, logical, and natural uniformities", said WilHam James, "the first Laws, were discovered, men were so carried away by the clearness, beauty, and simplification that resulted that they believed themselves to have deciphered authentically the eternal thoughts of the Almighty. His mind also thundered and reverberated in syllogisms. He also thought in conic sections, squares, and roots and ratios, and geometrised like Euclid. He made Kepler's laws for the planets to follow, he made velocity increase proportionately to the time in falhng bodies; he made the laws of the sines for light to obey when refracted; he established the classes, orders, families, and genera of plants and animals, and fixed the distances between them."


Far different is the account of itself which science has since learned to give. But this change of attitude is not a revolt against thought as such, or against reason as such ; it is only a loss of belief in the literal inspiration of the formulae proper to science. It would be just as extravagant to claim that the scientific investigator of the twentieth century sets down absolute truths in his laboratory notebook, and, armed with an infallible method, explores the real structure of an objective world, as it would be fantastic to claim that Jehovah dictated an absolute code of the good to Moses on Mount Sinai. To say that the development of a living being can best be described in a metrical or mechanical way is not to say that it is metrical or mechanical and nothing else. The physico-chemical embryologist is not committed to any opinion on what his material really is, but he is committed to the opinion that the scientific method is one way of describing it, and that it is best to apply that method in its full rigour if it is to be applied at all. In other words, following the train of thought of William James, he does not assert that the courts of Heaven as well as those of our laboratories resound with expressions such as "organisers of the second grade," and "so many milHgrams per cent." The mechanical theory of the world, which is, as many beHeve, bound up indissolubly with one of the ultimate types of human experience, can no longer be considered as necessarily involving the exclusion of other theories of the world. Or, put in another way, it is a theory of the world, and not a pocket edition of the world itself

But before bringing forward any arguments in support of this attitude and in defence of physico-chemical embryology, it will be well to consider briefly those theoretical tendencies in modern biology which go together under the inexact adjective "neo-vitalistic", for their influence in scientific thought has been far-reaching. To deal critically with them is not a waste of time, for, were we to adopt any one of them, we should find that the notion of embryology as complicated biophysics and biochemistry would have to be abandoned, and quite other means of approach (never, indeed, very well defined) would have to be used.

The Stumbling-block of Hormism

Hormism, or "Psychobiology," may be dealt with in a few words. Chiefly supported by A. Wagner in Germany, and by E. S. Russell and L. T. Hobhouse in this country, it holds that — to


use Lloyd Morgan's terminology — a physiological tale cannot be told separately from a psychological tale. Instead of expressing living processes in terms of physical causes and effects, the hormists wish to regard unconscious striving as the essential urge in life, and such conceptions as food, rest, fatigue, etc., as irreducible biological categories. These thinkers do not often acknowledge their debt to Galen of Pergamos, who put forward, as early as a.d. 170, an essentially similar conception as the basis of his biology. In the treatise On the Natural Faculties he says, "The cause of an activity I term a faculty.... Thus we say that there exists in the veins a blood-making faculty, as also a digestive faculty in the stomach, a pulsatile faculty in the heart, and in each of the other parts a special faculty corresponding to the function or activity of that part". He also said, "We call it a faculty so long as we are ignorant of the cause which is operating", but he never actually suggested any such underlying cause, and seems to have thought it impossible to ascertain. So do the hormists. According to them the actions of protozoa are to be described in terms of avoiding responses, seeking responses and the like, language which, as they claim, is much simpler than the complex terminology of surface tension and molecular orientation. Everything, of course, depends on what is meant by simple. To say that a protozoon seeks the light is evidently more naive than to say that a dimolecular photochemical reaction takes place in its protoplasm leading to an increase of lactic acid or what not on the stimulated side, but since the latter explanation fits into the body of scientific fact known already it is open to the biochemist to say that, for his part, he. considers the latter explanation the simpler. It is, in fact, simpler in the long run. Psychobiology or hormism differs from the other forms of neo-vitalism because it insists on retaining " commonsense " explanations in biology as categories of biological thought beneath which it is impossible to go. It dismisses the entelechy of dynamic Teleology, on the ground that it acts, as it were, in addition to the mechanistic schema, accepting the latter fully but interfering in it. It resembles much more finaUsm and organicism, but lays stress rather on the unconscious striving force which seems to animate colloidal solutions of carbohydrates, fats, and proteins. It resembles the Behaviourism of J. B. Watson superficially by emphasising animal behaviour, but it fundamentally differs, for it asks the question — Does an animal see the green light and the red light in this experiment


as we do, or does it see them as two shades of grey as colour-blind people do? while the behaviourist asks — Does it respond according to difference of light-intensity or difference of wave-lengths ? Hormism, in fact, recurs continually to psychical factors. Samuel Butler, for instance, one of its principal exponents, wrote, "I want to connect the actual manufacture of the things a chicken makes inside an egg with the desire and memory of the chicken so as to show that one and the same set of vibrations at once change the universal substratum into the particular phase of it required" (cf. ^ rov hwdixei, 6vTo^r 0^. "Democritus and Epicurus hold", says Plutarch, "that this unperfect fruit of the wombe receiveth nourishment at the mouth; and thereupon it commeth that so soon as ever it is borne it seeketh and nuzzeleth with the mouth for the brest head or nipple of the pappe : for that within the matrice there be certain teats; yea, and mouths too, whereby they may be nourished. But Alcmaeon affirmeth that the infant within the mother's wombe, feedeth by the whole body throughout for that it sucketh to it and draweth in maner of a spunge, of all the food, that which is good for nourishment." It would appear also that Democritus believed the external form of the embryo to be developed before the internal organs were formed.

1-3. Hippocrates: the Beginning of Observation

But the foregoing fragments of speculation do not really amount to much. The first detailed and clear-cut body of embryological knowledge is associated with the name of Hippocrates, of whom nothing certain is known save that he was born probably in the forty-fifth Olympiad, about 460 b.c, that he lived on the island of Cos in the Aegean Sea, and that he acquired greater fame as a


physician than any of his predecessors, if we may except the legendary names of Aesculapius, Machaon and Podalirius. It has not been believed for many centuries past that all the writings in the collection of Hippocratic books were actually set down by him, and much discussion has taken place about the authenticity of individual documents.

Most of the embryological information is contained in a section which in other respects (style, etc.) shows homogeneity. We are therefore rather interested in that unknown biological thinker who wrote the books in this class, for he could with considerable justice be referred to as the first embryologist. Littre discusses his identity, but there is no good evidence for any of the theories about it, though perhaps the most likely one is that he was Polybus, the son-in-law of Hippocrates. That the writings on generation are only slightly later than the time of Hippocrates is more or less clear from the fact that Bacchius knew of them, and actually mentions them.

For the most part the embryological knowledge of Hippocrates is concerned with obstetrical and gynaecological problems. Thus in the Aphorisms, d(f)opicr/iioi, the books on epidemics, eirchrifxiai, the treatise on the nature of women, irepl rywaLKelr)'? (f)V(rio'?, the discussions of premature birth, Trepl eirraixrjvov, the books on the diseases of women, irepl 'yvvaixeiaiv, and the pamphlet on superfoetation, there are many facts recorded about the embryo, but all with obstetrical reference. There are some curious notions to be found there, such as the association of right and left breasts with twin embryos and a prognostic dependent on this.

But the three books which are most important in the history of embryology are the treatise on Regimen, irepl StaLTr}<;, the work on generation, irepl jovr]<;, and the book about the nature of the infant, Trepl ^vaio<; TraiZiov. The two latter really form one continuous discussion, and it is not at all clear how they came to be split up into separate books. In the Regimen the writer expounds his fundamental physiological ideas, involving the two main constituents of all natural bodies, fire and water. Each of these is made up of three primary natures, only separable in thought and never found isolated, heat, dryness and moisture, and each of them has the power of attracting, eXKeiv, their like, an important feature of the system. Life consists in moisture being dried up by fire and fire being wetted by moisture alternately, rpo^-i'i, the nourishment (moisture) coming into


the body, is consumed by the fire so that fresh rpocfir] is in its turn required.

It is important to note that the Hippocratic school was far more akin in its general attitude to living things to modern physiology than the Aristotelian and Galenic physiology. For no considerations of final causes complicate the causal explanations of the Hippocratic school, and the author of the irepl SmtV?/? indeed devotes seven chapters to a detailed comparison of the processes of the body {a) with the processes of the inorganic world both celestial and terrestrial, and (b) with the processes used by men in the arts and crafts, such as iron-workers, cobblers, carpenters and confectioners. These discussions present distinct mechanistic features.

He then in Section 9 sets forth his theory of the formation of the embryo. "Whatever may be the sex", he says, "which chance gives to the embryo, it is set in motion, being humid, by fire, and thus it extracts its nourishment from the food and breath introduced into the mother. First of all this attraction is the same throughout because the body is porous but by the motion and the fire it dries up and solidifies — vtto Be r?)? Kivijcno^; Koi tov irvpoii ^rfpaiveTai koI arepeovraL — as it solidifies, a dense outer crust is formed, and then the fire inside cannot any more draw in sufficient nourishment and does not expel the air because of the density of the surrounding surface. It therefore consumes the interior humidity. In this way parts naturally solid being up to a point hard and dry are not consumed to feed the fire but fortify and condense themselves the more the humidity disappears — these are called bones and nerves. The fire burns up the mixed humidity and forwards development towards the natural disposition of the body in this manner ; through the solid and dry parts it cannot make permanent channels but it can do so through the soft wet parts, for these are all nourishment to it. There is also in these parts a certain dryness which the fire does not consume, and they become compacted one to another. Therefore the most interior fire, being closed round on all sides, becomes the most abundant and makes the most canals for itself (for that was the wettest part) and this is called the belly. Issuing out from thence, and finding no nourishment outside, it makes the air pipes and those for conducting and distributing food. As for the enclosed fire, it makes three circulations in the body and what were the most humid parts become the venae cavae. In the intermediate part the remainder of the water contracts and hardens


forming the flesh." In this account of the formation of the embryo, which seems at first sight a Httle fantastic, there are several interesting things to be remarked. Firstly, there is to be noted throughout it a remarkable attempt at causal explanations and not simply morphological description. The Hippocratic writer is out to explain the development of the embryo from the very beginning on machine-like principles, no doubt unduly simplified, but related directly to the observed properties of fire and water. In this way he is the spiritual ancestor of Gassendi and Descartes. The second point of interest is that he speaks of the embryo drying up during its development, a piece of observation which anyone could make by comparing a fourth-day chick with a fourteenth-day one, and which we express to-day in graphical form (see Fig. 220). Thirdly, the ascription of the main driving force in development to fire has doubtless no direct relation to John Mayow's discovery, two thousand years later, that there is a similarity between a burning candle and a living mouse each in its bell-jar, and may mean as much or as little as Sir Thomas Browne's remark, "Life is a pure flame, and we live by an invisible sun within us". Yet the essential chemical aspect of living matter is oxidation, and the development of the embryo no less than the life of the adult is subject to this rule, so that what may have been a mere guess on the part of the Hippocratic writer, may also have been a flash of insight due to the simple observation which, after all, it was always possible to make, namely, that both fires and li\dng things could be easily stifled.

Preformationism is perhaps foreshadowed in Section 26 of the same treatise. "Everything in the embryo is formed simultaneously. All the limbs separate themselves at the same time and so grow, none comes before or after other, but those which are naturally bigger appear before the smaller, without being formed earlier. Not all embryos form themselves in an equal time but some earlier and some later according to whether they meet with fire and food, some have everything visible in 40 days, others in 2 months, 3, or 4. They also become visible at variable times and show themselves to the light having the blend (of fire and water) which they always will have."

The work on Generation is equally interesting. The earlier sections deal with the differences between the male and the female seed, and the latter is identified with the vaginal secretion. Purely embryological


discussion begins at Section 14, where it is stated that the embryo is nourished by maternal blood, which flows to the foetus and there coagulates, forming the embryonic flesh. The proof alleged for this is that during pregnancy the flow of menstrual blood ceases; therefore it must be used up on the way out. In Section 15 the umbiHcal cord is recognised as the means by which foetal respiration is carried on. Section 1 7 contains a fine description of development with a very interesting analogy. "The flesh", it is said, "brought together by the spirit, TO TTvevfia, grows and divides itself into members, hke going to like, dense to dense, flabby to flabby, humid to humid. The bones harden, coagulated by the heat." Then a demonstration experiment follows : "Attach a tube to an earthen vessel, introduce through it some earth, sand, and lead chips, then pour in some water and blow through the tube. First of all, everything will be mixed up, but after a certain time the lead will go to the lead, the sand to the sand, and the earth to the earth, and if the water be allowed to dry up and the vessel be broken, it will be seen that this is so. In the same way seed and flesh articulate themselves. I shall say no more on this point". Here again was an attempt at causal explanation, rather than morphological description, in complete contrast to the later work of


Section 22 contains a suggestive comparison between seeds of plants and embryos of animals, but the identification of stalk with umbihcal cord leads to a certain confusion. Perhaps the most interesting passage of aU is to be found in Section 29. "Now I shall speak", says the unknown Hippocratic embryologist, "of the characters which I promised above to discuss and which show as clearly as human intelligence can to anyone who will examine these things that the seed is in a membrane, that the umbilicus occupies the middle of it, that it alternately draws the air through itself and then expels it, and that the members are attached to the umbilicus. In a word, all the constitution of the foetus as I have described it to you, you will find from one end to the other if you wiU use the following proof Take 20 eggs or more and give them to 2 or 3 hens to incubate, then each day from the second onwards tiU the time of hatching, take out an egg, break it, and examine it. You will find everything as I say in so far as a bird can resemble a man. He who has not made these observations before will be amazed to find an umbihcus in a bird's egg. But these things are so, and this is what I intended


to say about them." We see here as clearly as possible the beginnings of systematic embryological knowledge, and from this point onwards, through Aristotle, Leonardo, Harvey and von Baer, to the current number of the Archivf. Entwicklungsmechanik, the line runs as straight as Watling Street.

In Section 30 there is an important passage in which the author discusses the phenomena of birth. "I say", he says, "that it is the lack of food which leads to birth, unless any violence has been done; the proof of which is this ; — the bird is formed thus from the yolk of the egg, the egg gets hot under the sitting hen and that which is inside is put into movement. Heated, that which is inside begins to have breath and draws by counter-attraction another cold breath coming from the outside air and traversing the egg, for the egg is soft enough to allow a sufficient quantity of respiration to penetrate to the contents. The bird grows inside the egg and articulates itself exactly like the child, as I have previously described. It comes from the yolk but it has its food from, and its growth in, the white. To convince oneself of this it is only necessary to observe it attentively. When there is no more food for the young one in the egg and it has nothing on which to live, it makes violent movements, searches for food, and breaks the membranes. The mother, perceiving that the embryo is vigorously moving, smashes the shell. This occurs after 20 days. It is evident that this is how things happen, for when the mother breaks the shell there is only an insignificant quantity of liquid in it. All has been consumed by the foetus. In just the same way, when the child has grown big and the mother cannot continue to provide him with enough nourishment, he becomes agitated, breaks through the membranes and incontinently passes out into the external world free from any bonds. In the same way among beasts and savage animals birth occurs at a time fixed for each species without overshooting it, for necessarily in each case there must be a point at which intra-uterine nourishment will become inadequate. Those which have least food for the foetus come quickest to birth and vice versa. That is all that I had to say upon this subject."

The theory underlying this passage evidently is that the main food of the fowl embryo is the white and that the yolk is there purely for constructional purposes. Had the author not been strongly attached to this erroneous view he could not have failed to notice the unabsorbed yolk-sac which still protrudes from the abdomen of the


hatching chick, and if he had given this fact a little more prominence he could hardly have come to enunciate the general theory of birth which appears in the above passage. Moreover, had he been acquainted with the circulation of the maternal and foetal blood in viviparous animals, he could hardly have held that there was less food in a given amount of maternal blood at the end of development than at the beginning. At any rate, his attempted theory of birth was a worthy piece of scientific effort, and we cannot at the present moment be said to understand fully the principles governing incubation time (see p. 470).

The treatises on food and on flesh, trepl Tpo(f>rj<i and irepl aapKcovy are both late additions to the Hippocratic corpus, but contain points of embryological interest. Section 30 of the former contains some remarks on embryonic respiration, and Section 3 of the latter has a theory of formation of nerves, bones, etc. by difference of composition of glutinous substances, fats, water, etc. Section 6 supports the view that the embryo is nourished in utero by sucking blood from the placenta, and the proof given is that its intestine contains the meconium at birth. Moreover, it is argued, if this were not so, how could the embryo know how to suck after it is born?

1-4. Aristotle

After the Hippocratic writings nothing is of importance for our subject till Aristotle. It is true that in the Timaeus Plato deals with natural phenomena, eclectically adopting opinions from many previous writers and welding them into a not very harmonious or logical whole. But he has hardly any observations about the development of the embryo. The four elements, earth, fire, air, and water, are, according to him, all bodies and therefore have plane surfaces which are composed of triangles. Applying this semi-atomistic hypothesis to the growth of the young animal, he says, "The frame of the entire creature when young has the triangles of each kind new and may be compared to the keel of a vessel that is just ofT the stocks ;^ they are locked firmly together and yet the whole mass is soft and delicate, being freshly formed of marrow and nurtured on milk. Now when the triangles out of which meats and drinks are composed come in from without, and are comprehended in the body, being older and weaker than the triangles already there, the frame of the body gets the better of them and its newer triangles cut them up and so the


animal grows great, being nourished by a multitude of similar particles." This is as near as Plato gets to embryological speculation. His description has a causal ring about it, which is in some contrast with the predominantly teleological tone of the rest of his writings ; for instance, only a few pages earlier he has been speaking of the hair as having been arranged by God as "a shade in summer and a shelter in winter". It is also true that Plato may have said more about the embryo than appears in the dialogues. Plutarch mentions various speculations about sterility, and adds, "Plato directly pronounceth that the foetus is a living creature, for that it moveth and is fed within the bellie of the mother".

But all this was only the slightest prelude to the work of Plato's pupil, Aristotle. Aristotle's main embryological book was that entitled Trepl ^mcov yeveaeco'i, On the Generation of Animals, but embryological data appear in irepl ^axop, The History of Animals, irepl ^wmv fjLopicov, On the Parts of Animals, Trepl dva7rvofj<i, On Respiration, and Trepl ^Mcov Ktvrjcr€(o<i, On the Motion of Animals. All these were written in the last three-quarters of the fourth century B.C.

With Aristotle, general or comparative biology came into its own. That almost inexhaustible profusion of living shapes which had not attracted the attention of the earlier Ionian and Italo-Sicilian philosophers, which had been passed over silently by Socrates and Plato, intent as ever upon ethical problems, but which had been for centuries the inspiration of the vase-painters and other craftsmen {(^coypdcjioL), was now for the first time exhaustively studied and reduced to some sort of order. The Hippocratic school with their "Coan classification of animals", which Burckhardt has discussed, had indeed made a beginning, but no more. It was Aristotle who was the first curator of the animal world, and this comparative outlook colours his embryology, giving it, on the whole, a morphological rather than a physiological character.

The question of Aristotle's practical achievements in embryology is interesting, and has been discussed by Ogle. There is no doubt that he diligently followed the advice of the author of the Hippocratic treatise on generation and opened fowl's eggs at different stages during their development, but he learnt much more than the unknown Hippocratic embryologist did from them. It is also clear that he dissected and examined all kinds of animal embryos, mammalian and cold-blooded. The uncertain point is whether he also


dissected the human embryo. He refers in one place to an "aborted embryo", and as he was able to obtain easily all kinds of animal embryos without waiting for a case of abortion, it is likely that this was a human embryo. Ogle brings forward six or seven passages which all contain statements about human anatomy and physiology only to be explained on the assumption that he got his information from the foetus. So it is probable that his knowledge of biology was extended to man in this way, as would hardly have been the case if he had lived in later times, when the theologians of the Christian Church had come to very definite conclusions about the sanctity of foetal as well as adult life.

The Trept i^wcov ^eveaeoo<;^ the first great compendium of embryology ever written, is not a very well-arranged work. There are a multitude of repetitions, and the order is haphazard, so that long digressions from the main argument are common. The work is divided into five books, of which the second is much the most important in the history of embryology, though the first has also great interest, and the third, fourth, and fifth contain much embryological matter mixed up among points of generation and sexual physiology.

Book I begins with an introduction in which the relative significance of efficient and final causes is considered, and chapters i to 7 deal with the nature of maleness and femaleness, the nature and origin of semen, the manner of copulation in different animals and the forms of penis and testes found in them. Chapter 8 continues this, and describes the different forms of uterus in different animals, speaks of viviparity and oviparity, mentions the viviparous fishes (the selachians) and draws a distinction between perfect and imperfect eggs. Chapter 9 discusses the cetacea; 10, eggs in general; and 11 returns to the differences between uteri. In chapter 12 the question is raised why all uteri are internal, and why all testes are not, and in chapter 13 the relations between the urinary and the genital systems are discussed. Copulation now receives attention again, in 14 with regard to Crustacea, in 15 with regard to cephalopoda, and in 16 with regard to insecta. After this point the argument Hfts itself on to a more theoretical plane, and opens the question of pangenesis, into which it enters at length during the course of chapters 17 and 18, refuting eventually the widely-held view that the semen takes its origin from all the parts of the body so as to be able to reproduce in the offspring the characteristics of the parent.


The nature of semen receives a long discussion; it is decided at last that it is a true secretion, and not a homogeneous natural part (a tissue) nor a heterogeneous natural part (an organ) nor an unnatural part such as a growth, nor mere nutriment, nor yet a waste product. It is here that the theory is put forward that the semen supplies the "form" to the embryo and whatever the female produces supplies the matter fit for shaping. The obvious question has next to be answered, what is it that the female supplies? Aristotle concludes in chapters 19 and 20 that the female does not produce any semen, as earlier philosophers had held, but that the menstrual blood is the material from which the seminal fluid, in giving to it a form, will cause the complete embryo to be produced. This was not a new idea, but had already been suggested by the author of the Hippocratic ire pi yovi]^. What was quite new here, was the idea that the semen supplied or determined nothing but the form. Chapters 21 and 22 are rather confused ; they contain more arguments against pangenesis, and considerations upon the contrast between the active nature of the male and the passive nature of the female. Chapter 23, which closes the first book, compares animals to divided plants, for plants in Aristotle's view fertilise themselves.

Book II opens with a magnificent chapter on the embryological classification of animals, showing Aristotle, the systematist, at his best — his classification is reproduced in Chart I. But the chapter also includes a brilliant discussion of epigenesis or preformation, fresh development or simple unfolding of pre-existent structures, an antithesis which Aristotle was the first to perceive, and the subsequent history of which is almost synonymous with the history of embryology. The question in its acutest form was not settled until the eighteenth century, but since then it has become clear that there were elements of truth in the opinion which was the less true of the two. Chapter 2 is not so important, though it has some interesting chemical analogies; it compares semen to a foam, and suggests that it was this foam, like that of the sea, which gave birth to the goddess Aphrodite^. But chapter 3 returns to the high level of speculation and thought found in the opening part of the book, for it deals with the degree of aliveness which the embryo has during its passage through its developmental stages. Aristotle

^ To the Greeks all natural foams possessed a generative virtue, and a Zeus Aphrios was worshipped at Pherae in Thrace.



1/3 t— I





< o

o o









H -Z

T3.g u ,« ^ I.2 u:3 c« 

Q Pi O O O







I— I ■








Pi <


o V a -^

C3 C « -O

< " ^ « cr

^•£ u S «5 § o .-^ o

L <«  o


5 a,

in O ^

" " „

- " u "

u o -5 o

— ^^ ^ a

O O iJ «J

k ^ -S -C

3 "^

? z:

=: o

j3 •« he


g bo G

'^ t: ::

c3 c^ Ij

r! 3 ■ — I • —

5 C

bo V

s s

u v "

W> > rt "S bOX! h .S


"08 >


^ 'bO

o .5

u X!

>< s « 

r *j o

0-5 bo


bO O bo >'


J5 =« 


a Oh

._• C ? rt

J2 c .5 -o

-? S m bc

3.^ r>< CO

o il

C C3






bo bo

C bo



External Viviparo



c c

•3 ^.







fa o<


s -. =« o

2 "^ "^ u



S bo

3 2 ^

bci bo


C C —


tn rt

2 S

c O





does not here anticipate the form of the recapitulation theory, but he certainly suggests the essence of it in perfectly clear terms. This chapter has also an interest for the history of theological embryology, for its description of the entry of the various souls into the embryo was afterwards made the basis for the legal rulings concerning abortion. This chapter also discusses embryogeny as a whole, as does the succeeding one. Chapter 5 is a digression into the problem of why fertilisation is necessary by the male, but it has also some curious speculations as to what extent the hen's egg is alive, if it is infertile. The main thread is resumed in chapters 6 and 7, two very fine ones, in which embryogeny and foetal nutrition are thoroughly dealt with, but dropped again in the last section, chapter 8, which is devoted to an explanation of sterility. This ends the second book.

The third book is chiefly concerned with the application of the general embryological principles described in the previous book to the comparative field, and the fourth book contains a collection of minor items which Aristotle has not been able to speak of before.

But if the work as a whole tails off in a rather unsatisfactory manner, its merits are such that this hardly matters. The extraordinary thing is that, building on nothing but the scraps of speculation that had been made by the Ionian philosophers, and the exiguous data of the Hippocratic school, Aristotle should have produced, apparently without effort, a text-book of embryology of essentially the same type as Graham Kerr's or Balfour's. It is even very possible that Aristotle was unacquainted with any of the Coan school, for, though he often mentions Democritus, Anaxagoras, Empedocles and even Polybus, yet he never once quotes Hippocrates, and this is especially odd, for Aristotle is known to have collected a large library. Probably Hippocrates was only known to Aristotle as an eminent medical man; if this is so, Aristotle's achievements are still more wonderful.

The depth of Aristotle's insight into the generation of animals has not been surpassed by any subsequent embryologist, and, considering the width of his other interests, cannot have been equalled. At the same time, his achievements must not be over-estimated. Charles Darwin's praise of him in his letter to Ogle (which is too well known to quote) is not without all reservations true. There is something to be said for Lewes as well as Piatt. Aristotle's conclusions were sometimes not warranted by the facts at his disposal,


and some of his observations were quite incorrect. Moreover, he stood at the very entrance into an entirely unworked field of knowledge ; he had only to examine, as it were, every animal that he could find, and set down the results of his work, for nobody had ever done it before. It was like the great days of nineteenth-century physiology, when, as the saying was, "a chance cut with a scalpel might reveal something of the first importance".

As has already been said, Aristotle regarded the menstrual blood as the material out of which the embryo was made. "That, then, the female does not contribute semen to generation", says Aristotle, "but does contribute something, and that this is the matter of the catamenia, or that which is analogous to it in bloodless animals, is clear from what has been said, and also from a general and abstract survey of the question. For there must needs be that which generates and that from which it generates, even if these be one, still they must be distinct in form and their essence must be different; and in those animals that have these powers separate in two sexes the body and nature of the active and passive sex also differ. If, then, the male stands for the effective and active, and the female, considered as female, for the passive, it follows that what the female would contribute to the semen of the male would not be semen but material for the semen to work upon. This is just what we find to be the case, for the catamenia have in their nature an affinity to the primitive matter." Thus the male dynamic element {t6 appev iroLn^TtKov) gives a shape to the plastic female element {to OrjXv TradrjTiKov). Aristotle was right to the extent that the menstrual flow is associated with ovulation, but as he knew nothing of the mammalian ovum, and indeed, as is shown in his embryological classification, expressly denied that there was such a thing, his main menstruation theory is wrong. Yet it was not an illegitimate deduction from the facts before him.

These views of Aristotle's about the contribution of the female to the embryo are in striking contrast with certain conceptions of a century before which were probably generally held in Greece. There is a most interesting passage relating to them in the Eumenides of Aeschylus, when, during the trial scene, Apollo, defending Orestes from the charge of matricide, brings forward a physiological argument. "The mother of what is called her child", Apollo is made to say, "is no parent of it, but nurse only of the young Hfe that is sown in her (jpo(f)6<; 8e Kv^iajoLATE II

'■ jj.V^ki^-vfc . i ^ imj^oi^^

!^ Lfe<? l^i n 1^ tm

J. J O € ^4

^ t ;:: t o^^fi "




serves for nourishment; whiles the chick is unhatched and within the egge, the head is bigger than all the bodie besides; and the eies that be compact and thrust together be more than the verie head. As the chick within growes bigger, the white turneth into the middest, and is enclosed within the yolke. By the 20 day (if the eggs be stirred) ye shall heare the chick to peepe within the verie shell; from that time forward it beginneth to plume and gather feathers ; and in this manner it lies within the shell, the head resting on the right foot, and the same head under the right wing, and so the yolke by little and little decreaseth and faileth". But the best way to illustrate Pliny's embryology is to copy out some of his index, as follows :

The Table to the first Tome of Plinies Naturall Historie.

Egs diverse in colour 298

Egs of birds of 2 colours within the shell ibid.

Egs of fishes of i colour ibid.

Egs of birds, serpents, and fishes, how they differ ibid.

Egs best for an hen to sit upon 299

Egs hatched without a bird, onely by a kind heat ibid.

Egs how they be marred under an hen ibid,

wind-egs, called Hypenemia 300

how they be engendred 301

wind-egs, Zephyria ibid.

Egs drawne through a ring ibid.

Egs how they be best kept ibid.

The Table to the second Tome of Plinies Naturall Historie,

Egs of hens and their medicinable properties 351

yolke of hens egs, in what cases it is medicinable 352

Egs all yolke, and without white, be called Schista ibid,

skinne of an Hens egge-shell, good in Physicke ibid.

Hens Eggeshell reduced unto ashes, for what it serveth ibid,

the wonderfull nature of Hens Eggeshels ibid.

Hens Egges, all whole as they be, what they are good for 353

the commendations of Hens Egges, as a meat most medicinable ibid. Hens Egge, a proper nourishment for sicke folks, and may go

for meat and drinke both ibid.

Egge-shels, how they may be made tender and pliable ibid,

white of an Egge resisteth fire ibid,

of Geese Egges a discourse 354 the serpents egge, which the Latines call Anguinum, what it

is, and how engendred 355

This last item exhibits Pliny at his worst. It is worth quoting, apart from its intrinsic value, for it shows to what depths embryological knowledge descended within four hundred years after Aristotle collected his specimens on the shores of the lagoon of Pyrrha, and talked with the fishermen of Mitylene. "I will not overpasse one kind of eggs besides, which is in great name and request in France, and whereof the Greeke authors have not written a word ; and this is



the serpents egg, which the Latins call Anguinum. For in Summer time yerely, you shall see an infinit number of snakes gather round together into an heape, entangled and enwrapped one within another so artificially, as I am not able to expresse the manner thereof; by the means therefore, of the froth or salivation which they yeeld from their mouths, and the humour that commeth from their bodies, there is engendred the egg aforesaid. The priests of France, called Druidae^, are of opinion, and so they deliver it, that these serpents when they have thus engendred this egg do cast it up on high into the aire by the force of their hissing, which being observed, there must be one ready to catch and receive it in the fall again (before it touch the ground) within the lappet of a coat of arms or souldiours cassocks. They affirme also that the party who carrieth this egg away, had need to be wel mounted upon a good horse and to ride away upon the spur, for that the foresaid serpents will pursue him still, and never give over until they meet with some great river betweene him and them, that may cut off and intercept their chace. They ad moreover and say that the only marke to know this egg whether it be right or no, is this, that it will swim aloft above the water even against the stream, yea though it were bound and enchased with a plate of gold." But one must not be too severe upon Pliny, for he and his translator, Philemon Holland, provide an entertainment unequalled anywhere else.

To some extent the same applies to Plutarch of Chaeronea, who lived about the same time. Plutarch's writings, inspired as they were throughout by the desire to commend the ancient religion of Greece to a degenerate age, represent no milestone or turning-point in the history of embryology, yet there is a passage in the Symposiaques, or Table-questions which bears upon it. The third question of book 2 is "Whether was before, the hen or egg?" "This long time", says Plutarch, "I absteined from eating egges, by reason of a certaine dream I had, and the companie conceived an opinion or suspition of me that there were entred into my head the fantasies and superstitions of Orpheus or Pythagoras, and that I abhorred to eat an egge for that I believed it to be the principle and fountaine of generation." He then makes the various characters in the dialogue speak to the motion, and one of them, Firmus, ends his speech thus, "And

^ For further information about the serpent's eggs of the Druids, see Kendrick; they were probably fossil echinoderms.


now for that which remaineth (quoth he and therewith he laughed) I will sing unto those that be skilfull and of understanding one holy and sacred sentence taken out of the deepe secrets of Orpheus, which not onely importeth this much, that the cgge was before the henne, but also attributeth and adjudgeth to it the right of eldership and priority of all things in the world, as for the rest, let them remaine unspoken of in silence (as Herodotus saith) for that they bee exceeding divine and mysticall, this onely will I speake by the way; that the world containing as it doth so many sorts and sundry kinds of living creatures, there is not in manner one, I dare well say, exempt from being engendred of an egge, for the egge bringeth forth birdes and foules that fiie, fishes an infinit number that swimme, land creatures, as lizards, such as live both on land and water as crocodiles, those that bee two-footed, as the bird, such as are footlesse, as the serpent, and last of all, those that have many feet, as the unwinged locust. Not without great reason therefore is it consecrated to the sacred ceremonies and mysteries of Bacchus as representing that nature which produceth and comprehendeth in itselfe all things". This emphatic passage looks at first sight as if it was a statement of the Harveian doctrine omne vivum ex ovo. But the fact that no mammals are mentioned makes this improbable. Firmus then sits down and Senecius opposes him with the well-worn argument that the perfect must precede the imperfect, laying stress also on the occurrence of spontaneous, i.e. eggless, generation, and on the fact that men could find no "row" in eels. Three hundred years later, Ambrosius Macrobius handled the question again (see Whittaker), and the progress in embryological knowledge could be strikingly shown by the difference in treatment. It would be an interesting study to make a detailed comparison of them.

1-6. Galen

Another fifty years brings us to Galen of Pergamos, second in greatness among ancient biologists, though in spite of his multitudinous writings he does not quite take this high rank in embiyology. That knowledge of the development of the foetus was at this time specially associated with Peripatetic tradition appears from a remark of Lucian of Samosata, Galen's contemporary. In the satire called, The Auction of the Philosophies, Hermes, the auctioneer, referring to the Peripatetic who is being sold, says, "He will tell you all about the


shaping of the embryo in the womb". But Galen was now to weld together all the biological knowledge of antiquity into his voluminous works, and so transmit it to the Middle Ages.

Most of Galen's writing was done between a.d. 150 and 180. Out of the twenty volumes of Kiihn's edition of 1829, l^^s than one is concerned with embryology, a proportion considerably less than in the case of Aristotle. Galen's embryology is to be found in his Trepl (f)V(nKcov Svvdfiecov, On the Natural Faculties, which contains the theoretical part, and in his On the Formation of the Foetus, which contains the more anatomical part. There is also the probably spurious treatise et ^mov to Kara <yaa-Tp6<i, On the Question of whether the Embryo is an Animal.

It is important to realise at the outset that Galen was a vitalist and a teleologist of the extremest kind. He regarded the living being as owing all its characteristics to an indwelling Physis or natural entity with whose "faculties" or powers it was the province of physiology to deal. The living organism according to him has a kind of artistic creative power, a t6xvv> which acts on the things around it by means of the faculties, Swd/xei's, by the aid of which each part attracts to itself what is useful and good for it, rb oUelov, and repels what is not, to aXXorptov. These faculties, such as the "peptic faculty" in the stomach and the "sphygmic faculty" in the heart, are regarded by Galen as the causes of the specific functions or activity of the part in question. They are ultimate biological categories, for, although he admits the theoretical possibility of analysing them into simpler components, he never makes any attempt to do so, and evidently regards such an effort as doomed to failure, unlike Roux, whose "interim biological laws" are really conceived of as interim. "The effects of Nature", says Galen, "while the animal is still being formed in the womb are all the different parts of the body, and after it has been born an effect in which all parts share is the progress of each to its full size and thereafter the maintenance of itself as long as possible." Galen divides the effects of the faculties into three. Genesis, Growth, and Nutrition, and means by the first what we mean by embryogeny. "Genesis", he says, "is not a simple activity of Nature, but is compounded of alteration and of shaping. That is to say, in order that bone, nerve, veins, and all other tissues may come into existence, the underlying substance from which the animal springs must be altered; and in order that the substance so


altered may acquire its appropriate shape and position, its cavities, outgrowths, and attachments, and so forth, it has to undergo a shaping or formative process. One would be justified in calling this substance which undergoes alteration the material of an animal, just as wood is the material of a ship and wax of an image." In this remarkable passage, Galen expresses modern views about chemical growth and chemical differentiation.

Galen then goes on to treat of embryogeny in more detail. "The seed having been cast into the womb or into the earth — for there is no difference — ", he says (see p. 65), "then after a certain definite period a great number of parts become constituted in the substance which is being generated; these differ as regards moisture, dryness, coldness and warmth, and in all the other qualities which naturally derive therefrom", such as hardness, softness, viscosity, friability, lightness, heaviness, density, rarity, smoothness, roughness, thickness, and thinness. "Now Nature constructs bone, cartilage, nerve, membrane, ligament, vein, and so forth at the first stage of the animal's genesis, employing at this task a faculty which is, in general terms, generative and alterative, and, in more detail, warming, chilHng, drying and moistening, or such as spring from the blending of these, for example, the bone-producing, nerve-producing, and cartilageproducing, faculties (since for the sake of clearness these terms must be used as well) .... Now the peculiar flesh of the liver is of a certain kind as well, also that of the spleen, that of the kidneys and that of the lungs, and that of the heart, so also the proper substance of the brain, stomach, oesophagus, intestines and uterus is a sensible element, of similar parts all through, simple and uncompounded. . . . Thus the special alterative faculties in each animal are of the same number as the elementary parts, and further, the activities must necessarily correspond each to one of the special parts, just as each part has its special use. . . . As for the actual substance of the coats of the stomach, intestine, and uterus, each of these has been rendered what it is by a special alterative faculty of nature; while the bringing of these together, the combination therewith of the structures that are inserted into them, etc. have all been determined by a faculty which we call the shaping or formative faculty; this faculty we also state to be artistic — nay, the best and highest art — doing everything for some purpose, so that there is nothing ineffective or superfluous, or capable of being better disposed."


Thus the alterative faculty takes the primitive unformed raw material and changes it into the different forms represented by the different tissues, while the formative faculty, acting teleologically from within, organises these building-stones, as it were, into the various temples which make up the Acropolis of the completed animal. Galen next goes on to speak of the faculty of growth. "Let us first mention", he says, "that this too is present in the foetus in utero as is also the nutritive faculty, but that at that stage these two faculties are, as it were, handmaids to those already mentioned, and do not possess in themselves supreme authority."

Later on, until full stature is reached, growth is predominant, and finally nutrition assumes the hegemony.

So much for Galen's embryological theory. But before leaving the treatise On the Natural Faculties, it may be noted that he ascribes a retentive faculty to the uterus as well as to the stomach, and explains birth as being due to a cessation of action on the part of the retentive faculty, "when the object of the uterus has been fulfilled", and a coming into action of a hitherto quiescent propulsive faculty. This wholesale allotting of faculties can obviously be made to explain anything, and is eminently suited to a teleological account such as Galen's. It was not inconvenient as a framework within which all the biological knowledge of antiquity could be crystallised, but it was utterly pernicious to experimental science. Fifteen hundred years later it received what would have been the death-blow to any less virile theory, at the hands of Moliere in his immortal Malade Imaginaire :

Bachelirius. Mihi a docto doctore

Demandatur causam et rationem quare Opium facit dormire A quoi respondeo Quia est in eo Virtus dormitiva Cujus est nature Sensus assoupire. Chorus. Bene, bene, bene, bene respondere. Dignus, dignus est entrare In nostro docto corpore. Bene, bene, respondere.

But to return to Galen. The book on the formation of the embryo opens with a historical account of the views of the Hippocratic writers


with whom Galen was largely in agreement. It goes on to describe the anatomy of allantois, amnios, placenta, and membranes with considerable accuracy. The embryonic life consists, it says, of four stages: (i) an unformed seminal stage, (2) a stage in which the tria principia (a concept here met with for the first time) are engendered, the heart, liver and brain, (3) a stage when all the other parts are mapped out and (4) a stage when all the other parts have become clearly visible. Parallel with this development, the embryo also rises from possessing the life of a plant to that of an animal, and the umbilicus is made the root in the analogy with a plant. The embryo is formed, firstly, from menstrual blood, and secondly, from blood brought by the umbilical cord, and the way in which it turns into the embryo is made clearer as follows: "If you cut open the vein of an animal and let the blood flow out into moderately hot water; the formation of a coagulum very like the substance of the liver will be seen to take place". And in effect this viscus, according to Galen, is formed before the heart.

Galen also taught that the embryo excreted its urine into the allantois, and was acquainted with foetal atrophy. He gave a fairly correct account of the junction of the umbilical veins with the branches of the portal vein, and the umbilical with the iliac arteries, of the foramen ovale, the ductus Arantii and the ductus Botalli. He maintained that the embryo respired through the umbilical cord, and said that the blood passed in the embryo from the heart to the lungs and not vice versa. The belief that male foetuses were formed quicker than female ones he still entertained, and explained as being due to the superior heat and dryness of the male germ. He also associated the male conception with the right side and the female with the left and asserted that the intra-uterine movements are sooner felt in the case of the male than in the case of the female. Dry foods eaten by the mother, he thought, would lead to a more rapid development of the foetus than other kinds.

In this account of the Galenic embryology I have drawn not only upon the book on the formation of the foetus, but also upon his v7r6fMV7]/jba, Commentary on Hippocrates, his Trepl alricov av/jLTTTco/naTcov, On the Causes of Symptoms, and his book Trepl %peta? tmv fjuoplcor, On the Use of Parts. It is this latter work that had the greatest influence on the ages which followed Galen's Hfe. In the course of seventeen books, he tries to demonstrate the value and teleological significance of every


structure and function in the human and animal body, and to show that, being perfectly adapted to its end, it could not possibly be other in shape or nature than what it is. At the conclusion of this massive work with all its extraordinary ingenuity and labour, he says, "Such then and so great being the value of the argument now completed, this section makes it all plain and clear like a good epode — I say an epode, but not in the sense of one who uses enchantments (eVwSat?) but as in the melic poets whom some call lyric, there is as well as strophe and antistrophe, an epode, which, so it is said, they used to sing standing before the altar as a hymn to the Gods. To this then I compare this final section and therefore I have called it by that name". This is one of the half-dozen most striking paragraphs in the history of biology ; worthy to rank with the remarks of Hippocrates on the " Sacred Disease". Galen, as he wrote the words, must have thought of the altar of Dionysus in the Athenian or Pergamene theatre, made of marble and hung about with a garland, but they were equally applicable to the altar of a basilica of the Christian Church with the bishop and his priests celebrating the liturgy at it. What could be more charged with significance than this? At the end of the antique epoch the biology of all the schools, Croton, Akragas, Cos, Cnidus, Athens, Alexandria, Rome, is welded together and as it were deposited at the entrance into the sanctuary of Christendom. It was the turning-point, in Spengler's terminology, between ApoUinian civilisation and Faustian culture. Galen's words are the more extraordinary, for he himself can hardly have foreseen that the long line of experimentalists which had arisen in the sixth century B.C. would come to an end with him. But so it was to be, and thenceforward experimental research and biological speculation were alike to cease, except for a few stray mutations, born out of due time, until in 1453 the city of Byzantium should burst like .a ripe pod and, distributing her scholars all over the West, as if by a fertilising process, bring all the fruits of the Renaissance into being.



2-1. Patristic, Talmudic, and Arabian Writers

We are now at the beginning of the second century a.d. The next thousand years can be passed over in as short a time as it has taken to describe the embryology of Galen alone. The Patristic writers, who on the whole were careful to base their psychology on the physiology of the ancients, had little to say about the developing embryo. Most of their interest in it was, as would naturally be expected, theological; Tertullian, for instance, held that the soul was present fully in the embryo throughout its intra-uterine life, thus denying that kind of psychological recapitulation which had been suggested by Aristotle. "Reply," he says in his De Anima, "O ye Mothers, and say whether you do not feel the movements of the child within you. How then can it have no soul? " These views were not held by other Fathers, of whom St Augustine of Hippo {De Immortalitate et de quantitate ahimae) may serve as a representative, for he thought that the embryo was "besouled" in the second month and "besexed" in the fourth. These various opinions were duly reflected in the law, and abortion, which had even been recommended theoretically by Plato and defended practically by Lysias in the fourth or fifth century B.C., now became equivalent to homicide and punishable by death. This fact leads Singer to the view that the Hippocratic oath is late, perhaps early Christian. The late Roman law, which, according to Spangenberg, regarded the foetus as not Homo'", not even Infans'\ but only a Spes animantis'\ was gradually replaced by a stern condemnation of all pre-natal infanticide. "And we pay no attention", said the Bishops of the Quinisext Council, held at Byzantium in 692, "to the subtle distinction as to whether the foetus is formed or unformed." Other authorities, following St Augustine, took a more liberal view, and the canon law as finally crystallised recognised first the fortieth day for males and the eightieth day for females as the moment of animation, but later the fortieth day for both sexes. The embryo informatus" thus had no soul, the


^'^ embryo formatus" had, and as a corollary could be baptised. St Thomas Aquinas was of opinion that embryos dying in utero might possibly be saved : but Fulgentius denied it. As for the ancient belief that male embryos were formed twice as quickly as female ones, it lingered on until Goelicke took the trouble to disprove it experimentally in 1723.

Clement of Alexandria, in his book \6<yo<i TrporpeTTriKO'i Trpo? "EX\.7]va'i, has some remarks to make on embryology, but adds nothing to the knowledge previously gained. He adopts the Peripatetic view that generation results from the combination of semen with menstrual blood, and he uses the Aristotelian illustration of rennet coagulating milk. Lactantius of Nicomedia, who lived about the date of the Nicene Council (a.d. 325) perpetuated the deeply-rooted association of male with right and female with left in his book On the work of God, De opificio Dei. He also maintained that the head was formed before the heart in embryogeny, and seems to have opened hen's eggs systematically at different stages, so that to this extent he was a better embryologist than Galen. St Gregory of Nyssa, as we have already seen (p. 20), evolved a neo-vitalistic theory which he applied to the growth of the embryo.

Late Latin writers, other than the theologians, do not say much about it. There is a passage in Ausonius, however, which describes the development of the foetus {Eclog. de Rat. puerp.) but it is almost wholly astrological. Elsewhere he says:

juris idem tribus est, quod ter tribus; omnia in istis; forma hominis coepti, plenique exactio partu, quique novem novies fati tenet ultima finis.

Idyll II (Gryphus ternarii numeri), 4-6. (The power of 3, in 3 times 3 lies too, Thus 9 rules human form and human birth, And 9 times 9 the end of human life.)

But this is probably a late echo of the Pythagoreans rather than an early prelude to Leonardo da Vinci and the mathematisation of nature.

That great mass of Jewish writings known as the Talmud, which grew up between the second and sixth centuries a.d., also contains some references to embryology, and certain Jewish physicians, such as Samuel-el-Yehudi, of the second century, are said to have devoted


special attention to it. The embryo was called peri habbetten (fruit of the body), ]a2n ns. It grew through various definite stages:

(i) golem (formless, rolled-up thing), nbu, 0-1-5 months.

(2) shefir meruqqdm (embroidered foetus), api» T'Dit.

(3) ^ubbar (something carried), imi?, 1-5-4 months.

(4) walad (child), n*?!, 4-7 months.

(5) walad shel qaydmd (viable child), so'^^p '7tri'?i, 7-9 months.

(6) ben she-kallu khaddshdw (child whose months have been completed), rirnn I'^rir ]n.

The ideas of the Talmudic writers on the life led by the embryo in utero are well represented by the remark, "It floateth like a nutshell on the waters and moveth hither and thither at every touch"

ms o*» "rtr ':'SDn niia TUNb las •'^lan n»n n'?i rrch ity'^s •'sn lasi

And the classical passage, "Rabbi Simlai lectured: the babe in its mother's womb is like a rolled-up scroll, with folded arms lying closely pressed together, its elbows resting on its hips, its heels against its buttocks, its head between its knees. Its mouth is closed, its navel open. It eats its mother's food and sips its mother's drink: but it doth not excrete for fear of hurting"

bv rT* niioi "rsipa'!^ Q^ith las "'yan n»n n'^in rxh V^b's^^ •'in tJ^m ittNtr na» nmtyi n'?sis las:^ n»» '?2isi mns "nuai miio rsi rsin ^■'n i"?

It was thought, moreover, that the bones and tendons, the nails, the marrow in the head and the white of the eye, were derived from the father, "who sows the white", but the skin, flesh, blood, hair, and the dark part of the eye from the mother, "who sows the red". This is evidently in direct descent from Aristotle through Galen, and may be compared with the following passage from the latter writer's Commentary on Hippocrates: "We teach that some parts of the body are formed from the semen and the flesh alone from blood. But because the amount of semen which is injected into the uterus is small, growth and increment must come for the most part from the blood". It might thus appear that, just as the Jews of Alexandria were reading Aristotle in the third century B.C., and incorporating




him into the Wisdom Literature, so those of the third century a.d. were reading Galen and incorporating him into the Talmud. As for God, he contributed the life, the soul, the expression of the face, the functions of the different parts. This participation of three factors in generation, male, female, and god, is exceedingly ancient, as may be read in Robertson Smith. Some Talmudic writers held that development began with the head, agreeing with Lactantius, and others that it began at the navel, agreeing with Alcmaeon. Weber has given an account of the Talmudic beliefs about the infusion of the soul into the embryo. They do not seem to have embodied any new or striking idea.

Although the Talmud contained certain references of embryological interest, the first Hebrew treatise on biology was not composed till the tenth century, when Asaph Judaeus or Asaph-ha-Yehudi wrote on embryology about a.d. 950. His MSS. are exceedingly rare, but, according to Gottheil's description, they contain several sections on embryology. Steinschneider has given another description of them. For further details on the whole subject of Jewish embryology see Macht.

Arabian science, so justly famed for its successes in certain branches, was not of great help to embryology. Abu-1-Hasan ' Ali ibn Sahl ibn Rabban al-Tabari, a Moslem physician who flourished under the Caliphate of al-Mutawakldl about a.d. 850, wrote a book called The Paradise of Wisdom, in which an entire part was devoted to embryology, all the more interesting as it is a mixture of Greek and ancient Indian knowledge. Browne gives a description of it. Ibn Rabban's contemporary, Thabit ibn Qurra, is also said to have written on embryology. The great Avicenna, or, to give him his proper name, Abu 'Ali-1-Hasan ibn 'Abdallah ibn Sina, who lived from 978 to 1036, devoted certain chapters of his Canon Medicinae to the development of the foetus, but added nothing to Galen. His contemporaries, Abu-1-Qasim Maslama ibn Ahmad al-Majriti and Arib ibn Said al-Katib, a Spanish Moslem, wrote treatises on the generation of animals, but neither has survived.

What was alchemy doing all this time? It was engaged on many curious pursuits, but among them the interpretation of embryonic development was not one. Alchemical texts before the tenth century do make reference to eggs from time to time, but it is safe to say never with any trace of an interest in the development of the embryo


out of them. One example taken from Berthelot's collection will suffice; it comes from the "6th book of the Philosopher" (Syriac).

To make water of eggs

Take as many eggs as you wish, break them and put the whites in a glass flask, place this in another vessel and surround it with fresh horsedung up to the neck of the vessel. Leave it so for 15 days changing the dung every 5 days. Then distil the liquid in an alembic and taking a pound of the distillate add lime of eggs 2 ozs. Shake well and distil again. Do this 4 times. Take then of elixir of arsenic, 2 parts, of sulphur i part, of pyrites and magnesia, each i part. Pound in a mortar and add to the final distillate from the eggs. Do this for 7 days always working in the sunlight, once at sunrise, once in the middle of the day, and once at sunset. When this has been done, dry the mixture, pound it, and set it aside.

I could only find one reference to the embryo in a hen's egg among the vast number of alchemical directions of this time, and then only as a constituent of the egg which must be discarded. As we shall see, it is not until after the time of Paracelsus that the notion of applying chemical methods to eggs or embryos arises at all.

2-2. St Hildegard: the Lowest Depth

Not long after the death of Avicenna, St Hildegard was born. She lived from 1098 to 1180, and was Abbess successively of Disibodenberg and Bingen in the Rhineland. Her treatises on the world, which are an extraordinary medley of theological, mystical, scientific and philosophical speculation, have been described in detail by Singer, and, though in the books. Liber Scivias and Liber Divinorum Operum simplicis hominis, there is little of embryological interest, yet she does give an account of development and especially of the entry of the soul into the foetus.

This is illustrated in Plate HI taken from the Wiesbaden Codex B of the Liber Scivias. The soul is here shown passing down from heaven into the body of the pregnant woman and so to the embryo within her. The divine wisdom is represented by a square object with its angles pointing to the four corners of the earth in symbol of stabihty. From it a long tube-Hke process descends into the mother's womb and down it the soul passes as a bright object, "spherical" or "shapeless", illuminating the whole body. The scene shows the mother in the foreground lying down ; inside her there are traces of the foetal membranes; behind this ten persons are grouped, each carrying a


vessel, into one of which a fiend pours some noxious substance from the left-hand corner. St Hildegard describes and expounds the scene as follows: "Behold, I saw upon earth men carrying milk in earthen vessels and making cheeses therefrom. Some was of the thick kind from which firm cheese is made, some of the thinner sort from which more porous cheese is made, and some was mixed with corruption and of the sort from which bitter cheese is made. And I saw the likeness of a woman having a complete human form within her womb. And then by a secret disposition of the most high craftsman, a fiery sphere having none of the lineaments of a human body possessed the heart of the form and reached the brain and transfused itself through all the members. . . . And I saw that many circling eddies possessed the sphere and brought it earthward, but with ever renewed force it returned upwards and wailed aloud, asking, 'I, wanderer that I am, where am I?' 'In death's shadow.' 'And where go I?' 'In the way of sinners.' 'And what is my hope? ' ' That of all wanderers.' . . . As for those whom thou hast seen carrying milk in earthen vessels, they are in the world, men and women alike, having in their bodies the seed of mankind from which are procreated the various kinds of human beings. Part is thickened because the seed in its strength is well and truly concocted and this produces forceful men to whom are allotted gifts both spiritual and carnal.. . .And some had cheeses less firmly curdled, for in their feebleness they have seed imperfectly tempered and they raise offspring mostly stupid, feeble, and useless, . . . And some was mixed with corruption . , . for the seed in that brew cannot be rightly raised, it is invalid, and makes misshapen men who are bitter distressed and oppressed of heart so that they may not lift their gaze to higher things. . . .And often in forgetfulness of God and by the mocking devil a mistio is made of the man and the woman and the thing born therefrom is deformed, for parents who have sinned against me return to me crucified in their children". We have already traced the wanderings of the cheese-analogy, which, beginning fresh with Aristotle, was taken to Alexandria and incorporated in the Wisdom Literature, and so found its way to the Arabic of 'Ali ibn a'1-Abbas al-Majusi, or Haly-Abbas, as he was known in the West, a Persian. His Liber Totius appeared in Latin in 1523, but had been translated much earlier, at Monte Cassino between 1070 and 1085, by Constantine the African, who called it Liber de Humana Natura, and gave it out to be his own work. Thus


AN ILLUSTRATION FROM THE LIBER SCIVIAS OF ST HILDEGARD OF BINGEN (Wiesbaden Codex B) showing the descent of the soul into the embryo {ca. 1 150 a.d.).


St Hildegard obtained it, and worked it up into one of her visions. At this point embryology touched, perhaps, its low-water mark. But a great man was at hand, destined to carry on the Aristotelian tradition and to add to it much of originality, in the shape of Albertus of Cologne. Before speaking of him, however, a word must be said about that very queer character, Michael Scot (i 178-1234), who, according to Gunther, "appeared in Oxford in 1230 and experimented with the artificial incubation of eggs, having got an Egyptian to teach him how to incubate ostriches eggs by the heat of the Apulian sun". That "muddle-headed old magician", as Singer rightly calls him, was not the man to profit by it, but the point is interesting, especially as an Egyptian is mentioned. Haskins, in his curious studies of the scientific atmosphere of the court of the Emperor Frederick II of Sicily, has shown Scot, newly arrived fi"om his alchemical studies in Spain, assisting that very learned and unorthodox monarch in his artificial incubation experiments.

2-3. Albertus Magnus

Albertus Magnus of Cologne and Bollstadt was born in 1206, and died in 1280, six years after his favourite disciple, St Thomas Aquinas. The greater part of his life was spent in study and teaching in one or other of the houses of the Dominican friars, to which he belonged, though for a time he was Bishop of Regensburg. Albert resembles Aristotle in many points, but principally because he produced biological work with, as it were, no antecedents. Just as Aristotle's contributions to embryology were preceded by no more than the diffuse speculations of the Ionian nature-philosophers, so Albert's came immediately after the dead period represented by the visions of St Hildegard. In many ways, Albert's position was much less conducive to good work than Aristotle's.

Albert follows Aristotle closely throughout his biological writings, quoting him word for word in large amounts, but the significant thing is that he does not follow him slavishly. He resembled Aristotle in paying much attention to the phenomena of generation, as a rough computation shows, Aristotle devoting 37 per cent, of his biological writings to this subject, and Albert 31 per cent., to which Galen's 7 per cent, may with interest be compared. Albert is extremely inferior to Aristotle, however, in point of arrangement; for Aristotle, although some of his books, such as the De Generatione Animalium,


are sufficiently confused and repetitive, does yet succeed in infusing a clarity and incisiveness into his style. Albert, on the other hand, allows his argument to wander through his twenty-six books De Animalibus in the most complex convolutions, so that the sections on generation and embryology are found indiscriminately in the first, sixth, ninth, fifteenth, sixteenth, and seventeenth. In Book i he gives a kind of summary or skeleton of his views on the embryo. These follow Aristotle fairly closely; thus, he accepts the AristoteHan classification of animals according to their manner of generation, and thinks still that caterpillars are immature eggs ; he derives the embryo from the white, not the yolk, and he explains why soft-shelled eggs, being imperfect, are of one colour only. But there are new observations; for instance, he describes an ovum in ovo, which he has seen, calling it a natura peccatis, and he speaks definitely of the seed of the woman, thus departing from Peripatetic opinion, and adopting the Epicurean view. The female seed, he thinks, suflfers coagulation like cheese by the male seed, and to these two humidities there must be added a third, namely, the menstrual blood (corresponding to the yolk in the case of the bird). "When these three humidities therefore have been brought into one place, all the similar members except the blood and fat are formed from the two humidities of which one generates actively but the other passively. But the blood which is attracted for the nutriment of the embryo is double in virtue and double in substance. For a certain part of the blood is united with the sperm in such a way that it takes on some of the virtue of the seed because a certain part of the spermatic humour remains in it and from this are begotten the teeth and for this reason they grow again if they are pulled out at an age near the time of sperm-making and do not grow again at an age remoter from this, at which the virtue of the first generating principle has vanished from the blood. But another part of the blood is of twofold or threefold substance and from the thick part of the blood itself is generated the flesh. And this flows in and flows out and grows again if rubbed away. From the watery part of the same blood or of the nutritive humour are generated the fat and oil and this flows in and out more easily than the flesh itself, but other parts of the blood are its refuse and impurities and are not attracted to the generation of any part of the animal, but having been collected until birth are expelled with the embryo from the uterus in the foetal membranes, like the remnants


in the hen's egg after the chick has hatched. There is a similar virtue in the liver and heart of animals which organs after the animals are born form the flesh and fat from food in accordance with its twofold substance, and expel the refuse as we said before,"

In the sixth book, Albert contradicts Aristotle's opinion that male chick develops out of the sharp-ended egg, and one hopes that he is going to say there is no relationship between egg-shape and sex, but no, he goes on to say that the Aristotelian statement rested on a textual error (in which he was quite wrong), so that really Aristode agreed with Avicenna in saying that the males always develop from the more spherical eggs because the sphere is the most perfect of figures in solid geometry. These errors had a most persistent life : Horace has a passage in which they appear —

longa quibus facies ovis erit, ilia memento ut suci melioris, et ut magis alma rotundis ponere: namque marem cohibent callosa vitellu.m.

(When you would feast upon eggs, make choice of the long ones ; they are whiter and sweeter and more nourishing than the round, for being hard they contain the yolk of the male.)

They were finally abolished by two naturalists, Giinther and Biihle, who took the trouble to disprove them experimentally in the eighteenth century. Albertus refers here to artificial incubation: "For the alterative and maturative heat", he says, "of the egg is in the egg itself and the warmth which the bird provides is altogether external [extrinsecus est amminiculans] since in certain hot countries the eggs of fowls are put under the surface of the earth and come to completion of their own accord, as in Egypt, for the Egyptians hatch them out by placing them under dung in the sunlight". Next he speaks of monsters and of the modes of corruption of eggs which he divides into four: (i) decomposition of white, (2) decomposition of yolk, (3) bursting of the yolk-membrane, (4) antiquitas ovi. "And from the second cause it sometimes happens ", he says, "that in the corruption of the humours certain igneous parts are carried blazing to the shell of the egg and distribute themselves over it so that it shines in the dark like rotten wood; as happened in the case of that egg^ which Avicenna said he saw in the city called Kanetrizine in the country of the Gorascenes." Albert

^ See on this subject Zach.



is inclined to think that astrological influences may have an effect on foetal life, but he treats the suggestion with considerable scepticism, although he believes that thunder and lightning kill the embryos of fowls (a popular belief to which Fere tried not long ago to give a scientific foundation), and he regards the embryo of the crow as especially susceptible, though on what grounds he does not say.

The fourth chapter of the first tractate of the sixth book contains Albert's description of development of the chick, and is extremely interesting. He makes two principal mistakes: {a) he describes a quite non-existent fissure in the shell by which the chick may emerge, {b) he maintains that the yolk ascends after a day or two into the sharp end of the egg, adducing as the reason that there is found there more heat and formative force than elsewhere. On the other hand, he correctly describes {a) the pulsating drop of blood on the third day, and {b) he identifies it with the heart with its systolen et dyastolen sending out the "formative virtue" to all the parts of the growing body. He notices [c) that the differentiation of the chick at first proceeds rapidly and later more slowly. But the most notable characteristic of Albert's embryology is the way in which he is hampered by his inability to invent a technical terminology. Singer has studied the way in which anatomical terms, such as "syrach", etc., came into use, but whatever the causes were which produced them, they did not operate much in Albert's mind. He represents the point beyond which embryology could not advance, until it had created a new set of terms. This is well illustrated by the following passage:

"But fi'cni the drop of blood", he says, "out of which the heart is formed, there proceed two vein-like and pulsatile passages and there is in them a purer blood which forms the chief organs such as the liver and lungs and these though very small at first grow and extend at last to the outer membranes which hold the whole material of the egg together. There they ramify in many divisions, but the greater of them appears on the membrane which holds the white of the egg within it [the allantois]. The albumen, at first quite white, is changed owing to the power of the vein almost to a pale yellow-green tint [palearem colorem]. Then the path of which we spoke proceeds to a place in which the head of the embryo is found carrying thither the virtue and purer material from which are formed the head and the brain, which is the marrow of the head. In the formation of the head also are found the eyes and because they are of an aqueous humidity which is with difficulty used up by the first heat they are very large, swelling out and bulging from the chick's head. A short


time afterwards, however, they settle down a little and lose their swelling owing to the digestive action of the heat — and all this is brought about by the action of the formative virtue carried along the passage which is directed to the head, but before arriving there is separated and ramified by the great vein of the albumen-membrane, as may be clearly seen by anyone who breaks an egg at this time and notes the head appearing in the wet part of the egg and at the top of the other members. For what appears first in the making of a foetus are the upper parts because they are nobler and more spiritual being compacted of the subtler part of the egg wherein the formative virtue is stronger. When this happened one of the aforementioned two passages which spring from the heart branches into two, one of them going to the spiritual part which contains the heart and divides there in it carrying to it the pulse and subtle blood from which the lungs and other spiritual parts are formed, and the other going through the diaphragm \dyqfracmd\ to enclose within it at the other end the yolk of the Qgg, around which it forms the liver and stomach. It is accordingly said to take the place of the umbilicus in other animals and through it food is drawn in to supply the flesh for the chick's body, for the principle of generation of the radical members of the chick comes from the albumen but the food from which is made the flesh filling up all the hollows is from the yolk."

After ten days, Albert goes on to say, all the constituent organs are mapped out and the head is greater then than the rest of the body put together. He observes that the yolk liquefies early in development and that slimy concretions are present in the allantoic fluid later on (uric acid). But the passage quoted does demonstrate that before further progress could be made some better name must be found than "the interior membrane to which the first vessel proceeds" for a given structure.

Albert, however, was accomplishing a good work. One of his best amplifications of Aristotle was his description of the relationship between yolk and embryo in fishes. Just as his words about the chick demonstrate that he must have opened hen's eggs at different stages during incubation, so his words about fish eggs show that he must have dissected and examined them also. Thus (Book vi, tractate 2, chap, i) he says, "Between the mode of development [anathomiam generationis] of birds' and fishes eggs there is this diflference ; during the development of the fish the second of the two veins which extend from the heart does not exist. For we do not find the vein which extends to the outer covering of the eggs of birds which some wrongly call the umbilicus because it carries the blood to the outside parts, but we do find the vein which corresponds to the yolk vein of birds, for this vein imbibes the nourishment by which the limbs increase.


Therefore the generation of the fish embryo begins from the sharp end of the egg like that of birds and channels extend from the heart to the head and eyes and first in them appear the upper parts. As the growth of the young fish proceeds the yolk decreases in amount being incorporated into the members and it disappears entirely when development is complete. The beating of the heart, which some call panting, is transmitted through the pulsating veins to the lower part of the belly carrying life to the inferior members. While the young fish are small and not yet fully developed they have veins of great length which take the place of the umbilicus, but as they grow these shorten till they contract into the body by the heart as has been said about birds. The young fish are enclosed in a covering just like the embryos of birds, which resembles the dura mater and beneath it another containing the foetus and nothing else, while between the two there is the moisture rejected during the creation of the embryo". Albert also described ovoviviparous fishes but it is more difficult in that case to tell whether he had himself seen and dissected them. He notes also the prodigality of nature in producing so many marine eggs only destined to be eaten.

In Books IX and xv he treats of the Galenic views on generation and insists again that there is a seed provided by the female. In Book XVI he gives his opinions about the animation of the embryo, quoting the views of the ancients as given in Plutarch, e.g. Alexander the Peripatetic, Empedocles, Anaxagoras, Theodorus and Theophrastus, the Peripatetics, Socrates, Plato, the Stoics, Avicenna, and Aristotle, "who saw the truth", but— and it is interesting to notice it — never the Christian Fathers, whose writings must have been well known to him. In discussing the Aristotelian views he compares the menstrual blood to the marble and the semen to the man with a chisel in his hand.

On the question of epigenesis and preformation, he follows Aristotle almost word for word, using the same analogies, such as the "dead eye" and the sleeping mathematician. Here his scholasticism comes out clearly, for in rejecting altogether the theory that one part being formed then forms the next part, he says, not that A would have to be in some way like B, but is not, as Aristotle had, but simply "^Generans et generatum, est simul esset et non esset, quod omnino est impossibile^ — a high-handed and very unscientific manner of settling the question. In conformity with his theology and


in contradistinction from Aristotle he makes the vegetative and sensitive souls arrive automatically into the embryo but the rational soul only by a direct act of God.

His mammalian embryology presents some points of interest. He follows Hippocrates ("Ypocras") in an account of the co-operation of heat and cold in member-formation, and he holds very enlightened views about foetal nutrition, "It appears therefore that the embryo hangs from the cord and that the cord is joined with the vein and that the vein extends through the uterus and has blood running through it to the foetus like water through a canal. Round the embryo there are membranes and webs as we have seen. But those who think that the embryo is fed by little bits of flesh through the cord are wrong and lie, because if this were the case with man it would happen also with other animals and that it does not do so anybody can find out by investigation [per anathomyani].'"

Finally, it is typical that in Book xvii Albert repeats what he has already said in Book vi about the generation of the hen out of the tgg all over again with slight changes, but he adds the significant biochemical remark that "eggs grow into embryos because their wetness is like the wetness of yeast". The importance of Albert in the history of embryology is clear. With him the new spirit of investigation leapt up into being, and, though there were many years yet to pass before Harvey, the modern as opposed to the ancient period of embryology had begun. Albert's writings were often copied and printed in the next few centuries, and even as late as 1601 De Secretis Mulierum, an epitome of his books on generation, was published. In some sense, it still is, as it forms the backbone of the little book Aristotle's Masterpiece, of which thousands of copies are sold in England every year. The copy of the De Secretis in the Caius College Library has written across the title-page in faded ink "Simulacra sanctitas, duplex iniquitas, Nathan Emgross, Nov. 20. 161 3." But in spite of Mr Emgross, Albertus, rightly called Magnus, has had the happy fate of being beatified both by the Church and by science.

2-4. The Scholastic Period

St Thomas Aquinas (i 227-1 274) incorporated the Aristotelian theories of embryology into his Summa Theologica especially under the head De propagatione hominis quantum ad corpus. There are some striking passages, such as "The generative power of the female


is imperfect compared to that of the male; for just as in the crafts, the inferior workman prepares the material and the more skilled operator shapes it, so likewise the female generative virtue provides the substance but the active male virtue makes it into the finished product". How admirably this expresses the dominating sentiment of the Middle Ages! Aristotle might make a distinction between matter and form in generation, but the mediaeval mind, with its perpetual hankering after value, would at once enquire which of the two was the higher, the nobler, the more honourable.

St Thomas' theory of embryonic animation was complicated. He had a notion that the foetus was first endowed with a vegetative soul, which in due course perished, at which moment the embryo came into the possession of a sensitive soul, which died in its turn, only to be replaced by a rational soul provided directly by God, This led him into great difficulties, for, if this scheme were true, it was difficult to say that man generated man at all; on the contrary he could hardly be said to generate more than a sensitive soul which died before birth, and, on this view, what was to happen to original sin? As Harris has put it, Plato had said that the intellect was the man, using the body as a boatman uses a boat. Averroes had said precisely the opposite, namely, that the essence of humanity was in the body, and that the intellect was something extrinsic, not limited to the individual, but common to the race. Aristotle had taken the middle position, and given a soul to plants and animals, but, in doing so, he had made it into a vital rather than a psychological principle. The task of combining this -^vxv with the anima of the Fathers was what scholastic philosophy had before it. No wonder that St Thomas' account of embryonic animation was open to criticism. An echo of it appears in a poem of Jalalu'd-Din Rumi ( 1 207-1 273), the greatest of the Persian Sufi poets, and an exact contemporary of St Thomas Aquinas :

I died from mineral and plant became: Died from the plant, and took a sentient frame; Died from the beast, and donned a human dress; When by my dying did I e'er grow less?

Duns Scotus (i 266-1 308) objected to St Thomas' theory on the grounds already mentioned, and he himself abandoned the vegetative and sensitive souls altogether in his De Rerum Principio. This solution


was no better than that of St Thomas, for, agreeing with the latter as Duns did that the rational soul was not an ordinary form "educed " from the "potentiality" of the material, but rather an ad hoc creation of God, injected by divine power into the embryo at the appropriate moment, it was difficult to see how the spiritual effects of Adam's fall could be transmitted to the men of each generation. It was as if only acquired characteristics were inherited. But the further course of theological embryology need not be pursued here ; it runs in every century parallel with true scientific embryology, and it is not my purpose to do more than take a glance at its progress from time to time. In the Speculum Naturale, which was written about 1250, by Vincent of Beauvais, the embryology of Constantine the African appears again, and the embryology of Aristotle, Galen, and the scholastics is to be found in Dante Alighieri (i 265-1 321), who dealt with the subject in his Convivio, and especially in the Divina Commedia. In Canto XXV of the Purgatorio, Statins (the personification of human philosophy enlightened by divine revelation) is made to speak to the poet thus: "If thy mind, my son, gives due heed to my words and takes them home, they will elucidate the question thou dost ask. Perfect blood which is in no case drawn from the thirsty veins, but which remains behind like food that is removed from table, receives in the heart informing power for all the members of the human body, like the other blood which courses through the veins in order to be converted into those members. After being digested a second time it descends to the part whereof it is more seemly to keep silence than to speak, and thence it afterwards drops into the natural receptacle (the uterus) upon another's blood ; there the one blood and the other mingle. One is appointed to be passive, the other to be active according to the perfect place whence it proceeds (the heart). And being united with it, it begins to operate, first by coagulating it, and then by vivifying that to which it has given consistency, so that there may be material for it to work upon [e poi avviva, Cib che per sua materia fe' constare]. The active power having become a (vegetative) soul like that of a plant — only differing from it in this, that the former is in progress while the latter has reached its goal — thereafter works so much that it moves and feels like a sea-fungus and as the next stage it takes in hand to provide with organs the faculties which spring from it. At this point, my son, the power which proceeds from the heart of the begetter is expanded and developed, that power in which


Nature is intent on forming all the members, but how from being an animal it becomes a child, thou seest not yet, moreover this is so difficult a point that formerly it led astray one more wise than thou [Averroes], so that in his teaching he separated the active 'intellect' from the soul because he could not see any organ definitely appropriated by it. Open thy heart to the truth and know that as soon as the brain of the foetus is perfectly organised, the Prime Mover, rejoicing in this display of skill on the part of Nature, turns him towards it and infuses a new spirit replete with power into it which subsumes into its own essence the active elements which it finds already there, and so forms one single soul which lives and feels and is conscious of its own existence. And that thou mayst find my saying less strange, bethink thee how the heat of the sun passing into the juice which the grape distils, makes wine".

Having said this. Statins, Virgil and Dante pass on to the seventh ledge in Purgatory. It is interesting to see how Dante emphasises the dynamic teleological side of Aristotle and practically speaks of the soul enfleshing itself and arranging organs for its faculties. The reference to Averroes is explained by the fact that Averroes was a Traducianist, and held that all the soul was generated by man at the same time as the body, whereas both St Thomas and Dante, as Creationists, held that each fresh soul was a special creation of God inserted by him into the brain of the embryo. The mention of Dante's contemporary, Mondino de Luzzi (1270-1326), brings us to the more practical aspects of embryology at this period. Mondino is the most outstanding figure among the Bolognese anatomists of what is really the first period of the revival of biology. After him, as we shall see, biology languished for a couple of centuries until the advent of such men as Ulysses Aldrovandus in the sixteenth century, and Singer has shown that this was probably due to the fact that anatomy professors did not dissect in person. A fortiori embryotomy was infrequent.

But Mondino's Anathomia, published in 13 16, contained statements about the organs of generation which were rather important. He retains the notion of the seven-celled uterus, which had been introduced by Michael Scot, but he adopts a reasonable compromise between the opinions of Galen and Aristotle on the physiology of embryo formation. The distance between him and Leonardo da Vinci (1452-1519) would, however, be estimated rather at five or six centuries than at the century and a quarter that it actually was.


2-5. Leonardo da Vinci

Leonardo was not alone among the artists of the Renaissance in his anatomical interests, for Michael Angelo, Raphael, Diirer, Mantegna, and Verrochio all made dissections in order to increase their knowledge of the human body. But he penetrated more curiously into biology than they did, and he will always remain one of the greatest of biologists, for he first introduced the quantitative outlook. In this he was some four hundred years before his time.

Leonardo's embryology is contained in the third volume of his notebooks, Quaderni d' Anatomia, published in facsimile by the admirable labours of three Norwegian scholars, Vangensten, Fohnahn and Hopstock, in 191 1. His notebooks are a remarkable, and, indeed, charming miscellany of anatomical drawings, physiological diagrams, architectural and mechanical sketches and notes such as "Shirts, hose, and shoes", "Go and see Messer Andreas", "get coal", "the supreme fool (is the) necromancer, and enchanter".

His dissections of the pregnant uterus and its membranes are beautifully depicted, as can be seen from the figures which are here reproduced (Plate IV). He was acquainted with amnios and chorion, and he knew that the umbilical cord was composed only of vessels, though he seems to have thought the human placenta was cotyledonous. There is one drawing which the editors suppose to represent the developing hen's egg, but I do not feel that this ascription is likely. Indeed, Leonardo worked with eggs much less than with mammalian embryos, though there are references to the former. "See how birds are nourished in their eggs", he says in one place, to remind himself, perhaps, of possible experiments, and, elsewhere, "Chickens are hatched by means of the ovens of the fireplace". Again, "Ask the wife of Biagino Crivelli (was she the Lucrezia Crivelli, whose portrait Leonardo painted?) how the capon rears and hatches the eggs of the hen when he is inebriated", a subject recently reopened by Lienhart. "You must first dissect the hatched egg before you show the difference between the human liver in foetus and adult." Leonardo perpetuates a persistent error in the note, "Eggs which have a round form produce males, those which have a long form produce females".

Concerning the mammalian foetus, he says, "The veins of the child do not ramify in the substance of the uterus of its mother but in the placenta which takes the place of a shirt in the interior of the


uterus which it coats and to which it is connected but not united by means of the cotyledons". Thus in one sentence Leonardo falls into a mistake in saying that the human placenta is cotyledonous, but at the same time asserts a fact which it took all the ingenuity of the seventeenth century to prove to be true, namely, that the foetal circulation is not continuous with that of the mother, for the placenta is only connected to the uterine wall and not united with it. "The child", Leonardo goes on to say, "lies in the uterus surrounded with water, because heavy things weigh less in water than in the air and the less so the more viscous and greasy the water is. And then such water distributes its own weight with the weight of the creature over the whole body and sides of the uterus." The tendency towards quantitative and mathematical explanations is apparent at once.

Further notes are, "Note how the foetus breathes and how it is nourished through the umbilical cord and why one soul governs two bodies, as you see the mother desiring food and the child remaining marked (by a given amount of growth) because of it. Avicenna pretends that the soul generates the soul and the body the body. Per errata^'. The child, says Leonardo, secretes urine while still in utero, and has excrement in its intestines; at four months it has chyle in its stomach, made perhaps from menstrual blood. But it has no voice in utero, "when women say that the foetus is heard to weep sometimes within the uterus, this is rather the sound of some flatus . . . ". Nor does it breathe there (on this point Leonardo contradicts himself). "The child does not respire within the body of its mother because it lies in water and he who breathes in water is immediately drowned." "Breathing is not necessary to the embryo because it is vivified and nourished by the life and food of the mother." Nor does the embryonic heart beat. To us the statement that there is no respiration in the uterus is obviously false, but we mean by the word tissue respiration, whereas in Leonardo's time pulmonary respiration was intended; he was therefore perfectly right in denying that the embryo breathed, as certain anatomists before him had asserted.

His only reference to the soul runs thus: "Nature places in the bodies of animals the soul, the composer of the body, i.e. the soul of the mother, which first composes, in the womb, the shape of man and in due time awakens the soul which shall be the inhabitant thereof, which first remains asleep and under the tutelage of the soul of the mother which through the umbilical vein nourishes and vivifies


vf Hi^l


O V^xf^rtl

V-..,'.'.^/ I »-^ '•'

•V 'j'.'/** •»T^'"(' ♦'



it". This is not very revolutionary. But Leonardo was the first embryologist to make any quantitative observations on embryonic growth ; he defined, for instance, the length of a full-grown embryo as one braccio and the adult as three times that. "The child", he says, "grows daily far more when in the body of its mother than when it is outside of the body and this teaches us why in the first year when it finds itself outside the body of the mother, or, rather, in the first 9 months, it does not double the size of the 9 months when it found itself within the mother's body. Nor in 18 months has it doubled the size it was 9 months after it was born, and thus in every 9 months diminishing the quantity of such increase till it has come to its greatest height." Here Leonardo touches on one of the most modern quantitative aspects of embryology, and one almost expects to see him exemplify his words with a graph until one remembers with a shock that he lived two centuries before Descartes and five before Minot. His numerical data may also have included figures about the relative sizes of the parts, and the germ of the line of research so successfully pursued by Scammon in our own times may be found in the note "The liver is relatively much larger in the foetus than in the grown man". Other quantitative notes concern the length of the embryonic intestines as in the laconic "20 braccia of bowels" and the statement that "the length of the umbilical cord always equals the length of the foetal body in man though not in animals".^

He said little about heredity, but in one place he mentions a case of sexual intercourse between an Italian woman and an Ethiopian, the outcome of which assured him that blackness was not due to the direct action of the sun and that the "seed of the female was as potent as that of the male in generation". Finally, the best instance of the wideness of his thought appears in the note, "All seeds have an umbilical cord which breaks when the seed is mature. And similarly they have matrix and secundines as the herbs and all the seeds which grow in shells show". We have met this idea before in Hippocrates of Cos, and we shall find it again in Nathaniel Highmore.

It is no coincidence that pictures of weights and cogs and pulleys stand side by side in Leonardo's notes with anatomical drawings of the embryo. As Hopstock says, "Leonardo arrives at the conclusion that there is but one natural law which governs the world. Necessity.

^ Leonardo would have enjoyed Fog's statistical study of 8000 umbUical cords (1930).


Necessity is Nature's master and guardian, it is Necessity that makes the eternal laws". If Aristotle is the father of embryology regarded as a branch of natural history, Leonardo is the father of embryology regarded as an exact science.

2-6. The Sixteenth Century: the Macro- Iconographers

After such a man, the writings of his contemporaries, such as the mythical Johannes de Ketham, Alessandro Achillini and Gabriele de Gerbi, appear beyond description inferior. De Ketham's embryology has been described by Ferckel. De Gerbi included in his Liber Anatomiae corporis humani et singulorum membrorum illius a section entitled De Generatione Embrjonis, but there is nothing to be said about it except that it is a verbose compilation of the views of Aristotle and Galen taken from Avicenna. The work of Nolanus in 1532 presents certain points of interest, but it is of little importance. Petrus Crescentius in his work on husbandry of 1548 mentions artificial incubation in ovens, but rather as a lost art. About this time also Hieronymus Dandinus Cesenas, a Jesuit, wrote a treatise on Galen's division of organs into white and red, those proceeding from the semen and those proceeding from the blood: it is cited by Aldrovandus, but I have not been able to consult it.

The most remarkable feature of the first half of the century was the encyclopaedic group of zoologists which now arose. Thus Belon and Rondelet, whose well-illustrated catalogues of animals were appearing from 1550 onwards, did a good service to comparative embryology in figuring the ovoviviparous selachians and viviparous cetacea. Gesner belongs to this group. All of them reproduce thin versions of Aristotle, when they speak of generation as such, and this is what differentiates them from Ulysses Aldrovandus, of whom I shall speak presently. Figs. 3 and 4 show Rondelet's pictures of a viviparous dolphin and an ovoviviparous selachian.

But the end of the twilight period was now at hand, for, within thirty years after the death of de Gerbi in 1505, four great embryologists were born as well as the greatest anatomist of any age, Andreas Vesalius (1514), of whom I shall say no more, for he had no opportunities for dissecting human embryos, and took hardly any interest in foetal development. But in 1522 Ulysses Aldrovandus was born, and in the following year Gabriel Fallopius, in 1530 Julius Caesar Arantius and in 1534 Volcher Goiter. Only three

SECT. 2]



more years bring us to the birth of Andreas Laurentius and of Hieronymus Fabricius ab Aquapendente, the teacher of William Harvey.

The senior member of this group, Ulysses Aldrovandus, was the first biologist since Aristotle to open the eggs of hens regularly during

Fig. 3. A (viviparous) dolphin: from Rondelet's De piscibus marinis of 1554.

their incubation period, and to describe in detail the appearances which he found there. In his Ornithologia, published at Bonn in 1597, he set out to describe all the known kinds of birds, discussing in turn not only their zoological and physiological characteristics, but

Fig. 4. An (ovoviviparous) shark: from Rondelet's De piscibus marints of 1554.

also their significance as presages and for augury, their mystical meaning, their use as allegories and for eating, and finally all the legends respecting them, Generositas, Temperantia, Liberalitas, aquilae one finds. Beginning with the eagle, he proceeds to the vulture, the owl, the bat (the only viviparous bird!), the ostrich, the harpy (!),


the parrot, the crow, and so to the fowl. Side by side with a reference to the famous poem of Prudentius {Multi sunt Presbyteri, translated by J. M. Neale) about the steeple-cock, we find an excellent account of the generation of the chick in the c:gg. The book is illustrated sumptuously, but unfortunately there is only one picture of embryological interest, namely, a chick in the act of hatching.

In Aldrovandus' embryology there is much discussion of Aristotle and Galen, but traces of an independent spirit abound. Pliny's view that the heart was formed in the white is "exploded", and Aldrovandus says that it is formed on the yolk-membrane. He refutes the opinion of Galen also that the liver is first formed, in connection with which he says, "In order that I might bring to an end this controversy between the philosophers and the physicians I followed with the keenest curiosity and diligence the incubation of 22 hen's eggs, opening one each day; thus I found Aristotle's doctrine to be the truest. And because apart from the fact that these matters are most worthy of being looked into they provide also the greatest pleasure and entertainment I have thought it well to describe them as clearly and briefly as possible".

Aldrovandus also contradicts Albertus, and propounds a new theory, namely, that the spiritualia (the organs in the thorax) are formed from the seed of the cock {ex maris semine sunt). This seed he aflfirms to be present in the egg, and he identifies it with the chalazae, thus anticipating Fabricius ab Aquapendente, but not going quite so far, and explicitly opposing Gaza, who had said not long before that the chalazae were simply congealed water. Aldrovandus' admiration for Aristotle is extreme, and, though he differs from him about the chalazae, he defends the Aristotelian opinion that the chick was made from the white but nourished from the yolk. His argument for this is new, however; it is that, during incubation, the latter liquefies but the former hardens; now in all digestion liquefaction takes place, and in all growth hardening, therefore, etc. This argument is a great deal more cogent than most of those which were current between 1550 and 1650. He goes out of his way to castigate Albertus for saying that the yolk moves up into the sharp point of the egg, for experience assures him that it does not, "as I have observed by cutting open an egg after one day's incubation". A striking instance of his powers of observation was his description of the "egg-tooth" of embryonic birds, a discovery made anew in


the nineteenth century by Yarrell and Rose. The chick was perfect in form, according to him, on the tenth day.

The peculiarity of Aldrovandus lies in the fact that he incorporated so many elements into one book, and was able to produce a collection of chapters in which good scientific observation sat at the closest quarters with literary allusion and semi-theological homily. So wellproportioned a mixture as the Ornithologia is not often found. As a final instance three consecutive paragraphs may be mentioned, in the first of which he discusses Plutarch's arid problem about the priority of egg or hen, next he makes some very reasonable remarks about teratology, suggesting that monsters come from yolks which are physico-chemically abnormal in some way, while in the third he expresses strong scepticism concerning the tale that the basilisk is sometimes hatched out from a hen's egg — Ego ne jurantibus quidem crediderim'\ he says. This last notion is found in the fourteenthcentury poem of Prudentius alluded to above, and appears again in the Miscellaneous Exercitations of Caspar Bartholinus the younger, whose second chapter is devoted to showing "That the basilisk hatcheth not from the egg of the hen", a conclusion which has been amply confirmed in the light of subsequent experience. Bartholinus gives a bibliography of this curious legend.

Aldrovandus and his disciple Volcher Coiter the Frisian, as he described himself, were alike in not suffering from the prevailing vice of the age, verbosity. Colter's Externarum et Internarum principalium humajii corporis partium tabulae et exercitationes, which appeared at Nuremberg in 1573 — a beautifully printed book — contained a brief section entitled De ovorum gallinaceorum generationis primo exordio progressuque et pulli gallinacei creationis ordine. His Latin style betrays his German origin, for the constructions are very Teutonic, although the meaning is always perfectly clear. Coiter says, "In the year 1564 in the month of May at Bologna, being instigated by that excellent professor of philosophy outstanding in varied sciences and arts. Doctor Ulysses Aldrovandus, and by other doctors and students, I ordered 2 broody fowls to be brought and under each of them I caused 23 eggs to be placed, and in the company of these persons I opened one every day so that we could see firstly the origin of the veins and secondly what organ is first formed in the animal". What follows is practically a repetition of the facts available in Aristotle, but described with much greater clearness than either


Aristotle or Aldrovandus had been able to bring to the matter. On the third day, he saw the globulus sanguineus which in vitello manifeste pulsabat, and so solved his first problem. He decides that the first organ to be formed is the heart, and quotes Lactantius' experiments. He explains the large size of the eye as due to the fact that the most complicated part of the body needs the longest time for its manufacture. He correctly describes the various membranes, and the faeces subviridies in the intestines at hatching. Once he contradicts Aristotle, maintaining that on the tenth day the body as a whole is larger than the head, and once he contradicts Albertus, denying that any yolk can be found in the stomach at hatching. He concludes his tractate by a succinct and clear account of the opinions of Aristotle and Hippocrates about embryonic development. His importance is that he drew the attention of scientific thinkers to the problems arising out of the hen's egg, and assisted in the formation of that iconographic phase in embryology which was later to find its climax in the plates of Fabricius, and its close in Harvey's Exercitations.

Gabriel Fallopius, who belongs to this time, must be mentioned as the discoverer of the organs which bear his name, but his services to embryology were only indirect. A. Benedictus, who was now growing old, and Caesar Cremonius, who was still young, may be remembered as the principal upholders of pure Aristotelianism at this time. Realdus Columbus also wrote on the embryo. B. Telesius, in his De Natura Rerum of 1565, studied the hen's egg and suggested that the parts of animals were formed by the pressure of the uterus acting as a mould: he was thus the middle term between Galen and Buffon.

Julius Caesar Arantius has already been referred to. His De Humano Foetu was an important book, but, though it appeared in 1564, just at the time when the macro-iconographic school was at its height, it dealt with a rather different field and cannot be considered as a constituent of that group. He begins by relating that a pregnant woman was killed by an accident at Bologna a couple of years before, so that he had an opportunity of testing whether the opinions about certain points in generation, which he had formed on a priori grounds during the previous fifteen years, were true or not. In the first place, he found on dissection that the placenta was not cotyledonous, and he spoke thus of its formation: "Blood flows


out from the spongy substance of the uterus and this blood growing in bulk forms a soft and fungus-like mass of flesh, rather like the substance of the spleen, which adheres to the surface of the uterus and transmits to the foetus in proportion as it grows the nourishment for it which reaches the uterus in the form of blood and spirits". Then, going on to discuss the functions of the jecor uterinae, as he calls the placenta (with what justice may be seen by turning to Section 8-5), he devotes a chapter to De vasorum umbilicalium origine, and, contradicting Hippocrates, Galen, Erasistratus, and Aetius, says that the maternal and foetal blood-vessels do not pass into each other by a free passage. "This is repugnant to sense", he writes, "and as may be seen by ocular inspection, these vessels do not reach the inner membrane of the uterus, for the substance of the placenta is placed between their ramifications and the proper substance of the womb." He was thus the first to maintain that the maternal and foetal circulations are separate, but he naturally did not, and could not, speak of circulations, since he lived before Harvey. Nor could he have satisfactorily proved his point with the means then at his command, and, as we shall see, it was to take another century before the proof was given. Apart from this valuable contribution to embryology, Arantius gave some admirable anatomical descriptions of the foetal membranes.

Hieronymus Fabricius ab Aquapendente, the pupil of Fallopius, has always been given an important place in the history of embryology by those who have written on him. As one comes upon him in the process of tracing out that history itself, however, he does not take such a high place. With the statement, for instance, that "Fabricius carried embryology far beyond where Goiter had left it and elevated it at one bound into an independent science" I find that I cannot agree. Embryologists who called themselves that and nothing else did not appear till the end of the eighteenth century, and it seems to me doubtful whether the anatomical advances in embryology made by Fabricius are not counterbalanced by the erroneous theories which he invented at the same time. His De Formatione Ovi et Pulli pennatorum, and his De Formato Foetu of 1604 show far more scholasticism and mere argumentativeness than is to be found in Goiter, and are remarkable for their bulk. Fabricius seems to have had a genius for exsuccous and formal discussions. He spends much time, for example, in taking up the problem of whether



the yolk of the hen's egg is more earthy than the white, and looking at it from all possible angles. He disagrees at last with Aristotle and decides that the white is the more earthy. Bones, he says, are white, but also very earthy. The albumen is colder, stickier, and heavier than the yolk, "sequitur, terrestrius esse^\ And this particular example is the more flagrant because the actual matter of it is fundamentally physico-chemical. But, in addition, he introduced a number of grave errors and misleading theories into embryology, so that subsequently Harvey had to spend a large part of his time refuting them. Fabricius was, indeed, a good comparative anatomist, and it is upon that ground that he deserves praise: his plates, some of which are reproduced herewith, were far better than anything before and for a long time afterwards. He dissected embryos of man, rabbit, guinea-pig, mouse, dog, cat, sheep, pig, horse, ox, goat, deer, dogfish, and viper, a comparative study which had certainly never been made previously.

In his first tractate he begins by dealing with a question not unlike that of how the sardines got into the tin, i.e. how the contents got into the hard-shelled egg. He rejects Aristotle's idea that the egg is formed in the oviduct by a kind of umbilicus, and ascribes its growth there to transudation through the blood-vessels. He marks a definite advance upon Aristotle when he says that silkworms and other insects are born into their larval state from an egg, though he still terms the chrysalis an egg, and therefore holds that they are generated twice. Then follows his discussion of what part of the egg the chick comes from. The chalazae, he says, are not semen, for the semen is not present at all in the fertilised egg. His argument sounds peculiar when he says that both the white and yolk of the egg are the food of the embryo, for neither of them is absent at the end of incubation, therefore neither of them is its material. Hippocrates had said, "^ex luteo gigni, ex albo nutriri; Aristotle had said, "ex albo fieri, ex luteo nutrirV\ The latter was the view generally held in the sixteenth century, as may be gathered from Ambrosius Calepinus' dictionary, Scaliger's Commentary on Aristotle, and the treatise on the soul of Johannes Grammaticus.

Fabricius now says both nourish, neither makes. This distinction between food and building-materials seems to us unnecessary, but it had a great influence on later thought. Fabricius devotes much time to proving, as he thinks, that albumen and white are of the




same nature, and adduces the fact that "in cooking the white hardens first, whether the egg be boiled or poached, but the yolk hardens also if the heat is more", comparing the heat of the kitchen to the innate heat of the chick. "But you will say", he goes on, "if the albumen and the yolk are the food of the chick in the egg, what then must we decide the material of the chick to be, since we have already said that the semen is not present in the eggs. You will find this material from an enumeration of the parts of the egg — there remains only the shell, the two membranes, and the chalazae; — nobody will assign the membranes or the shell as the material of the chick, therefore the chalazae alone are the fitting substance out of which it can be made." Having discovered this truth by the infallible processes of logic, Fabricius brings all kinds of arguments forward to support it; he adduces the three nodes in the chalazae as the precursors of brain, heart, and liver; tadpoles, he thinks, resemble significantly the chalazae, being "armless legless spines". The eyes are transparent, so are the chalazae, therefore the latter must give rise to the former. The liver is formed as soon as the heart but is practically invisible as it does not palpitate. One of his most gratuitous errors was the suggestion, now newly introduced, that the heart (and other organs) of the foetus has no proper function, no munus publicum, but beats only in order to preserve its own life. Then there is a considerable section called De Ovorum utilitatibus, which almost does for the hen's egg what Galen's De Usu Partium did for the human body, and in which such questions as Why the shell is hard and porous? and Why there are any membranes in the egg? are taken up and answered with an elaborate display of common sense. The influence of Galen is perceptible in a passage about a liver-like substance being formed if blood is freshly shed into hot water, in the usual terminology of formative faculties, and in the division of fleshes into white and red, though the former is not specifically derived fi"om the semen nor the latter from the menstrual blood. The human placenta is described as cotyledonous, and needless confusion is caused by the doctrine that the "liquors, humours, or rather, excrements, around the foetus, are two in number, sweat and urine, the former in the amnios, the latter in the allantois". But the drawings and illustrations of Fabricius' work are beautiful and accurate — so much so, indeed, that it will always remain a mystery how the man who figured the early stages of the


development of the chick as Fabricius did, showing the bloodvessels radiating from the minute heart, should have been able to propound the thesis that the chalazae were the material of the embryo.

The other biologist to whom Harvey was most indebted was Andreas Laurentius of Montpellier, whose Historia Anatomica (printed with his other works in 1628) contained a whole book (viii) devoted to embryology, but which presents us with nothing except a commentary on Hippocrates and Aristotle. The only evidences of life are furnished by two polemics, one of which was against Simon Petreus of Paris, who had propounded some new views about the foetal circulation. Laurentius gave also a table showing the changes which occur in the heart and lungs of the foetus at birth.

It was about this time that the embryological observations of that many-sided genius, Hieronymus Cardanus, began to attract attention. His main thesis was that the limbs of the embryo were alone derived from the yolk, while the rest of the body came from the white. This was a well-meant attempt to mediate between the two traditions headed respectively by Aristotle and Hippocrates, but the arguments in support of it were not even remarkable for ingenuity. Constantinus Varolius treated of the formation of the embryo in a book which appeared in 1591, but very inadequately. He had certainly opened hen's eggs, and describes the fourth-day embryo as forma minimi faseoli. But nearly every one of his marginal headings begins with the word Cur, and this tells its own story, for the didactic style rarely hides genuine works of research. Johannes Fernelius, a rather earlier worker, in his De Hominis Procreatione followed Aristotle and Galen in nearly all particulars, and made no real contribution to embryology. On its practical obstetrical side, the sixteenth century produced some remarkable compilations of ancient gynaecological writings. The first of these was that of Caspar Wolf, which was published at Ziirich in 1566, and, after having been enlarged by Caspar Bauhin in 1586, subsequently formed the backbone of the most important and famous one, namely, that of Israel Spach (Strassburg, 1597). Although these composite textbooks represented no real embryological progress, they yet showed that great interest in development was alive, an interest which, though doubtless utilitarian in its origin, could hardly fail to lead to advances of a theoretical nature. (See Fig. 5.)

Fig. 5. Illustration from W. H. Ryff's Anatomia of 1541.


The obstetrical literature intended for midwives is also of great interest. It was about this time that the first popular guides to their subject began to appear, founded not upon mere superstition and the remnants of ancient knowledge derived in roundabout fashion through Syriac and Arabic, but either upon a careful study of Galen and Aristotle, or upon the results of dissections and living speculation. The principal representative of the former class is that of Jacob Rueff, which appeared in 1554 and was called De Conceptu et Generatione Hominis. Although written in Latin, it was evidently a popular work, for the illustrations given in it are such as would naturally be incorporated in such a book. It is the illustrations which give it its importance, and I reproduce them in Fig. 6. I think they show very clearly what the general ideas were at this period about mammalian embryology, and thus afford us a precious insight into what was in the minds of such writers as Riolanus the elder, Mercurialis, Saxonia, Rondeletius, Venusti, Holler and Vallesius. There are many points which their expositions of foetal growth and development leave vague, and without Rueff it would be difficult or impossible to picture in what manner they imagined it to go on. Rueff 's text follows Galen and Aristotle with fidelity, as does theirs — with the exception of a few minor ideas not quite consonant with this.

In (a) of Fig. 6 Rueff portrays the mixture of semen and menstrual blood in the womb, or, as he loosely refers to it, of both seeds, coagulating into a pink egg-shaped mass surrounded with a fine pellicle, {b) shows the same mass in the uterus and wrapped round with the three coats, amnion, chorion, and allantois — a lamentable but interesting misrepresentation of the facts. Then in {c) it is shown that upon the surface of the yolk-like mass of semen and blood appear "three tiny white points not unlike coagulated milk", these being the first origins of the liver, the heart, and the brain. Next {d) shows the first blood-vessels springing from the heart, four in number, and distributing themselves over the surface of the mass. It is plain that Rueff must either have opened hen's eggs himself and seen the early growth of the blastoderm or have been told about it by some observer such as Goiter or Aldrovandus. He could not have copied his pseudo-blastoderm pictures from their works, for in 1554 none of them had appeared, and, as far as I know, there were no similar illustrations in existence at that time.

After this point the pictures grow even more fanciful, and, in (^),

Fig. 6. Illustrations from Jacob Rueff's De Conceptu et Generatione Hominis of 1554 (arranged by Singer) showing the Aristotelian coagulum of blood and seed in the uterus.


the first outline of the cranium is seen taking shape in the upper part of the "egg". In (/) the blood-vessels have suddenly assumed the outline of a human being, and in (g) the finished product is seen. Rueff gives what seems to be a mnemonic in hexameters :

iniectum semen, sex primis certe diebus est quasi lac : reliquisque novem sit sanguis ; at inde consolidat duodena dies; bis nona deinceps effigiat; tempusque sequens producit ad ortum talis enim praedicto tempore figura consit.

Rueff gives some excellent diagrams of the foetus in utero with relation to the rest of the body, and the various positions which are familiar to obstetricians. His teratology is less happy, for he attributes the production of monsters to the direct action of God, though he does venture upon a few speculations concerning "corrupt seed". But his principal significance for this history is that, in his picture of the yolk-like mass of mixed semen and blood and the pseudoblastoderm upon it, he throws a good light on the conceptions of the time.

Rueff 's book was subsequently translated into English, and had many editions as The Expert Midwife.

The principal representative of the second class of popular books of this period is that of Euch. Rhodion, or Rosslein, which was translated into English, and published as his own work, by Thomas Raynold, " physition ", in 1 545, under the title of The Byrth ofMankynde otherwyse named The Woman" s Book (cf. d'Arcy Power). It was the first book in the English language to contain copper engravings. They were variants of the traditional Soranus-Moschion figures. The Rosslein-Raynold book pays less attention to Galenic theory than does that of Rueff, and includes much better drawings of actual dissections. Another famous obstetrical book was that of Scipio Mercurius; for further information here see Spencer.

The minor embryologists of the sixteenth century included among them Ambroise Pare, the founder of modern surgery. His teaching on generation involved nothing original, but it seems to have been Galenism interpreted by a very intelligent and well-balanced, unspeculative mind. The three-bubble theory appears in him very clearly; thus, we read, "The seed boileth and fermenteth in the womb, and swelleth into three bubbles or bladders" — the brain, the


liver, and the heart. Fare's illustrations are copied wholesale from Vesalius and Rueff, without acknowledgment. The last author to take the three-bubble theory quite seriously was A. Deusingius, who wrote in 1665, after Harvey. Others who deserve a mention, but no more, were Severinus Pinaeus, L. Bonaciolus and Felix Platter. None of them made any advance, and the illustrations of the former's De Virginitatibus notis graviditate et partu were almost ludicrous.

Hieronymus Capivaccius, F. Licetus, J. Costaeus and V. Cardelinus, who wrote in 1608, were the last true supporters of the ancient theories, such as that the male embryo was twice as hot and developed twice as quickly as the female.



3-1. The Opening Years of the Seventeenth Century

iEmilius Parisanus, a Venetian, now dealt with embryology in the fourth, fifth, and sixth books of his De Subtilitate. They were entitled as follows: "(4) Of the principles and first instruments of the soul and of innate heat, (5) Of the material of the embryo and of its efficient cause, (6) Of the part of the animal body which is first made, and of the mode and order of procreation". Parisanus is very wordy, but he has the merit of giving many quotations from the lesser known authors, and providing (as a rule) accurate references. He held that the spleen was formed in all development before the heart, and that neither heart nor lungs moved in utero. With regard to the controversy over the function of white and yolk, he was in agreement with Fabricius, but he firmly opposed the view that the chalazae were the first material of the chick, as much, it must be confessed, because of the opinion of Aristotle as from his own observation. Nevertheless, his own observations were noteworthy, and he will always be remembered for his discovery of the fact that the heart of the chick begins to beat some time before any red blood appears in it.

Parisanus was the last of the macro-iconographic group of sixteenthcentury embryologists. Their labours established the fundamental morphological facts about the developing embryo; the first great step in the history of embryology. But there were numerous errors in their work, and Harvey, who occupies a terminal or boundary position, was destined to correct them. He marks the transition from the static to the dynamic conception of embryology, from the study of the embryo as a changing succession of shapes, to the study of it as a causally governed organisation of an initial physical complexity, in a word, from Goiter and Fabricius to Descartes and Mayow. Iconography did not die : on the contrary, the improvement of the microscope gave it new life, and the micro-iconographic school emerged with its principal glory, Malpighi.


Harvey sums up the work of the macro-iconographic period in the historical introduction contained in Ex. xiv of his De Generatione Animalium. I give it in full in the beautiful seventeenth century English into which Harvey's Latin was translated under his guidance by the physician, Martin Llewellyn,

"We have already discovered the Formation, and Generation of the Egge; it remains that we now deliver our Observations, concerning the Procreation of the Chicken out of the Egge. An undertaking equally difficult, usefull, and pleasant as the former. For Nature's Rudiments and Attempts are involved in obscurity and deep night, and so perplext with subtilties, that they delude the most piercing wit, as well as the sharpest eye. Nor can we easier discover the secret recesses, and dark principles of Generation than the method of the fabrick and composure of the whole world. In this reciprocal interchange of Generation and Corruption consists the ^Eternity and Duration of mortal creatures. And as the Rising and Setting of the Sun, doth by continued revolutions complete and perfect Time; so doth the alternative vicissitude of Individuums, by a constant repetition of the same species, perpetuate the continuance of fading things.

"Those Authors which have delivered any thing touching this subject, do for the most part tread a several path, for having their Judgements prepossessed with their own private opinions, they proceed to erect and fashion principles proportionable to them.

"Aristotle of old, and Hieronymus Fabricius of late, have written so accurately concerning the Formation and Generation of the Foetus out of the Egge, that they seem to have left little to the industry of Posterity, And yet Ulysses Aldrovandus hath undertaken the description of the Pullulation or Formation of the chicken out of the Egge, out of his own Observations ; wherein he seems rather to have directed and guided his thoughts by the Authority of Aristotle, than by his own experience.

"For Volcherus Goiter, living at Bononia at the same time did by the advice of the said Aldrovandus (whom he calls Tutor) dayly employ himself in the opening of Egges sat upon by the Hen, and hath discovered many things truer than Aldrovandus himself, of which he also could not be ignorant. Likewise iEmilius Parisanus (a Venetian Doctor) despising other mens opinions hath fancied A new procreation of the Chicken out of the Egge.


"But because somethings, (according to our experience) and those of great moment and consequence, are much otherwise than hath been yet delivered, I shall declare to you what dayly progress is made in the egge, and what parts are altered, especially about the first dayes of Incubation; at which time all things are most intricate, confused, and hard to observe, and about which authors do chiefly stickle for their own observations, which they accomodate rather to their own preconceived perswasions (which they have entertained concerning the Material and Efficient Causes of the generation of Animals) than to truth herself.

"Aldrovandus, partaking of the same error with Aristotle, saith (which none but a blind man can subscribe to) that the Yolk doth in the first dayes, arise to the Acute Angle of the Egge ; and thinks the Grandines to be the Seed of the Cock; and that the Pullus is framed out of them, but nourished as well by the yolk as the white ; which is clean contrary to Aristotle's opinion, who conceived the Grandines to conduce nothing to the fecundity of the egge. Volcherus Goiter delivers truer things, and more consonant to Autopsie, yet his three Globuli are meer fables. Nor did he rightly consider the principle from whence the Foetus is derived in the Egg. Hieronymus Fabricius indeed contends, that the Grandines are not the seed of the cock, and yet he will have the body of the Chicken to be framed out of them (as out of its first matter) being made fruitful by the seed of the cock. He likewise saw the Original of the Chicken in the Egge; namely the Macula, or Cicatricula annexed to the membrane of the Yolke but conceived it to be onely a Relique of the stalk broken off, and an in-firmity of blemish onely of the Egge, and not a principle part of it. Parisanus hath plentifully confuted Fabricius his opinion concerning the Chalazae or Grandines, and yet himself is evidently at a loss in some certaine circles and points of the Principle parts of the Foetus (namely the Liver and the Heart) and seems to have observed a Principium or first Principle of the Foetus, but not to have known which it was, in that he saith, that the Punctum Album in the Middle of the Circles is the Cocks Seed out of which the Chicken is made. So that it comes to pass that while each of them desire to reduce the manner of the Formation of the Chicken out of the Egge to their own opinions they are all wide from the mark."

Before discussing how Harvey put them right, however, there are a number of other matters to be mentioned. Parisanus' work was


published in 1623, and twenty-five years were to elapse before Harvey's Exercitations were to be put before the learned world by George Ent. In that time not a few events of importance for the history of embryology took place.

It will be convenient to speak first of Adrianus Spigelius, whose De Formato Foetu appeared in 1 63 1 . In this book the plates of the gravid uterus which had been prepared some years before for Julius Casserius were now published. They had more influence than Spigelius' text, perhaps, in contributing to the permanent fame of his book.

He gives for the most part straightforward anatomical descriptions, but he returns to the notion of a cotyledonous placenta in man, and he combats Arantius' opinions about the placenta. Arantius had said that the function of the jecor uterinae was to purify the bloodsupply to the foetus, a thoroughly modern idea, but Spigelius opposes this on two grounds, firstly, because the foetus has its own organs for purifying blood, and secondly, because, if Arantius was right, the placenta would always be as red as blood, but this is not the case in such animals as the sheep. Spigelius himself thought that the placenta was for the purpose of preventing severe loss of blood at birth, as would be the case if the embryo was joined to the mother with only one big vessel and not a great many little ones.

However, Spigelius upholds the view, taken by Rufus of Ephesus and by Vesalius, that the allantois contains the foetal urine, which has to be separated from the amniotic liquid in which the embryo is, because it would corrode the embryonic skin [ne cuti tenellae aliquod damnum urinae acrimonia inferret). This passage is interesting, as showing biochemical rudiments. The first discussion of the vernix caseosa, or sordes, as he calls it, appears in Spigelius, who, however, hazards no guess as to its nature. He is happy in his refutation of Laurentius, who had affirmed that the foetal heart did not beat in utero, and he shows some advance on all previous writers save Arantius in declaring that the umbilical vessels take vital spirits away from the foetal heart, not exclusively to it. He gave, moreover, the first denial of the presence of a nerve in the umbilical cord, and also made the first observation of the occurrence of milk in foetal breasts at birth (for the endocrinological explanation of this see Section 15). Finally, he abolished at last the notion that the meconium in the foetal intestines argued eating in utero on the part of the embryo.


Riolanus the younger, the correspondent and almost exactly the contemporary of Harvey, was Professor in Paris and published his Anthropographia in 1618. As he was a keen advocate of the ancient views, his section on the formation of the foetus has little importance. Yet it contains the first known instance of the use of the lens in embryology, the germ of that powerful instrument which was to lead in due course to so many discoveries. "In aborted embryos", said Riolanus, "the structure is damaged and can often not be properly seen, even when you make use of lenses [conspicilid] which make objects so much bigger and more complicated than they ordinarily seem."

The De Formatrice Foetus of Thomas Fienus, Professor at Louvain and a friend of Gassendi, published in 1620, is interesting because it is the middle term between Aristotle and Driesch. As the titlepage informs us, he sets out to demonstrate that the rational soul is infused into the human embryo on the third day after conception. This by itself would not be very attractive, but the most cursory inspection shows that Fienus' interests were not at all theological. He divides the book up into seven main questions, (i) What is the efficient cause of embryogeny? He concludes that it is neither God, nor Intelligence, nor anima mundi (influence of Neo-platonism here as on Galileo). (2) Is it in the uterus or in the seed? In the latter, says Fienus, adding a list of authorities who agree with this view — Haly-Abbas, Gaietanus, Zonzinas, Turisanus, Fernelius, Vallesius, Peramatus, Saxonia, Carrerius, Zegarra, Mercurialis, Massaria, and Archangelus, ^' solus Fabio Pacio utero imprudenter adscribit (!). (3) Is it heat? Fienus nearly decided that it was, and, if he had done so, would have shown a modern mind, but no, he gave his opinion against it, saying, "the process (of development) is so divine and wonderful that it would be ridiculous to ascribe it to heat, a mere naked and simple quality". After weighing various other alternatives in questions (4), (5) and (6), he asks whether it is ^'^ anima seminis post conceptum adveniens (7), and concludes that it is. It is here that he becomes really interesting, for he quotes with approval certain writers, e.g. Alexander Aphrodisias [Organicum corpus esse organicum ab anima et anima praeexistere organizationi) , Themistius {Anima fabricatur architecturaque sibi domicilium et accommodatum instrumentum) and Marsilio Ficino in his commentary on Plato's Timaeus [Priusquam adultum sit corpus, anima tota in illius fabrica occupatur), and


then maintains with them that the soul is the principle which organises the body from within, arranging an organ for each of its faculties and preparing a residence for itself, not merely allowing itself to be breathed into a being which has already organised itself "The conformation of the foetus is a vital, not a natural, action", he says. He develops this idea in the remainder of the book; according to him, the seed first coagulates the menstrual blood into an amorphous cake, taking three days to do so, after which, the rational (not vegetative or sensitive) soul (entelechy), which has entered the uterus with the seed, finding a suitable mass of shapeless material, enters into it and begins to give it a shape. Fienus was attacked by several writers, and published a defence of his views.

Later writers on the same subject included Fidelis, Teichmeyer, Albertus, de Reies, Torreblanca and de Mendoza. The Spanish influence here is perhaps significant. Hieronymus Florentinus, who adopted the same standpoint as Fienus in 1658 was forced to recant it.

In 1625 Joseph de Aromatari, a Venetian, included in his epistle on plants the first definite statement of the preformationist theory since Seneca, but he did not develop the idea. He had noted that in bulbs and some seeds the rudiments of many parts of the adult plant can be seen even without glass or microscope, and this led him to suggest that probably in all animals as well as plants a similar thing was true. "And as for the eggs of fowls", he said, "I think the embryo is already roughly sketched out in the egg before being formed at all by the hen [quod attinet ad ova gallinarum, existimamus quidem pullum in ovo delineatum esse, antequam formatur a gallina]. This suggestion did not begin to bear its malignant fruits till the time of Swammerdam and Malpighi.

Johannes Sinibaldi's Geneanthropia might be mentioned as belonging to this time. It was a compilation of facts relating to the generation of man, but it expressly excluded from its field any discussion of the embryo. It is no more important for our subject than the queer Ovi Encomium of Erycius Puteanus, another of Gassendi's friends, which has already been referred to (p. 8).

3-2. Kenelm Digby and Nathaniel Highmore

Much more significant was the controversy between Sir Kenelm Digby and Nathaniel Highmore. In 1644, Sir Kenelm, whose in


triguing personality will be sufficiently familiar to anyone even slightly acquainted with seventeenth-century England, and whose biographic details may be found in John Aubrey, published a work with the following title: Two treatises, in the one of which, The Nature of Bodies, in the other, The Nature of Man's Soule is looked into, in way of discovery of the Immortality of Reasonable Soules. It was inscribed in a charming dedication to his son, and consisted, in brief, of a survey of the whole realms of metaphysics, physics, and biology from a very individual point of view.

One of Sir Kenelm's principal objects in writing was apparently to attack the old terminology of " qualities " in physics and "faculties " in biology. To say, as contemporary reasoning did, that bodies were red or blue because they possessed a quality of redness or blueness which caused them to appear red or blue to us, or again, to say that the heart beat because it was informed by a sphygmic faculty, or, to take the famous example, that opium sent people to sleep because it contained in it a dormitive virtue, appeared mere nonsense and word-spinning to Digby, "the last refuge of ignorant men, who not knowing what to say, and yet presuming to say something, do often fall upon such expressions".

Digby, like Galileo and Hobbes, wished to explain all phenomena by reference to two "virtues" only, those of rarity and density, "working by means oflocall Motion". Chapters twenty-three, twentyfour, and twenty-five contain his opinions and experiments in embryology. He begins by opening the question of epigenesis or preformation, practically for the first time since Albert the Great. "Our main question shall be", he says, "whether they be framed entirely at once, or successively, one part after another? And if this latter way, which part first?" He declares for epigenesis, but after a manner of his own, refuting "the opinion of those who hold that everything containeth formally all things". "Why should not the parts be made in generation", he asks, "of a matter like to that which maketh them in nutrition? If they be augmented by one kind of juyce that after severall changes turneth at the length into flesh and bone; and into every sort of mixed body or similar part whereof the sensitive creature is compounded, and that joyneth itself to what it findeth there already made, why should not the same juyce with the same progresse of heat and moisture, and other due temperaments, be converted at the first into flesh and bone though none be formerly there to joyn


it self unto?" He gives a clearly deterministic account of development. "Take a bean, or any other seed and put it in the earth, and let water fall upon it; can it then choose but that the bean must swell? The bean swelling, can it choose but break the skin? The skin broken, can it choose (by reason of the heat that is in it) but push out more matter, and do that action which we may call germinating? Can these germs choose but pierce the earth in small strings, as they are able to make their way? . . . Thus by drawing the thrid carefully along through your fingers, and staying at every knot to examine how it is tyed ; you see that this difficult progresse of the generation of living creatures is obvious enough to be comprehended and the steps of it set down; if one would but take the paines and afford the time that is necessary to note diligently all the circumstances in every change of it. . . . Now if all this orderly succession of mutations be necessarily made in a bean, by force of sundry circumstances and externall accidents ; why may it not be conceived that the like is also done in sensible creatures, but in a more perfect manner, they being perfecter substances? Surely the progresse we have set down is much more reasonable than to conceive that in the seed of the male there is already in act, the substance of flesh, bone, sinews, and veins, and the rest of those severall similar parts which are found in the body of an animall, and that they are but extended to their due magnitude by the humidity drawn from the mother, without receiving any substantiall mutation from what they were originally in the seed. Let us then confidently conclude, that all generation is made of a fitting, but remote, homogeneall compounded substance upon which outward Agents, working in the due course of Nature, do change it into another substance, quite different from the first, and do make it lesse homogeneall than the first was. And other circumstances and agents do change this second into a third, that third, into a fourth; and so onwards, by successive mutations that still make every new thing become lesse homogeneall than the former was, according to the nature of heat, mingling more and more different bodies together, untill that substance bee produced which we consider the period of all these mutations." This passage is indeed admirable, and well expresses the most modern conception of embryonic development, that of the ovum as a physico-chemical system, containing within itself only to a slight and varying degree any localisation answering to the localisation of the adult, and ready to change itself, once the



appropriate stimulus has been received, into the completed embryo by the actions and reactions of its own constituents on the one hand and the influence of the fitting factors of the environment upon the other. Digby has not received his due in the past; he stands to embryology as an exact science, much in the same relationship as Bacon to science as a whole.

"Generation is not made", he says, "by aggregation of like parts to presupposed like ones; nor by a specificall worker within; but by the compounding of a seminary matter with the juice which accrueth to it from without and with the steams of circumstant bodies, which by an ordinary course of nature are regularly imbibed in it by degrees and which at every degree doe change it into a different thing ..." (see p. 317). "Therefore to satisfie ourselves herein, it were well we made our remarks on some creatures that might be continually in our power to observe in them the course of nature every day and hour. Sir lohn Heydon, the Lieutenant of his Majesties Ordnance (that generous and knowing Gentleman, and consummate Souldier both in theory and practice) was the first that instructed me how to do this, by means of a furnace so made as to imitate the warmth of a sitting hen. In which you may lay severall eggs to hatch, and by breaking them at severall ages you may distinctly observe every hourly mutation in them if you please." Sir Kenelm then goes on to describe the events that take place in the incubating egg, which he does very accurately, though briefly. In vivipara, he says, the like experiments have been made, and the like conclusions come to by "that learned and exact searcher into nature. Doctor Harvey" — these he must have learnt of by word of mouth, for Harvey's book had not at that time been published. As regards heredity, he adopts a pure theory of pangenesis, and has more to say about it than any other writer of his time. He is sure that the heart is first formed both in ovipara and vivipara, "whose motion and manner of working evidently appears in the twinckling of the first red spot (which is the first change) in the egge".

Sir Kenelm Digby not only anticipated the outlook of the physicochemical embryologist, but he also foreshadowed with considerable accuracy Wilhelm Roux's definition of interim embryological laws. "Out of our short survey", he says, "of which (anserable to our weak talents, and slender experience) I perswade myselfe it appeareth evidently enough that to effect this worke of generation there needeth


not to be supposed a forming virtue or Vis Formatrix of an unknown power and operation, as those that consider things suddenly and in grosse do use to put. Yet in discourse, for conveniency and shortnesse of expression we shall not quite banish that terme from all commerce with us; so that what we mean by it be rightly understood, which is the complex assemblement, or chain of all the causes, that concur to produce this effect, as they are set on foot to this end by the great Architect and Moderatour of them, God Almighty, whose instrument Nature is : that is, the same thing, or rather the same things so ordered as we have declared, but expressed and comprized under another name." Thus Sir Kenelm admits that it is allowable to speak of the "complex assemblement" of causes, as if it were one formative virtue, and this corresponds to Roux's "secondary components" or interim embryological laws. But that the portmanteau generalisations can be resolved into ultimate physico-chemical processes, Digby both believes and spends two entire chapters in trying to show. Digby has been one of the two seventeenth-century Englishmen most underestimated in the history of biology, but his place is in reality a very high one. How far he was in advance of his time may be gauged from the work of his contemporary Sperlingen, whose book of 1641 was thoroughly scholastic and retrograde.

His Treatise on Bodies evoked several answers. Undoubtedly the most interesting from the progressive side was that of Nathaniel Highmore, who will always be well remembered in embryological history. Highmore's The History of Generation came out in 1651, so that Harvey must have known of it, and it is one of the puzzles of this period why Harvey did not make any mention of it in his work, especially as J. D. Horst in a letter to Harvey refers to Highmore as his pupil. Harvey replying in 1655 said he had not seen Highmore for seven years. Highmore's title-page expressly states that his book is an answer to the opinions of Sir Kenelm Digby. But before discussing in what the answer consisted, we may look at the plate which is bound in immediately after the dedication (to Robert Boyle). It is interesting in that it shows again the idea initiated by Leonardo, namely, that all growing things, plants as well as animals, have an umbilical cord, and in that the drawings of the chick embryos and eggs are more quaint than accurate (Plate VI).

Highmore first describes the Aristotelian doctrine of form and matter, and then censures both it and the extensions of it with their


"qualities", etc., much as Digby himself had done. "Some of our later philosophers have showed us that those forms w'^*^ they thought and taught to bee but potentially in the matter, are there actually subsisting though till they have acquired fitting organs, they manifest not themselves. And that the effects which were done before their manifestation (as the forming and fashioning of the parts wherein they are to operate) can rise from nothing else than from the Soul itselfe. This likewise I shall leave to the Readers enquiry, and shall follow that other way of introducing Forms, and Generation of creatures (as well animals as vegetables) which gives Fortune and Chance the preheminency in that work." He then describes Sir Kenelm's opinions, quoting from him in detail, and dissents from them mainly on the ground that they do not sufficiently account for embryogeny, as it were, from a technical point of view. That they subvert the "antique principals of philosophy" does not worry Highmore, but in his view their detailed mechanisms do not explain the facts, a much more serious drawback. Highmore is himself by way of being an Atomist, and it is because embryology was first treated by him from an atomistic standpoint that he derives his importance. "The blood, that all parts may be irrigated with its benigne moisture, is forc'd by several channels to run through every region and part of the body; by which meanes every part out of that stream selects those atomes which they finde to be cognate to themselves. Amongst which the Testicles abstract some spiritual atomes belonging to every part, which had they not here been anticipated, should have been attracted to those parts, to which properly they did belong for nourishment. . . . These particles passing through the body of the Testicles, and being in this Athanor cohobated and reposited into a tenacious matter, at last passe through infinite Meanders through certain vessels, in which it undergoes another digestion and pelicanizing." Highmore objects, therefore, more to Digby 's theory of pangenesis than to his description of embryogenesis. He goes on to give a long description of the development of the chick in the egg, mentioning in passing that the albumen corresponds to the semen and the blood of vivipara and the yolk to their milk. "Fabritius, who hath taken a great deal of pains in dissections. . . supposes the chick to be formed from the chalazae, that part which by our Women is called the treddle. But this likewise is false, for then every egge should produce 2 chickens, there being one treddle


l^: j:


9 pp ^^^E


■^ ^


0y: 6^^

0a: ^f

Tat-. 1^

(P^: 6.




at each end of the egg, which serve for no other end than for Hgaments to contain the yolk in an equilibrium, that it might not by every moving of the egg be shakt, broke, and confused with the white." Highmore was the first to draw attention to the increase of brittleness which takes place in the egg-shell during incubation, and he holds still to the Epicurean view that the female produces a kind of seed, though he thinks that the chick embryo is nourished in the early stages by the amniotic liquid.

Perhaps the most interesting reply to Digby from the traditional angle was that of Alexander Ross. In his Philosophicall Touchstone he upheld the Galenic view that the liver must be first formed in generation, for the nourishment is in the blood and the blood requires a liver to make it : ergo, the liver must be the earliest organ. Such arguments could dispense with observations. Ross also mentions Digby's suggestion that the "formative virtue" was only a bundle of natural causes, but he claims that the notion was an old one in schoolphilosophy, being included in the phrase causa causae^ causa causati.

3-3. Thomas Browne and the Beginnings of Chemical Embryology

There are references to embryology in Sir Thomas Browne's Pseudodoxia Epidemica, or Inquiries into very many vulgar Tenents and commonly received Truths, which was published at this time. The twenty-eighth chapter of the third book contains a number of difficult problems in the embryology of the period, in most cases stated without any solution. "That a chicken is formed out of the yolk of the Egg was the opinion of some Ancient Philosophers. Whether it be not the nutrient of the Pullet may also be considered ; since umbilical vessels are carried into it, since much of the yolk remaineth after the chicken is formed, since in a chicken newly hatched, the stomack is tincted yellow and the belly full of yelk which is drawn at the navel or vessels towards the vent, as may be discerned in chickens a day or two before exclusion. Whether the chicken be made out of the white, or that be not also its aliment, is likewise very questionable, since an umbilical vessel is derived unto it, since after the formation and perfect shape of the chicken, much of the white remaineth. Whether it be not made out of the grando, gallature, germ, or tred of the egg, as Aquapendente informeth us,


seemed to many of doubt; for at the blunter end it is not discovered after the chicken is formed, by this also the white and the yelk are continued whereby it may conveniently receive its nutriment from them both.. . .But these at last and how in the Cicatricula or little pale circle formation first beginneth, how the Grando or tredle, are but the poles and establishing particles of the tender membrans firmly conserving the floating parts in their proper places, with many observables, that ocular Philosopher and singular discloser of truth, Dr Harvey hath discovered, in that excellent discourse of generation, so strongly erected upon the two great pillars of truth, Experience, and Reason.

"That the sex is discernable from the figure of eggs, or that cocks or hens proceed from long or round ones, experiment will easily frustrate.. . .Why the hen hatcheth not the egg in her belly? Why the egg is thinner at one extream? Why there is some cavity or emptiness at the blunter end? Why we open them at that part? Why the greater end is first excluded [cf p. 233]? Why some eggs are all red, as the Kestrils, some only red at one end, as those of kites and buzzards? Why some eggs are not oval but round, as those of fishes ? etc. are problems whose decisions would too much enlarge this discourse." And elsewhere, "That (saith Aristotle) which is not watery and improlifical will not conglaciate; which perhaps must not be taken strictly, but in the germ and spirited particles; for Eggs, I observe, will freeze, in the albuginous part thereof". Again, "They who hold that the egg was before the bird, prevent this doubt in many other animals, which also extendeth unto them; for birds are nourished by umbilical vessels and the navel is manifest sometimes a day or two after exclusion.. . .The same is made out in the eggs of snakes, and is not improbable in the generation of Porwiggles or Tadpoles, and may also be true in some vermiparous exclusions, although (as we have observed in the daily progress of some) the whole Magot is little enough to make a fly without any part remaining. . . . The vitreous or glassie flegm of white of egg will thus extinguish a coal."

These citations show Sir Thomas to have been more than simply the supreme artist in English prose which is his common title to remembrance. In picking his way carefully among the doubtful points and difficult problems which previous embryologists had propounded but not answered, he usually managed to give the right


answer to each. But in addition to this, he was also an experimentahst, he had made both anatomical and physical experiments on eggs, and he was prepared to put any disputed point to the test of "ocular aspection", if this could be done. His experimental contributions to embryology come out more clearly in his Commonplace Books which were published by Wilkin in 1836.

"Runnet beat up with the whites of eggs seems to perform nothing, nor will it well incorporate, without so much heat as will harden the tgg. . . . Eggs seem to contain within themselves their own coagulum, evidenced upon incubation, which makes incrassation of parts before very fluid.. . .Rotten eggs will not be made hard by incubation or decoction, as being destitute of that spirit or having the same vitiated. . . . They will be made hard in oil but not so easily in vinegar which by the attenuating quality keeps them longer from concoction, for infused in vinegar they lose the shell and grow big and much heavier then before. ... In the ovary or second cell of the matrix the white comes upon the yolk, and in the later and lower part, the shell is made or manifested. Try if the same parts will give any coagulation unto milk. Whether will the ovary best?... The whites of eggs drenched in saltpeter will shoot forth a long and hairy saltpeter and the egg become of a hard substance. Even in the whole egg there seems a great nitrosity, for it is very cold and especially that which is without a shell (as some are laid by fat hens) or such as are found in the egg poke or lowest part of the matrix, if an hen be killed a day or two before she layeth. . . . Difference between the sperm of frogs and eggs, spawn though long boiled, would not grow thick and coagulate. In the eggs of skates or thornbacks the yolk coagulates upon long docoction, not the greatest part of the white. . . . In spawn of frogs the little black specks will concrete though not the other. ... In eggs we observe the white will totally freeze, the yolk, with the same degree of cold will grow thick and clammy like the gum of trees, but the sperm or tread hold its former body, the white growing stiff that is nearest to it."

The only conclusion that can be drawn from these remarkable observations is that it was in the " laboratory " in Sir Thomas' house at Norwich that the first experiments in chemical embryology were undertaken. His significance in this connection has so far been quite overlooked, and it is time to recognise that his originality and genius in this field shows itself to be hardly less remarkable than in so many


others. To have occupied himself with the chemical properties of those substances which afford the raw material of development was a great step for those times, but it was not until some twenty-five years later that Walter Needham carried this new interest into the mammalian domain, and made chemical experiments there.

3-4. William Harvey

The Latin edition of William Harvey's book on the generation of animals appeared in 1651, and the English in 1653. The frontispiece of the former which is reproduced as the frontispiece of this book is a very noteworthy picture, and derives a special interest from the fact that on the egg which Zeus holds in his hands is written, "^x ovo omnia'\ — a conception which Harvey is continually expounding (see especially the chapter, "That an egg is the common Original of all animals"), but which he never puts into epigrammatic form in his text, so that the saying, omne vivum ex ovo, often attributed to him, is only obliquely his.

The De Generatione Animalium was written at different times during his life, and not collected together for publication until George Ent, of the College of Physicians, persuaded Harvey to give it forth about 1650. As early as 1625 Harvey was studying the phenomena of embryology, as is shown among other evidences by a passage in his book where he says, "Our late Sovereign King Charles, so soon as he was become a man, was wont for Recreation and Health sake, to hunt almost every week, especially the Buck and Doe, no Prince in Europe having greater store, whether wandring at liberty in the Woods and Forrests or inclosed and kept up in Parkes and Chaces. In the three summer moneths the Buck and the Stagge being then fat and in season were his game, and the Doe and Hind in the Autumme and Winter so long as the three seasonable moneths continued. Hereupon I had a daily opportunity of dissecting them and of making inspection and observation of all their parts, which liberty I chiefly made use of in order to the genital parts". Nor was Harvey less diligent in examining the generation of ovipara. John Aubrey, in his Brief Lives, says, " I first sawe Doctor Harvey at Oxford in 1642 after Edgehill fight, but I was then too young to be acquainted with so great a Doctor. I remember that he came often to Trin. Coll. to one George Bathurst, B.D. who kept a hen in his chamber to


hatch egges, which they did dayly open to discerne the progress and way of generation". Aubrey mentions a conversation he had with a sow-gelder, a countryman of Httle learning, but much practical experience and wisdom, who told him that he had met Dr Harvey, who had conversed with him for two or three hours, and "if he had been", the man remarked, "as stiff as some of our starched and formall doctors, he had known no more than they". Harvey seems also to have learnt all he could from the keepers of King Charles' forests, as several passages in his book show. Nor was the King's own interest lacking. "I saw long since a foetus", he says, "the magnitude of a peasecod cut out of the uterus of a doe, which was complete in all its members & I showed this pretty spectacle to our late King and Queen. It did swim, trim and perfect, in such a kinde of white, most transparent and crystalline moysture (as if it had been treasured up in some most clear glassie receptacle) about the bignesse of a pigeon's Ggge, and was invested with its proper coat." And, again — "My Royal Master, whose Physitian I was, was himself much delighted in this kinde of curiosity, being many times pleased to be an eye-witness, and to assert my new inventions".

Harvey's book is composed of seventy-two exercitations, which may be divided up for convenience into five divisions. In Nos. i to 10 he speaks of the anatomy and physiology of the genital organs of the fowl, and the manner of production of eggs. Nos. 11 to 13 and also Nos. 23 and 36 deal with the hen's egg in detail, describing its parts and their uses, while in Nos. 14 to 23 the process of the "generation of the foetus out of the hen egge is described. The greater part of the book, comprising Nos. 25 to 62, as well as Nos. 71 and 72, is theoretical, and treats of the embryological theories held by Aristotle on the one hand, and the physicians, following Galen, on the other, instead of which it propounds new views upon the subject. Finally, Nos. 63 to 70, as well as the two appendices^ or "particular discourses", are concerned with embryogenesis in viviparous animals, especially in hinds and does.

It will be best to refer to certain details and main points of interest in Harvey's discussions, before trying to assess his principal contributions to the science as a whole. Harvey is the first, since Aristotle, to refer to the "white yolk" of birds. "For between the yolk", he says, "which is yet in the cluster and that which is in the midst of the eg when it is perfected this is the difference in chief, that though


the former be yellowish in colour and in appearance, yet its consistence representeth rather the white, and being sodden, thickeneth like it, growing compact and viscous and may be cut into slices. But the yolk of a perfect eggc being boiled groweth friable and of a more earthy consistence, not thick and glutinous like the white." All of Harvey's observations on the formation of the egg in the oviduct contained in this chapter are interesting, and may with advantage be compared with the studies of Riddle upon the same subject, where the chemical explanation will be found for many of Harvey's simple observations. Harvey's controversy with Fabricius on the question of whether the egg is produced with a hard shell or only acquires its external hardness upon standing in the air, which follows immediately on the above citation, is interesting. "Fabricius seemeth to me to be in errour, for though I was never so good at slight of hand to surprise an egge in the very laying, and so make discovery whether it was soft or hard, yet this I confidently pronounce that the shell is compounded within the womb of a substance there at hand for the purpose, and that it is framed in the same manner as the other parts of the egg are by the plastick faculty, and the rather, because I have seen an exceeding small egge which had a shell of its own and yet was contained within another egge, greater and fairer than it, which egge had a shell too."

Harvey was the first to note that the white of the hen's egg is heterogeneous, in the sense that part of it is much more liquid than the rest, and that the more viscous part seems to be contained in an exceedingly fine membrane, so that if it is sliced across with a knife, its contents will flow out. He also set right the errors of Fabricius, Parisanus and others, by showing that the chalazae were neither the seed of the cock nor the material out of which the embryo was formed, and, most important of all, by demonstrating that the cicatricula was the point of origin of the embryo. He denied, as against popular belief, that the hen contributed anything to the developing egg but heat, "For certain it is that the chicken is constituted by an internal principle in the egge, and that there is no accession to a complete and perfect egge by the Hennes incubation, but bare cherishing and protection; no more than the Hen contributeth to the chickens which are now hatched, which is only a friendly heat, and care, by which she defendeth them from the cold, and forreign injuries and helpeth them to their meat". Whether future work will still affirm


that nothing is given to the egg by the hen except heat is beginning now to be in doubt, if the results of Chattock are correct.

In the description of the development of the embryo in the hen's egg, which remains to this day one of the most accurate, Harvey says with regard to the spot on the yolk, which had, of course, been seen and mentioned by many previous observers, "And yet I conceive that no man hitherto hath acknowledged that this Cicatricula was to be found in every egge nor that it was the first Principle of the Egge". His description of the beginning of the heart, that "capering bloody point" or "punctum saliens'\ is too famous to need more than a reference. He thought that the amniotic liquid was of "mighty use", "For while the embryos swim there, they are guarded and skreened from all concussion, contusion, and other outward injuries, and are also nourished by it".

Thus he made no advance on the opinion which had for long been held, namely, that the amniotic liquid or colliquamentum served for sustenance. "I believe", he says, "that this colliquamentum or water wherein the foetus swims doth serve for his sustenance and that the thinner and purer part of it, being imbibed by the umbilicall vessels, does constitute and supply the primo-genital parts, and the rest, like Milk, being by suction conveyed into the stomack and there concocted or chylified, and afterwards attracted by the orifices of the Meseraick Veins doth nourish and enlarge the tender embryo." His arguments for this are, ( i ) that swallowing movements take place, and (2) that the gut of the chicken is "stuft" with excrement which could hardly arise from any other source. He was thus led to divide the amniotic liquid into two quite imaginary constituents, a purer and "sincerer" part, which could be absorbed straight into the blood without chylification, and a creamless milky part which could not be treated so simply.

"About the fourth day", says Harvey, "the egg beginneth to step from the life of a plant to that of an animall." "From that to the tenth it enjoys a sensitive and moving soul as Animals do, and after that, it is compleated by degrees and being adorned with Plumes, Bill, Clawes and other furniture, it hastens to get out." These and other passages which deal with the forerunner of the theory of recapitulation are interesting, but we have already met essentially the same idea in Aristotle. Harvey contributed nothing new to it. The first point on which he went definitely wrong was the statement that


he made that the heart does not pulsate before the appearance of the blood. No doubt his lack of microscopical facilities or of the desire to use them affords the reason for this error, but it was a very unfortunate one, for it was to a large extent upon it that he formulated his doctrine "the life is in the blood". For example, he says, "I am fully satisfied that the Blood hath a being before any other part of the body besides, and is the elder brother to all other parts of the foetus ".

The yolk, Harvey thought, supplied the place of milk, "and is that which is last consumed, for the remainder of it (after the chicken is hatched and walks abroad with the Henne) is yet contained in its belly". He thus ranged himself with Alcmaeon and Abderhalden. All his remarks about the relationships of yolk and white in nutrition are worth consideration; in noting, for instance, that the yolk is the last to be consumed, he comes very near to anticipating the knowledge of the succession of energy-sources which we now possess (see Section T"]). "In that Physitians affirme, that the Yolke is the hotter part of the ^gg&, and the most nourishing, I conceive that they understand it, in relation to us, as it is become our nourishment, not as it doth supply more congruous aliment to the chicken in the tggt. And this appeares out of our history of the Fabrick of the chicken ; which doth first prey upon and devoure the thinner part of the white, before the grosser; as it were a more proper diet, and did more easily submit to transmutation into the substance of the foetus. And therefore the yolke seems to be a remoter and more deferred entertainment than the white; for all the white is quite and clean spent, before any notable invasion is made upon the yolke." A comparison between these simple facts and our knowledge of embryonic nutrition is most interesting (see Section 6-9).

In connection with Minot's distinction of the periods of embryonic growth, it is curious that Harvey says, "And now the foetus moves and gently tumbles, and stretcheth out the neck though nothing of a brain be yet to be seen, but merely a bright water shut up in a small bladder. And now it is a perfect Magot, differing only from those kinde of wormes in this, that those when they have their freedom crawle up and down and search for their living abroad, but this worm constant to his station, and swimming in his own provision, draws it in by his Umbilicall Vessels".

Sometimes Harvey confesses himself puzzled by problems which could only be solved by chemical means, yet it does not occur to


him that this is the case. For instance, he enquires why heat will develop a chick out of a good egg but will only make a bad one worse. "Give me leave to add something here", he writes, "which I have tried often; that I might the better discerne the scituation of the foetus and the liquors at the seventeenth day to the very exclusion. I have boiled an eggc till it grew hard, and then pilling away the shell and freeing the scituation of the chicken, I found both the remaining parts of the white, and the two parts of the yolk of the same consistence, colour, tast, and other accidents, as any other stale egge, thus ordered, is. And upon this Experiment, I did much ponder whence it should come to passe that Improlifical eggs should, from the adventitious heat of a sitting Henne, putrifie and stink; and yet no such inconvenience befall the Prolifical. But both these liquors (though there be a Chicken in them too, and he with some pollution and excrement) should be found wholesome and incorrupt; for that if you eat them in the dark after they are boyled, you cannot distinguishe them from egges that are so prepared, which have never undergone the hen's incubation." Harvey was never afraid of trying such tests on himself; in another place, for example, he says, "Eggs after 2 or 3 days incubation, are even then sweeter relished than stale ones are, as if the cherishing warmth of the hen did refresh and restore them to their primitive excellence and integrity". "And the yolke (at 14 days) was as sweet and pleasant as that of a newlaid cgge, when it is in like manner boyled to an induration." Another matter on which Harvey set Fabricius right was on the question whether at hatching the hen helps the chicken out or the chicken comes out by itself. The latter was the belief held by Harvey, who said of Fabricius' arguments on this point that they were "pleasant and elegant, but not well bottomed".

On the great question of preformation v. epigenesis, Harvey keenly argued in favour of the latter view. "There is no part of the future foetus actually in the egg, but yet all the parts of it are in it potentially. ... I have declared that one thing is made out of another two several wayes and that as well in artificial as natural productions, but especially in the generation of animals. The first is, when one thing is made out of another thing that is pre-existent, and thus a Bedstead is made out of Timber, and a Statue out of a Rock, where the whole matter of the future fabrick was existent and in being, before it was reduced into its subsequent shape, or any tittle of the


designe begun. But the other way is when the matter is both made and receiveth its form at the same time. ... So Hkewise in the Generation of Animals, some are formed and transfigured out of matter already concocted and grown and all the parts are made and distinguished together per metamorphosin, by a metamorphosis, so that a complete animal is the result of that generation; but some again, having one part made before another, are afterwards nourished, augmented, and formed out of the same matter, that is, they have parts, whereof some are before, and some after, other, and at the same time, are both formed, and grow. . . . These we say are made per epigenesin, by a post-generation, or after-production, that is to say, by degrees, part after part, and this is more properly called a Generation, than the former. . . . The perfect animals, which have blood, are made by Epigenesis, or superaddition of parts, and do grow, and attain their just future or ciKfir} after they are born. . . . An animal produced by Epigenesis, attracts, prepares, concocts, and applies, the Matter at the same time, and is at the same time formed, and augmented.. . .Wherefore Fabricius did erroniously seek after the Matter of the chicken (as it were some distinct part of the egg which went to the imbodying of the chicken) as though the generation of the chicken were effected by a Metamorphosis, or transfiguration of some collected lump or mass, and that all the parts of the body, at least the Principall parts, were wrought off at a heat or (as himselfe speaks) did arise and were corporated out of the same Matter." Nothing could be more plain than Harvey's teaching on epigenesis, so that he has precedence over Caspar Wolff on this matter.

On the relation between growth and differentiation Harvey has some valuable things to say. The term "nutrition" he restricted to that which replaces existent structures, and the term "augmentation" or "increment" to that which contributes something new. That process which led to greater diversity of form and complexity of shape he called "formation" or "framing". "For though the head of the Chicken, and the rest of its Trunck or Corporature (being first of a similar constitution) do resemble a Mucus or soft glewey substance; out of which afterwards all the parts are framed in their order; yet by the same Operatour they are together made and augmented, and as the substance resembling glew doth grow, so are the parts distinguished. Namely they are generated, altered, and formed at once,


they are at once similar and dissimilar, and from a small similar is a great organ made." Harvey was thus very certain that the processes of growth in size and differentiation in shape went on quite concurrently, though he had no inkling of changes in the relative rapidity of each process. On this point he goes further than Fabricius. Fabricius thought that growth was a more or less mechanical process, taking its origin from the properties of elementary substances, but that differentiation was brought about by some more spiritual or subtle activity. "Fabricius", says Harvey, "affirmes amisse, that the Immutative Faculty doth operate by the qualities of the elements, namely. Heat, Gold, Moisture, and Dryness (as being its instruments) but the Formative works without them and after a more divine manner; as if (forsooth) she did finish her task with Meditation, Choice, and Providence. For had he looked deeper into the thing, he would have seen that the Formative as well as the Alterative Faculty makes use of Hot, Cold, Moist, and Dry, (as her instruments) and would have deprehended as much divinity and skill in Nutrition and Immutation as in the operations of the Formative Faculty her self." "I say the Concocting and Immutative, the Nutritive and Augmenting Faculties (which Fabricius would have to busie themselves only about Hot, Cold, Moist, and Dry, without all knowledge) do operate with as much artifice, and as much to a designed end, as the Formative faculty, which he affirms to possess the knowledge and fore-sight of the future action and use of every particular part and organ." Thus although in nearly every respect Harvey makes an advance on Fabricius, yet here he is retrograde, for, in the former's thought, the growth process at least had struggled towards a deterministic schema; with Harvey this movement is rigidly suppressed. "All things are full of deity" {Jovis omnia plena), said he, "so also in the little edifice of a chicken, and all its actions and operations, Digitus Dei, the Finger of God, or the God of Nature, doth reveal himself"

There can be no doubt that Harvey's leanings were vitalistic. In the following passage, he argues against both those who wished to deduce generation from properties of bodies (like Sir Kenelm Digby) and the Atomists ; in other words, against the outlook of those types of mind which in later times were to build up biophysics and biochemistry. Aubrey notes that Harvey was "disdainfull of the chymists and undervalued them".

"It is the usual error of philosophers of these times", says he, "to


seek the diversity of the causes of parts out of the diversity of the matter from whence they should be framed. So Physicians affirm, that the different parts of the body are fashioned and nourished by the different materials of blood or seed ; namely the softer parts, as the flesh, out of a thinner matter, and the more earthy parts as the bones, out of grosser and harder. But this error now too much received, we have confuted in another place. Nor are they lesse deceived who make all things out of Atomes, as Democritus, or out of the elements, as Empedocles. As if (forsooth) Generation were nothing in the world, but a meer separation, or Collection, or Order of things. I do not indeed deny that to the Production of one thing out of another, these forementioned things are requisite, but Generation her self is a thing quite distinct from them all. (I finde Aristotle in this opinion) and I my self intend to clear it anon, that out of the same White of the Egge (which all men confesse to be a similar body, and without diversity of parts) all and every the parts of the chicken whether they be Bones, Clawes, Feathers, Flesh, or what ever else, are procreated and fed. Besides, they that argue thus assigning only a material cause, deducing the causes of Natural things from an involuntary or casual concurrence of the Elements, or from the several disposition or contriving of Atomes ; they doe not reach that which is chiefly concerned in the operations of nature, and in the Generation and Nutrition of animals, namely the Divine Agent, and God of Nature, whose operations are guided with the highest Artifice, Providence, and Wisdome, and doe all tend to some certaine end, and are all produced, for some certaine good. But these men derogate from the Honour of the Divine Architect, who hath made the Shell of the Egge with as much skill for the egge's defence as any other particle, disposing the whole out of the same matter and by one and the same formative faculty." But although these are Harvey's theories, it is significant that in his preface he says, "Every inquisition is to be derived from its causes, and chiefly from the material and efficient", thus expressly excluding formal and final considerations. Certainly, as far as his practical work went, he was unaffected by them, and in the case of the egg-shell, for example, Harvey was not the man to say, "it is present for the protection of the embryo", and then to do or say nothing more. Such an explanation, though he might gladly accept it, was no bar to further exploration both by way of experiment and observation.


Harvey not only follows Aristotle in his good discoveries and true statements about the egg, but also, unfortunately, in his less useful parts, as, for example, when he devotes several pages to the discussion of how far the egg itself is alive, and whether there is any soul in subventaneous or unfruitful eggs. He decides that there is only a vegetative soul. On the other hand, he admirably refutes the opinion of those physicians — who were not few in number — who declared that the foetal organs were all functionless during foetal life. "But while they contende", he says, "that the mother's Blood is the nutriment of the foetus in the womb, especially of the Partes Sanguineae, the bloody parts (as they call them) and that the Foetus is at first, as if it were a part of the mother, sustained by her blood and quickened by her spirits, in so much that the heart beats not and the liver sanguifies not, nor any part of the Foetus doth execute any publick function, but all of them make Holy-Day and lie idle; in this Experience itself confutes them. For the chicken in the egge enjoyes his own Blood, which is bred of the liquors contained within the egge, and his Heart hath its motion from the very beginning, and he borroweth nothing, either blood or spirits, from the Hen, towards the constitution either of the sanguineous parts or plumes, as those that strictly observe it may plainly perceive." We have already seen how the Stoics in antiquity believed that the embryo was a part of the mother until it was born ; from this idea the transition would be easy to the belief that all the organs in the embryo were functionless and dependent on the activity of the corresponding ones in the maternal organism.

One of Harvey's most important services to thought lay in his abolishing for good the controversy which had gone on ever since the sixth century B.C. about which part of the egg was for nutrition and which for formation. He had the sense to see that the distinction was a useless and baseless one — "There is no distinct part (as we have often said) or disposed matter out of which the Foetus may be formed and fashioned. . . . An egge is that thing, whose liquors do serve both for the Matter and the Nourishment of the foetus.. . .Both liquors are the nourishment of the foetus."

As regards spontaneous generation, Harvey considered that even the most imperfect and lowest animals came out of eggs. "We shall show", he writes, "that many Animals themselves, especially insects do germinate and spring from seeds and principles not to be discerned


even by the eye, by reason of their contract invisible dimensions (like those Atomes, that fly in the aire) which are scattered and dispersed up and down by the winds ; all which are esteemed to be Spontaneous issues, or born of Putrefaction, because their seed is not anywhere seen." Unfortunately, he never returned to this subject, for, as he himself informs us in another place, all the papers and notes in his house in London were destroyed at the time of the Civil War, so that what he had written on the generation of insects irretrievably perished.

Another point on which Fabricius had been in error was the appearance of bone and cartilage in the embryo. According to him, "Nature first stretcheth out the Chine Bone, with the ribbes drawn round it, as the Keel, and congruous principle, whereon she foundeth and finisheth the whole pile". This armchair conceit Harvey was easily able to destroy by a mere appeal to experience, but by experience also he came upon a fact less easily to be explained, namely, that the motion of the foetus began when as yet there was hardly any nervous system. "Nor is it less new and unheard of, that there should be sense and motion in the foetus, before his brain is made; for the Foetus moves, contracts, and extends himself, when there is nothing yet appears for a braine, but clear water." On the basis of this paradox Harvey may be said to be the discoverer of myogenic contraction, but he already could claim that distinction, for the first heart-beats are accomplished long before there are any nerves to the heart, as he himself points out. "We may conclude from this fact", he remarks, "that the heart and not the brain is the first principle of embryonic life", and he gives instances of physiological actions not under the conscious control of the individual, such as the reflexes, as we should call them, of the intestinal tract, and the emetic action of infusion of antimony which cannot be tasted much and "yet there passeth a censure upon it by the Stomack" and a vomit ensues. Thus, twenty-five years before Francis Glisson, Harvey had formulated, from embryological studies, the view that irritability was an intrinsic property of living tissues.

Both Harvey and Fabricius were very puzzled about the first origin of the blood. "What artificer", says Harvey, "can transform the two liquors into blood, when there is yet no liver in being?" It was to be a long time before this question was answered by Wolflf 's discovery of the blood islands in the blastoderm, and, even now, the


chemistry of the appearance of haemoglobin is one of the most obscure corners of chemical embryology. The older observers explained it by considering the yolk to be akin to blood and ready to turn into it at the slightest inducement.

Another problem which neither Fabricius nor Harvey did anything to solve was the nature of the air-space at the blunt end of the egg. "Fabricius recounts several conveniences arising from it, according to its several magnitudes, which I shall declare in short, saying, It contains aire in it, and is therefore commodious to the Ventilation of the egge, to the Respiration, Transpiration, and Refrigeration, and, lastly, to the Vociferation of the Chicken. Whereupon, that cavity is at the first very little, afterwards greater, and at last greatest of all, according as the several recited uses do require."

As regards the placenta, Harvey took the side of Arantius and denied any connection between the maternal and foetal circulations. "The extremities of the umbilicall vessels", he said, "are no way conjoined to the extremities of the Uterine vessels by an Anastomosis, nor do extract blood from them, but are terminated in that white mucilaginous matter, and are quite obliterated in it, attracting nourishment from it." "Wherefore these caruncles may be justly stiled the Uterine Cakes or Dugs, that is to say, convenient and proportionate organs or instruments designed for the concocting of that Albuginous Aliment and for preparing it for the attraction of the veins." From this it would appear that Harvey regarded the uterine milk as the special secretion of the placenta, conveyed to the foetus through the umbilical cord. The nature of the uterine milk is still very imperfectly understood (see Section 21). Its discovery is usually attributed to Walter Needham, but various remarks in this chapter (Ex. lxx) seem to show that Harvey was well acquainted with it. In later times, it was regarded by some (Bohnius and Charleton in 1686, Zacchias in 1688 and Franc in 1722) as the sole source of foetal nourishment. Mercklin spoke of it in 1679 as materia albuginea, ovique albo non absimili". Harvey often calls the placenta the uterine liver, no doubt only for this reason, but the remarkable appropriateness of the term was to become apparent in Claude Bernard's day. As regards the matter of the continuity of the maternal and foetal circulations, he criticises van Spieghel. "There came forth a book of late", he says, "wrote by one Adrianus Spigelius, wherein he treateth concerning the use of the umbilicall arteries and doth


demonstrate by powerfull arguments that the Foetus doth not receive its Vital Spirits by the arteries from the Mother, and hath fully answered those arguments which are alledged to the contrary. But he might also as well have proved by the same arguments that the blood neither is transported into the Foetus from the mother's veines by the propagations of the umbilicall veins which is made chiefly manifest by the examples drawn from the Hen-Egge and the Caesarean Birth."

The least satisfactory parts of Harvey's book are the Exercitations Lxxi and lxxii on the innate heat and the primigenial moisture. Here he becomes very wordy and highly speculative, and gives us little but a mass of groundless arguments. He devotes many pages to proving that the innate heat is the blood and to drawing distinctions between blood and gore, the one in the body, the other shed. In one place he speaks of the processes of generation as so divine and admirable as to be "beyond the comprehension and grasp of our thoughts or understanding". Two centuries previously Frascatorius had said precisely the same thing about the motion of the heart, and it was ironical that the very man who let the light in on cardiac physiology should in his turn despair of the future of our knowledge of embryonic development.

Harvey did not say much about foetal respiration, and his few remarks are contained in one of the "additional discourses". He is puzzled exceedingly by the question. But he comes very near indeed to the truth when he says, "Whosoever doth carefully consider these things and look narrowly into the nature of aire, will (I suppose) easily grant, that the Aire is allowed to animals, neither for refrigeration, nor nutrition sake. For it is a tryed thing, that the Foetus is sooner suffocated after he hath enjoyed the Aire, than when he was quite excluded from it, as if the heat within him, were rather inflamed than quenched by the aire". Had Harvey pursued this line of thought, and looked still more narrowly into the nature of air, he might have anticipated Mayow. He does say that he proposes to treat of the subject again, but he never did.

The mainspring of Harvey's researches on the does and hinds can be realised by a reference to Rueff's figures in Fig. 6. According to the Aristotelian theory, the uterus after fertile copulation would be full of blood and semen ; according to the Epicurean theory (held by the "physitians") it would be full of the mixed semina. If this


coagulated mass exists, said Harvey, it ought to be possible to find it by dissection, and this was what he tried to do. It soon became plain, as may be read in Ex. lxviii, not only to Harvey but to the King and the King's gamekeepers, that no such coagulum existed,

^t^^^^^jr. S>'^'^^^ ^>^T*-^^J-^ r "^ ^^^ A--r> >^r'-/f >^'^^ S" ^,-^^ ^ , ■

x> f.^^ ^^^^^/s S^^ . rr- f

Fig. 7. Manuscript notes of Dr William Harvey.

and the result was made still more certain by means of segregation experiments which the King carried out at Hampton Court. Accordingly there was nothing to be done but to abandon all the older theories completely, and have recourse to some sort of hypothesis in which an aura seminalis, an "incorporeal agent" or a "kinde of contagious property" should bring about fertilisation. This was


a perfectly sound deduction from Harvey's experiments, and did not then appear anything like so unsatisfactory as it does now, for Gilbert of Colchester was not long dead, the "lodestone" was beginning to be investigated by the virtuosi, and even such extravagances as Sir Gilbert Talbot's Powder "for the sympatheticall cure of wounds" were only with difficulty distinguishable from the real effects of magnetic force. Harvey's idea of fertilisation by contagion has recently been in a sense revived by the work of Shearer (see Section 4*2).

But to Harvey himself the subject of the action of the seed was hid in deep night, and he confessed that, when he came to it, he was "at a stand". Some very interesting light is thrown upon his mind in this connection by a copy of the De Generatione Animalium annotated by himself, and now in the possession of Dr Pybus, by whose courtesy and by that of Dr Singer, who has transcribed the notes, I have been enabled to study it. It was given by Harvey to his brother Eliab, whose name it still bears. The notes, which are on the fly-leaves, are written in much the same way as those famous ones which Harvey used for his lectures at the College of Physicians in London, and which have been reproduced in facsimile. There is the same mixture of Latin and English, and the same signs, such as WI, to denote thoughts claimed as original. A page is reproduced in Fig. 7.

For the most part, the notes are uninteresting and nothing but a confusion of Aristotelian terms. But one page is concerned with the mode of action of the seed, and here we can, as it were, see Harvey's mind wrestling with this most difficult of problems. He sees that odour and the sense of smell may give a clue. That his thoughts on this point were doomed to frustration as soon as eggs and spermatozoa were discovered does not detract from the interest of the struggle.

Quod facit semen fecundum

What makes the seed fertile is on the analogy of an injection. In fact, the injection causes disease in many cases, and that from a distance, both by another. . .and by the same. . . A Venereal (?) disease corrupts coitus with a woman in whose uterus is the poison.

They do not [or do not yet ?] come forth in actuality but lie dormant as in fuel [? fomite]. Again, rabies in dogs lies dormant for many days on my own observation W4. Again, smallpox for days. Again, the generative seed, just as it (passes) from the male, lies dormant in the woman as infuel(?).

Or else like a . . . , like light in stone . . . , the pupil in the eye, in sense motion, ... in the body.


Like ferment, vapour, odour, rottenness ... by rule. Or like the smell given off by flowers.

Like heat, inflammation (?) A in chalk (heat ?) both the wet form

Like what is first ... in the art of cooking . . . principles of vegetation and propagation. A Dormice by hibernating. . .cleansing by water and all kinds of lotions, again for insects, as for their seeds as well (?). Or when a soul is a god present in nature, that is divine which it brings about without an organic body by means of law.

See Aristode Marvels concerning odours and smells given off. Whether on sense and everything that can be smelt gives off something and so the objects of disperses (?) what is not without heat, or by destroying . . . sense. attracts to itself

A Amongst inflammable (objects are) fire, naphtha, paper

A WI manus et odore car . . . anatomia manair. ...

A Anat . . . post 4°"^ poras. otium inclinente die rursus quod prius et

olefrere vid. . . . Galen. ... A Mr. Boys spainel in Paris lay all ye third night and morning in getting dogg. Whelping dogg's sent (scent) are a stronger sent, vesting in vestigio alios ord . . . gr . . . lepris odore lepris esse libidine esse. Hors, the mare, hors, the cow, a bull per mutta millsa. A ... si lepra fracedo in farioli fader cupidinitus. Dogg ye otter in aquas fracedo vasorum ex sulpore?

Just as Aristotle put much of his best embryological work into his Historia Animalium and not into the work with the appropriate title, so Harvey has some admirable observations on the embryonic heart scattered through his De Motu Cordis et Sanguinis in Animalibus. Turning now to consider Harvey's influence on embryology, we must admit that it was in certain respects reactionary.

1 . He did not break with Aristotelianism, as a few of his predecessors had already done, but on the contrary lent his authority to a moribund outlook which involved the laborious treatment of unprofitable questions.

2. His opposition to atomism and to "chymistry" precluded any close co-operation between his followers and those of the DescartesGassendi tradition.

3. Fabricius had elaborated a vitalistic theory of differentiation, but had allowed growth to be "natural" or mechanical. Harvey, however, made both growth and differentiation the results of an immanent spirit, a sort of divine legate.

But these failings are far outweighed by his positive services. It must always be remembered that he had no compound microscope.


and had to rely, like Riolanus, on "perspectives", or simple lenses of very low power.

1. There can be no doubt that the doctrine omne vivum ex ovo was a tremendous advance on all preceding thought. Harvey's scepticism about spontaneous generation antedated by less than a century the experiments of Redi. It is important to note that he was led to his idea of the mammalian ovum by observations on small conceptions surrounded by their chorion and no bigger than eggs, for the true ovum itself was not discovered until the time of de Graaf and Stensen.

2. He identified definitely and finally the cicatricula on the yolkmembrane as the spot from which the embryo originated.

3. He denied the possibility of generation from excrement and from mud, saying that even vermiparous animals had eggs.

4. He discussed the question of metamorphosis (preformation) and epigenesis, and decided plainly for the latter, at any rate for the sanguineous animals.

In addition to these achievements, there are others, perhaps less striking, but equally important.

5. He destroyed once and for all the Aristotelian (semen-blood) and Epicurean (semen-semen) theories of early embryogeny. This was perhaps the biggest crack he made in the Peripatetic teaching on development; but, in spite of it, Sennertus, van Linde and Sylvius adhered to the ancient views, and Cyprianus, in 1700, had the distinction of being the last to support them in a scientific discussion, though Sterne, as late as 1 759, referred to them in a way that shows they still lived on in popular thought.

6. He handled the question of growth and differentiation better than any before, anticipating the ideas of the present century.

7. He settled for good the controversy which had lasted for 2200 years as to which part of the egg was nutritive and which was formative, by demonstrating the unreality of the distinction.

8. He set his predecessors right on a very large number of detailed points, such as the nature of the placenta.

9. He made a great step forward in his theory of foetal respiration, though here he did not consolidate the gain.

10. He affirmed that embryonic organs were active, and that the embryo did not depend on external aid for its principal physiological functions.


But all these titles to remembrance, great as they are, do not account for the pecuhar fascination of Harvey. A little of it is perhaps due to his imaginative style, which comes out clearly in Martm Llewellyn's English version. A word of censure is due to Willis for transmuting it in his translation into the dull and pedestrian style of 1847. None who reads the 1653 edition of Harvey can ever forget such metaphors as this, "For the trunck of the body hitherto resembles a skiff without a deck, being in no way covered up by the anteriour parts"; or the vigour of diction which promotes such remarks as, "In a hen-egge after the tenth day, the heart admits no spectators without dissection"; or again, "For while the foetus is yet feeble. Nature hath provided it milder diet and solider meats for its stronger capacity, and when it is now hearty enough, and can away with courser cates, it is served with commons answerable to it. And hereupon I conceive that perfect eggs are not onely party-coloured, but also furnished with a double white"; or, lastly, "An egge is, as it were, an exposed womb ; wherein there is a substance concluded, as the Representative and Substitute or Vicar of the breasts".

In this connection, it would be a pity not to quote from the verses which Llewellyn prefixed to his translation of Harvey's book. After describing the controversies that followed the De Motu Cordis he wrote

A Calmer Welcome this choice Peice befall,

Which from fresh Extract hath deduced all,

And for Belief, bids it no longer begg

That Castor once and Pollux were an Egge :

That both the Hen and Houswife are so matcht,

That her Son born, is only her Son hatcht;

That when her Teeming hopes have prosp'rous bin.

Yet to conceive, is but to lay, within.

Experiment, and Truth both take thy part:

If thou canst 'scape the Women ! there's the Art.

Live Modern Wonder, and be read alone,

Thy Brain hath Issue, though thy Loins have none.

Let fraile Succession be the Vulgar Care;

Great Generation's Selfe is now thy Heire.

Curiously enough, the "calmer welcome" which Martin Llewellyn hoped for actually happened. Harvey's book was so well reasoned and based on such good observations that it produced only two


answers, and they were of little importance. Janus Orcham took exception to Harvey's finding no seed in the uterus and suggested that it had vaporised like a steam, but his Aristotelian leanings were promptly detected and castigated by Rallius. Matthew Slade, taking the pseudonym of Theodore Aides, published in 1667 his Dissertatio epistolica contra D. G. Harveium, which was, in his own words, "a detection of one or two errors in that golden book on the generation of animals of William Harvey, greatest of physicians and anatomists". The errors were purely anatomical, and ab Angelis defended Harvey against Slade's attack, claiming that the "errors" were not errors at all. A manuscript work of Slade's appears to be extant.

Harvey's influence was evidently speedily felt by his contemporaries. Strauss soon wrote a rather poor book on the bird's egg in imitation of him. But the best instance is that in 1655, very soon after the publication of Harvey's book, William Langly, "an eminent senator and physician of Dordrecht", made a great many experiments on the development of the hen's egg. Buffon says that he worked in 1635, i.e. before Harvey, but this is not the case, for in his observations which were published by Julius Schrader in 1674 the later date is given several times. Langly mentions Harvey more than once, and evidently followed his example in careful observation, for his text is concise and accurate and his drawings very noteworthy.

Julius Schrader included Langly's work in a composite volume containing a well-arranged epitome of Harvey's book on generation and some observations of his own on the hen's egg. The book was dedicated to Matthew Slade and J. Swammerdam. On the practical side Schrader added nothing memorable to Harvey and Langly, but it is noteworthy that the mammalian embryo was throughout these centuries more popular material than that of the chick. Out of fifty embryologists between Harvey and Haller, the names of Langly, Schrader, Malpighi and Maitre-Jan practically exhaust the list of those who studied the egg of the hen. This rather unfortunate orientation of mind doubtless sprang from the strong influence of medicine, and especially obstetrics, on seventeenth and eighteenth century embryology.

3-5. Gassendi and Descartes: Atomistic Embryology

Harvey's death took place in 1657. The following year saw the publication of Pierre Gassendi's Opera Omnia, and thus brought in


an entirely new phase in embryology. Together with Rene Descartes' treatise on the formation of the foetus, Gassendi's De generatione animalium et de animatione foetus marks a quite different attitude to the subject. Harvey had adopted a rather contemptuous position about the "corpuscularian or mechanical philosophy", which was then coming in, and had expected even less help from it in the solution of his problems than from his equally despised "chymists". Gassendi now set out to show that the formation of the foetus could be explained on an atomistic basis: and, using the Galenic physiology and the new anatomy as a framework, he set forth his theory in full. As we read it through at the present day, however, we cannot avoid the confession that it was not a success. In spite of his frequent quotations from Lucretius and his persuasive style, it does not carry conviction. The truth of the matter was that the time was not ripe for so great a simplification. The facts were insufficiently known, and that Gassendi is not quite as interested in them as he is in his theory is shown by the circumstance that he only mentions Harvey once.

Gassendi examines in turn the Aristotelian and the Epicurean doctrines of embryogeny and rejects them both, the former on the ground that the change from tgg to hen is too great and difficult for anything so shadowy and ghost-like as a "form" to accomplish, and the latter because it leaves no room for teleology. He therefore adopts as the basis of his system atomism + preformationism, alleging that all the germs of living things were made at the creation, but that they come to their perfection as atomic congregations in an atomistic universe. Thomas' monograph is a valuable help to the study of this very interesting thinker.

At exactly the same time, Descartes was speculating on the same subject. Added to his posthumous De Homine Liber (1662) is a treatise on the formation of the foetus. He may also have written a work On the generation of animals, for a manuscript with that title was found among his papers after his death, and was believed to be in his handwriting. There is evidence, however, that it is not his, and though it was published in Cousin's edition of his works, we may safely neglect it, agreeing, in the words of that editor, that it is "a fragment in which very mediocre and often quite false ideas struggle to light through the medium of a style devoid alike of clarity and of grandeur ". It must be admitted, however, that even his main treatise is very


confused. It suffers from containing in its earlier part a great deal of matter which really belongs to the physiological text-book which immediately preceded it. Thus it begins abruptly in the middle of a disquisition on the error of attributing bodily functions to the soul. Before long, however, it warms to its theme, and a conception of growth is outlined. "When one is young, the movement of the little threads which compose the body is less slow than it is in old age, because the threads are not so tightly joined one to the other, and the streams in which the solid particles run are large, so that the threads become attached to more matter at their roots than detaches itself from their extremities, so that they grow longer and thicker, in this way producing growth." The fourth part of the book is called, strangely enough, a Digression, in which the formation of the animal is spoken of. The mixture of seeds is then described, and a theory of the formation of the heart is attempted by means of an analogy with fermentation. The explanation is unconvincing, but has a certain interest as showing chemical notions beginning to permeate biological thought. However, Descartes' way of looking at development was thoroughly novel, as is illustrated by the following citation. "How the heart begins to move.. . .Then, because the little parts thus dilated, tend to continue their movement in a straight line, and because the heart now formed resists them, they move away from it and take their course towards the place where afterwards the base of the brain will be formed, they enter into the place of those that were there before, which for their part move in a circular manner to the heart and there, after waiting for a moment to assemble themselves, they dilate and follow the same road as the aforementioned ones, etc." Descartes, in fact, with premature simplification, was trying to erect an embryology more geometrico demonstrata. That he failed in the attempt was as obvious to his contemporaries as it is to us — "We see", said Garden, "how wretchedly Descartes came off" when he began to apply the laws of motion to the forming of an animal". In doing so, he was many years before his time; Borelli had done all that could be done at that period in that direction, and, significantly enough, he left embryology alone. The rest of Descartes' book is exactly like the citations which have been given, only applied to each organ and part in turn; he practically uses the traditional teaching as a scaffolding in which to interweave his mechanical theory, and he discovers no new facts.


But in the history of embryology these men and their writings have a very great significance. Impressed by the unity of the world of phenomena, they wished to derive embryology as well as physics from fundamental laws. This attempt, which resulted in a GalenEpicurus synthesis on the one hand and a Galen-Descartes synthesis on the other, must be regarded as a noble failure. Its authors did not realise what a vast array of facts would have to be discovered before a mechanical theory could with any justice be applied to explain them. Gassendi and Descartes were like the Ionian nature-philosophers, propounding general laws before the particular instances were accurately known. Their ineffectiveness arises from the fact that they did not themselves appreciate this, and consequently worked out their idea in a prolix detail, the whole of which was inevitably doomed to the scrap-heap from the very beginning. But the spark was not to die ; and if anywhere in this history we are to find the roots of physico-chemical embryology, we must pause to recognise them here.

Much less well known, but not without interest, was the Dissertatio de vita foetus in utero of Gregorius Nymmanus, which appeared in the same year as the second edition of Descartes' book, 1664. Nymmanus writes with a very beautiful Latin style, and expresses himself with great clearness. His proposition is, he says, "That the foetus in the uterus lives with a life of its own evincing its own vital actions, and if the mother dies, it not uncommonly survives for a certain period, so that it can sometimes be taken alive from the dead body of its mother". In supporting this thesis, Nymmanus answers the arguments of those who had held that the lungs and heart of the foetus were inactive in utero. Fabricius, Riolanus and Spigelius all proved, says Nymmanus, that the mother and the foetus by no means necessarily die at the same time. "The essential life", he says, "is the soul itself informing and activating the body, the accidental life is the acts of the soul which it performs in and with the body." Though the foetus cannot be said to have life in the latter sense, it can in the former. The foetus, says Nymmanus, prepares its own vital spirits and the instruments of its own soul; there is no nerve between it and its mother. If, he says, the foetal arteries got their sphygmic power from the maternal heart, they would stop pulsating when the umbilical cord was tied, but this is not the case. The pulse of the embryo is therefore due to the foetal heart itself. Galen, says Nymmanus, was


aware of this, but did not understand the meaning of it. Again, the foetus in utero moves during the mother's sleep, and vice versa. Nymmanus' dissertation is an interesting study in the transition from theological to scientific embryology which took place all through the seventeenth century, and may be followed in the writings of Varandaeus, de Castro, Dolaeus, Hildanus, Scultetus, Ammanus, Augerius and Garmannus. The problem of animation-time, a more metaphysical aspect of the same question, was still being handled, but less attention was being paid to it than formerly. Honoratus Faber's De Generatione Animalium of 1666 does not belong to its period. Its author, a Jesuit, proceeds in scholastic fashion to lay down four definitions, three axioms, one hypothesis, and seventy-seven propositions, in the last of which he summarises his conclusions. He is interesting in that he displayed a disbelief in spontaneous generation, thereby anticipating Redi, and he is careful to mention the work of Harvey, but nevertheless his treatise is of little value. His chief importance is that he is an epigenesist, and therefore demonstrates to us how the true opinion was becoming accepted, when Malpighi's brilliant observations and bad theory sent it out of favour, and prepared the way for the numerous controversies of the following century.

3-6. Walter Needham and Robert Boyle

It was in 1666 also that the following appeared in the Philosophical Transactions of the Royal Society :

A way of preserving birds taken out of the egge, and other small f actus' s: communicated by Mr. Boyle.

When I was sollicitous to observe the Processe of Nature in the Formation of the Chick, I did open Hens Eggs, some at such a day, and some at other daies after the beginning of the Incubation, and carefully taking out the Embryo's, embalmed each of them in a distinct Glass (which is to be carefully stopt) in Spirit of Wine; Which I did, that so I might have them in readinesse to make on them, at any time, the Observations, I thought them capable of affording; and to let my Friends at other seasons of the year, see, both the differing appearances of the chick at the third, fourth, seventh, fourteenth, or other daies, after the eggs had been sate on, and (especially) some particulars not obvious in chickens, that go about, as the hanging of the Gutts out of the Abdomen, etc. How long


the tender Embryo of the Chick soon after the Punctum saliens is discoverable, and whilst the bodie seems but a little organized Gelly, and some while after that, will be this way preserv'd, without being too much shrivel'd up, I was hindred by some mischances to satisfie myself; but when the Faetus's, I took out, were so perfectly formed as they were wont to be about the seventh day, and after, they so well retained thjeir shape and bulk, as to make me not repent of my curiosity; And some of those, which I did very early this Spring, I can yet shew you.

Boyle said in conclusion that he sometimes also "added Sal Armoniack, abounding in a salt not sowre but urinous".

In the same year that Nymmanus' book appeared, Nicholas Stensen, that great anatomist, later a Bishop, who was also to all intents and purposes the founder of geology, published his De musculis et glandulis specimen, in which Goiter's observations on the vitelline duct and the general relations between embryo and yolk in the hen's tgg were made again and confirmed. About this time also Deusingius described his case of abdominal pregnancy, and was thus the first anatomist to draw attention to this phenomenon.

In 1667 Stensen published his Elementorum myologiae specimen, in which he described the female genital organs of dogfishes. He demonstrated eggs in them and affirmed that the "testis" of women ought to be regarded as exactly the same organ as the "ovary" or "roe" of ovipara. At the time he carried the suggestion no further, but it was an extremely fruitful one, and it is surprising that it did not create more interest, for it was exactly what Harvey had been looking for. Nothing obvious having been found in the uteri of King Charles' does, and the conviction yet being very strong that viviparous conceptions really came from eggs, Stensen's minute ova supplied the fitting answer to the question. Thus Harvey and Stensen between them substituted the modern knowledge of mammalian ova for the ancient theory of the coagulum all in the space of fourteen years. The other event for which the year 1667 is remarkable is the De Formato Foetu of Walter Needham. Needham was a Cambridge physician who went to Oxford to study in the active school of physiological research which such men as Christopher Wren, Richard Lower, John Ward and Thomas Willis were making famous. His book on the formation of the embryo, written later (and dedicated to Robert Boyle), after he had been in practice in Shropshire for some time.


is important because it is the first book in which definite chemical experiments on the developing embryo are reported, and also because it contains the first practical instructions for dissections of embryos.

Sir Thomas Browne had, as we have already seen, made experiments of a chemical nature on the constituents of birds' eggs and of the eggs of amphibia, but he did not analyse them after any development had been allowed to take place. He may therefore be regarded as the father of the static aspect of physico-chemical embryology, while Walter Needham may be regarded as the founder of the dynamic aspect. The practical difficulties of these pioneers of animal chemistry may be seen in such a book of practical instructions as Salmon's General Practise ofChymistry of 1678. They had no satisfactory glassware, no pure reagents, the methods of heating were incredibly clumsy, and there was no means of measuring either heat or atmospheric pressure.

In the review of Needham's book which is to be found in the Philosophical Transactions of the Royal Society for September 1667 there occurs the sentence, "These humors (the amniotic, allantoic, etc.) he saith, he hath examined, by concreting, distilling, and coagulating them; where he furnishes the Reader with no vulgar observations". What were these observations ? They are to be found in the chapter entitled "The nature of the humours":

"I now proceed to speak of this other nutritive liquor round about the urine itself which latter is plainly separated by the kidneys and the bladder. These liquors also proceed from the blood and seem similar to its serum but yet they are different from it. For when fire is applied to them in an evaporating basin [cochlea] they do not coagulate, as the blood-serum always does. Indeed, not even the colliquamentous liquid of the egg itself coagulates in this manner, although it is formed from juices which are evidently liable to coagulation — in the same way humours differ from themselves before and after digestion, filtration, and the other operations [mangonial of nature. All, when distilled, give over a soft and clear water [mollem et lenem] very like distilled milk. This property is common to the liquor of the allantoic space, along with the rest. Because when the salts are not yet made wild and exalted the serum of the blood remains still quite soft and does not give proof of a tartaric or saline nature. Indeed, the first urine of an infant is observed by nurses to be not at


all salt, but in older animals, when I distilled it in an alembic, I seemed to observe a little volatile salt at the small end [in capitello]. Coagulations attempted by acids happened differently in respect of the different humours. For when I poured a decoction of alumina into the liquor of the cow's amnios it exhibited a few rather fine coagulations but they were clearly white. The allantoic juice, however, was precipitated like urine. Spirits of vitriol and vinegar brought about less results than alumina in each case. Spontaneous concretions I found also in the later months; these I discovered in both places. They are more frequent and larger, however, within the allantoic membrane."

From the above excerpt, which contains the account of all that Needham did on the chemical composition of the embryonic liquids, it can be seen that he treated the whole matter more dynamically than Browne. He was the first to describe the solid bodies in the amniotic fluid (see Jenkinson) and his chemical experimentation was all pioneer work.

His book has other merits, however. In the first chapter, he refutes the theory which Everard had propounded, that the uterine milk was identical with the contents of the thoracic duct, conveyed by lymphatic vessels to the uterus from the lac teals of Aselli, instead of elsewhere, and he shows that arteries must be the vessels bringing the material to the womb. The second chapter deals with the placenta "where he giveth a particular account of the double Placenta or Cake, to be found in Rabbets, Hares, Mice, Moles, etc., and examines the learned Dr Wharton's doctrine, assigning a double placenta to at least all the viviparous animals, so as one half of it belongs to the Uterus, the other to the Chorion, shewing how far this is true, and declaring the variety of these Phaenomena. Where do occur many uncommon observations concerning the difference of Milk [uterine] in ruminating and other animals, the various degrees of thickness of the uterin liquor in oviparous and viviparous creatures". He describes the human placenta very correctly indeed. "The use of the placenta is known to be to serve for conveighing the aliment to the foetus. The difficulty is only about the manner. Here are examined three opinions, of Curvey, Everhard, and Harvey. The two former do hold that the foetus is nourished only from the Amnion by the mouth ; yet with this difference, that Curvey will have it fed by the mouth when it is perfect, but whilst it is yet imperfect, by filtration


through all the pores of the body, and by a kind of juxtaposition : but Everhard, supposing a simultaneous formation of all the instruments of nutrition together and at first, and esteeming the mass of bloud by reason of its asperity and eagerness unfit for nutrition, and rather apt to prey upon than feed the parts, maintains, that the liquor is sucked out of the amnion by the mouth, concocted in the stomack, and thence passed into the Milky Vessels even from the beginning. Meantime they both agree in this, that the embryo doth breath but not feed through the umbilicall vessels. This our Author undertakes to disprove; and having asserted the mildness of, at least, many parts of the bloud, and consequently their fitness for nutrition, he defends the Harveyan doctrine of the colliquation of the nourishing juyce by the Arteries and its conveyance to the foetus by the veins."

In the third chapter Needham gives the first really comparative account of the secondary apparatus of generation, enunciating the rather obvious rule that in any given case the number of membranes exceeds the number of separate humours by one. He affirms that all the humours are nutritive save the allantoic. It had previously been held that all fish eggs were of one humour only, but he points out that a selachian egg has its white and yolk separate. He gives the results of his chemical experiments at this point, and suggests that the noises heard from embryos in utero and in ovo may be due to the presence of air or gas in the amniotic cavity, thus forming a link between Leonardo and Mazin. In his fourth chapter he deals with the umbilical vessels and the urachus, and here he claims priority over Stensen for the discovery of the ductus intestinalis in the chick, referring to Robert Boyle, Robert Willis, Richard Lower and Thomas Millington, to whom, he says, he showed the duct before Stensen published his observations on it. The fifth chapter is concerned with the foramen ovale, and the arterial and venous canals, and with the foetal circulation in general. The sixth is about respiration or "biolychnium", and in it Needham writes against the conception of a vital flame, alleging cold-blooded animals, etc., in his favour, but here he takes a retrograde step, for he argues that the use of the lungs is not for respiration but to "comminute the bloud and so render it fit for a due circulation". "The seventh and last chapter contains a direction for the younger Anatomists, of what is to be observed in the dissection of divers animals with young, and first, of what is


common to the viviparous, then, what is pecuHar to severall of them, as, a sow, mare, cow, ewe, she-goat, doe, rabbet, bitch, and a woman, lastly, what is observable in an Egg, skate, salmon, frog, etc. All is illustrated with divers accurate schemes."

The subsequent course of chemical embryology in the seventeenth century may be put in a very few words. Marguerite du Tertre incorporated in her obstetrical text-book of 1677 the results of some similar experiments to those of Needham. "If you heat the (amniotic) liquor", she says, "it does not coagulate, and if you boil it it flies away leaving a crass salt like urine, but if you heat the serosity of blood, it solidifies as if it were glue." The same observation was recorded by Mauriceau in 1687, who concluded, with some common sense, that, as there was so little solid matter present, the liquid could not be very nutritive; and by Case in 1696, who said, "In this juice the plastic and vivifying force resides, for although to our eyes it looks in colour and consistency like the serum of the blood, yet it is absolutely \toto coelo] different; for if a little of the former is slowly evaporated \si in cochleari super ignem defines] no coagulation will ever appear." Lister said this once more in 171 1, but with Boerhaave's work of 1732 the subject entered a new phase.

In 1670 Theodore Kerckring published an adequate work on foetal osteology, and, two years later, de Graaf and Swammerdam, making full use of the opportunities afforded them by the invention of the microscope, described in detail the ova of mammalia, thus demonstrating the truth of Stensen's suggestion of some years before. It is important to note that these workers mistook the "Graafian follicles" for the eggs — a mistake which was not rectified till the time of von Baer. Stensen himself published not long after an account of these eggs also, but he was by then too late to gain the priority of demonstration. Portal's claim that Ferrari da Grado, who lived in the fifteenth century, was the true discoverer of mammalian ova has been disproved by Ferrari; and, although it is true that Volcher Goiter described what we now call the Graafian follicles, he did not recognise in any way their true nature.

De Graaf 's discovery was confirmed in 1678 by Caspar Bartholinus, and, in 1674, by Langly, whose original observations had been made, so it was said, in 1657, the year of Harvey's death. If this is true, Langly has the priority of observation, Stensen of theory and de Graaf of demonstration.


3-7. Marcello Malpighi: Micro- Iconography and Preformationism

In the year 1672, Marcello Malpighi, who had for many years previously been working on various embryological problems with the aid of the simple microscope, published his tractates De Ovo Incubato and De Formatione Pulli in Ovo. In spite of its great importance, there is not much to be said about it, for it is anything but a voluminous work. The plates in which Malpighi represented the appearances he had seen in his examination of the embryo at different stages are beautiful, and some of them are reproduced. Description of the embryo was now pushed back into the very first hours of incubation, and it is interesting to note that Malpighi could not have done his work without Harvey, whose name he mentions on his first page, and who pointed out the cicatricula as the place where development began, and therefore, as Malpighi must have reasoned, the place where microscopic study would be very profitable. Now for the first time the neural groove was described, the optic vesicles, the somites, and the earliest blood-vessels.

Malpighi opened the modern phase of the controversy preformation versus epigenesis by supporting the former view. Embryogeny, he held, is not comparable to the building of an artificial machine, in which one part is made after another part, and all the parts gradually "assembled", but takes place rather by an unfolding of what was already there, like a Japanese paper flower in water. He was led to this belief by the fact that development goes on after fertilisation as the tgg passes down the oviduct, and in the most recently laid eggs gastrulation is already over, so that in his researches he could never find an absolutely undivided egg-cell. It is curious to note that he says his experiments were done "mense Augusti, magno vigente calore'\ so that more than a usual degree of development would have taken place overnight. Had he examined the cicatriculae in hens' eggs before laying, he would very probably not have formed this theory, and the epigenesis controversy would have been settled with Harvey. Another influence which was unfavourable to the epigenetic position was that it was Aristotelian, and therefore unfashionable. Yet Malpighi's view was much more sensible than many which succeeded it, for he did not maintain a perfectly equal swelling up of all parts existing at the start, but rather an unequal unfolding,

SECT. 3]



a distribution of rate of growth at different times and in different regions of the body. Thus he says, "Now, as Tully says, Death truly belongs neither to the living nor to the dead, and I think that something similar holds of the first beginnings of animals, for when we enquire carefully into the production of animals out of their eggs, we always find the animal there, so that our labour is repaid and we see an emerging manifestation of parts successively, but never the first origin of any of them".

Ti£ JCr

I>e Cue .


  1. 1? J


Ti^ lOI.

Fig. 8. Malpighi's drawings of the early stages of development in the chick embryo.

What had been an unfounded speculation for Seneca in antiquity and for Joseph de Aromatari and Everard in late times was now set upon an apparently firm experimental basis by Malpighi.

It is most instructive to note the difference in the attitudes of Langly and Schrader respectively towards the preformation question. Langly has no doubts about it, nor has Faber; they both follow Harvey and epigenesis unquestioningly, but Schrader, although he believes in epigenesis on the whole, is not at all certain about it. His friend, Matthew Slade, he says, brought the epistle of Joseph de Aromatari to his attention, and what with that and the unexplained observations of Malpighi on the pre-existence of the embryo, he is not willing to deny all value to preformationist doctrine. Others were bolder. It was immediately seized upon by Malebranche, the

1 68


Streeter of his age, who, in his Recherche de la Verite of 1672, reaHsed its philosophical possibilities, and gave it a kind of metaphysical sanction. That mystical microscopist, Swammerdam, made use of it as an explanation of the doctrine of original sin. In a remarkably short space of time it was a thoroughly established piece of biological theory.

Malebranche refers to it in his Recherche de la Verite in the chapter where he treats of optical illusions and emphasises the deceitfulness and inadequacy of our senses. "We see", he says, "in the germ of a fresh Qgg which has not been incubated an entirely formed chicken. We see frogs in frogs' eggs and we shall see other animals in their


Fig. 9. Malpighi's drawings of the chick embryo's blood-vessels.

germs also when we have sufficient skill and experience to discover them. We must suppose that all the bodies of men and animals which will be born until the consummation of time will have been direct products of the original creation, in other words, that the first females were created with all the subsequent individuals of their own species within them. We might push this thought further and belike with much reason and truth, but we not unreasonably fear a too premature penetration into the works of God. Our thoughts are, indeed, too gross and feeble to understand even the smallest of his creatures." Malebranche, who was a priest of the Oratory of the Cardinal de Berulle, took an ardent interest in the scientific life of his time — for example, in a letter to Poisson, the Abbe Daniel wrote, "Reverend Father, M. Malebranche has written to me saying that he has installed an oven in which he has hatched eggs. He has already opened



lu Ouc


ILLUSTRATIONS FROM MALPIGHI'S DE OVO INCUBATO OF 1672 Showing the early stages of the development of the chick, somites, area vasculosa, etc.


some and has been able to see the heart formed in them and beating, together with some of the arteries" (Blampignon).

Swammerdam's support for preformation came from a different angle. He had been investigating insect metamorphosis, and, having hardened the chrysalis with alcohol, had seen the butterfly folded up and perfectly formed within the cocoon. He concluded that the butterfly had been hidden or masked {larvatus) in the caterpillar, and thence it was no great step to regard the Qgg in a similar light. Each butterfly in each cocoon must contain eggs within it which in their turn must contain butterflies which in their turn must contain eggs, and so on. Before long, Swammerdam extended this theory to man. "In nature", he said, "there is no generation but only propagation, the growth of parts. Thus original sin is explained, for all men were contained in the organs of Adam and of Eve. When their stock of eggs is finished, the human race will cease to be."

In 1684 Zypaeus reported that he had seen minute embryos in unfertilised eggs, and there were other similar claims. ^Hinc recentiores physiologV\ said Schurigius in 1732, ^^ hominem in ovulis delineatum quoad omnes partes in exiguis staminibus ante conceptionem existere statuunt"

Swammerdam cannot be regarded simply as one of the principal pillars of the preformation theory. His own embryological researches, which were made chiefly on the frog, were remarkable in many ways. He was the first to see and describe the cleavage of the eggcell and later segmentation. He said that there was a time during the development of the tadpole when its body consisted of granules {greynkens or klootkens), but as these grew smaller and much more numerous they escaped his penetration. Leeuwenhoek also saw these cells, and his account was published long before Swammerdam's, but his observations on the rotating embryos oi Anodon and the eggs of fleas were equally interesting.

3-8. Robert Boyle and John Mayow

In 1674 John Mayow, a young Oxford physician, published his tractate, De Respiratione Foetus in Utero et Ovo, which was included as one of the parts of his Tractatus Quinque medico-physici in that year. Mayow was the first worker to realise that gaseous oxygen, or, as he termed it, the " nitro-aerial " vapour, was the essential factor in the burning of a candle and the respiration of a living animal. His work


was forgotten until Beddoes drew attention to it in 1 790, but since then many have praised it and Schultze makes him the equal of Harvey.

The reason why he became interested in embryology is given in the opening sentences of his work. "Since the necessity of breathing", he says, "is so essential to the sustaining of life that to be deprived of air is the same as to be deprived of common light and vital spirit, it will not be out of place to enquire here how it happens that the foetus can live though imprisoned in the straits of the womb and completely destitute of air." He first of all gives an account of the opinions held about foetal respiration and the umbilical cord. He says that he disagrees (i) with the view that the embryo breathes per OS while it is in the womb, for there is no air in the amnion and the suctio infantuli proves nothing; and (2) with the view propounded by Spigelius that the umbilical vessels existed to supply blood to the placenta for the nourishment of the latter. If this were the case, he says, the membranes in the hen's egg could not be formed before the vitelline vein, as they are, and in cases of foetal atrophy the placenta would always die and be corrupted too, which does not happen. Nor does he support the view of Harvey (3) that the umbilical vessels supply blood for the concoction and colliquation of the food of the foetus, for why should not the embryonic body prepare its own nutritious juice before birth just as it does afterwards. He further thinks the theory (4) that the umbilical vessels are for carrying off surplus foetal nourishment quite untenable and as little likely as the theory (5) that they exist for the object of allowing a foetal circulation — for this could just as well be accomplished through the vessels which exist in the embryonic body.

Mayow decides therefore for the opinion of divino sene Hippocrate and Everard that the umbilicus is a respiratory mechanism, carefully dissociating himself, however, from the hypothesis of Riolanus that the umbilical cord with all its windings is so arranged to cool the blood passing through it. He then says, "We observe, in the first place, that it is probable that the albuminous juice exuding from the impregnated uterus is stored with no small abundance of aerial substance, as may be observed from its white colour and frothy character [Needham's uterine milk]. And in further indication of this, the primogenial juices of the egg, which have a great resemblance to the seminal juice of the uterus, appear to abound in air particles. For if the white or the yolk of an tgg be put into a glass from which


the air is exhausted by the Boyhan pump these liquids will immediately become very frothy and swell up into an almost infinite number of little bubbles and into a much greater bulk than before — a sufficiently clear proof that certain aerial particles are most intimately mixed with these liquids. To which I add that the humours of an tgg when thrown into the fire, give out a succession of explosive cracks which seem to be caused by the air particles rarefying and violently bursting through the barriers which confined them. Hence it is that the fluids of the egg are possessed of so fermentative a nature. For it is indeed probable that the spermatic portions of the uterus and its carunculae are naturally adapted for separating aerial particles from arterial blood. These observations premised, we maintain that the blood of the embryo, conveyed by the umbilical arteries to the placenta or uterine carunculae, brings not only nutritious juice, but along with this a portion of nitro-aerial particles to the foetus for its support, so that it seems that the blood of the infant is impregnated with nitro-aerial particles by its circulation through the umbilical vessels quite in the same way as in the pulmonary vessels. And therefore I think that the placenta should no longer be called a uterine liver but rather a uterine lung". These splendid words, informed by so much insight and scientific acumen, show that, by the time of Mayow, chemical embryology had definitely come into being. He died at the early age of thirty-six, and we may well ponder how different the subsequent course of this kind of study would have been if he had lived a little longer.

The second part of Mayow's treatise is concerned with respiration in the hen's egg during its development, and it may be noted that his observations on the air contained in the liquids before development probably account for the facts which have been reported at one time and another concerning an alleged anaerobic life of embryos in early stages. Mayow is wrong in supposing that the gas which he pumped out from white and yolk was purely "nitro-aerial", but he shows the greatest good sense in his reminder that the amount of nitro-aerial particles required by embryos must be comparatively small owing to their small requirement for "muscular contraction and visceral concoction". His remarks on the effect of heat on the developing egg are not so clear as the remainder of the treatise, but he seems to mean that the heat will disengage the nitro-aerial particles from the liquids, and so aid in respiration, an idea which was later


used by Mazin. His fundamental mistake here was that he failed to realise that the egg-shell was permeable to air; and this vitiates all his reasoning about the respiration of the egg. "It will not be irrelevant", he says, "to enquire here whether the air which is contained in the cavity in the blunter end of every egg contributes to the respiration of the chick." He first notes that the cavity in question lies between two membranes and not between the shell-membrane and the shell as Harvey himself had supposed ; and then he goes on to say that he disagrees with the opinion of Fabricius, who had asserted that the air in the air-space serves for the respiration of the chick. His reasons are (i) that there would not be enough therein for the needs of the embryo which would use it, as it were, in one gulp, and (2) that the air in it cannot pass through the inner membrane, an error into which he was led by observing that, if an egg-shell with its contents removed and its air-space intact, was put into a vacuum, the air-space would swell up until it was as big as the egg itself. Mayow sees now what had escaped the attention of all previous observers, namely, that the egg-contents are not "rarefied or expanded, but are on the contrary condensed and forced into a narrower space than before". Such a condensation could, he thinks, take place in four ways, (a) by an increase in propinquity of discrete particles, (b) by a subsidence of motion on the part of a congregation of particles into rest, (c) by the extraction of some subtle spirit from amongst the particles, and, (d) by a decrease in elasticity on the part of some elastic substance previously present. We should at the present time choose the third alternative as being the truest, in view of the loss of water and carbon dioxide which the egg suffers as it develops, but Mayow chose the fourth, thinking it probable that the "air distributed among the juices of the egg loses its elastic force on account of the fermentation produced among these juices by incubation". Now since the egg-contents are compacted into smaller bulk by the process of incubation, a vacuum would be created somewhere if Nature had not, with her customary prudence, inserted a small amount of air into the air-space which might in due course expand and avoid this. His proof for this was an inaccurate observation; he thought he saw, in eggs at a late stage, when the contents were removed, the air-space collapse to the normal size which it occupies in unincubated eggs. He expressly says that his theory does not depend upon the conception of horror vacui, but that, by the compressive


action of the imprisoned air, the fluids of the egg would be forced into the umbiHcal vessels, and the particles composing the embryonic body packed more tightly together. "The internal air appears to perform the same work as the steel plate bent round into numerous coils by which automata are set in motion."

With this ingenious but erroneous supposition Mayow concludes what is undoubtedly the first great contribution to physiological or biophysical embryology. His views on foetal respiration were soon generally accepted, as the writings of Zacchias, Viardel, Pechlin and John Ray show, but Sponius as late as 1684 was asserting that the lungs of the foetus were functional in utero, absorbing from the amniotic liquid the nitro-aerial particles which P. Stalpartius supposed the placenta to be secreting into it. It is interesting to note that by Mayow's own air-pump method Bohn found nitro-aerial particles in the uterine milk in 1686, and Lang found them in the amniotic liquid in 1 704. The problem had by then arrived at a stage beyond which it could not progress in the absence of quantitative methods.

The year 1675 saw the publication of Nicholas Hoboken's useful treatise on the anatomy of the placenta, and of the English edition of P. Thibaut's Art of Chymistry. I mention the latter here, because of a reference to the special conditions of embryonic life which is found in it. As yet no real help was being given to embryology by contemporary chemistry.

The Magistery and Calx of Egg-shells.

Obs. 2. That you must use the eggshells of hens and not of ducks, geese, or turkeys because that hens eggshells easier calcin'd being thinner by reason that a hen is a more temperate animall; waterfowl are hotter and by reason of their heat do concoct and harden their eggshells more than other fowl ; and from thence it comes that you must have a greater quantity of your Dissolvant, employ more heat, and spend more time to calcine the eggs of waterfowl than those of hens.

About this time also Francis Willoughby published his famous book on birds, an attempt to bring Aldrovandus up to date, in which a good picture is given of the embryological knowledge of the time, although no new observations or theories are given. Another contemporary review is that of Barbatus.

In 1677, spermatozoa were discovered, as announced by Hamm and Leeuwenhoek in the Philosophical Transactions of the Royal Society,


though Hartsoeker afterwards claimed that he had seen them as early as 1674, but had not had sufficient confidence to publish his results. There is a reference to this in the letters of Sir Thomas Browne, who, writing to his son, Dr Edward Browne, on December 9, 1679, said, "I sawe the last transactions, or philosophicall collections, of the Royal Society. Here are some things remarkable, as Lewenhoecks finding such a vast number of little animals in the melt of a cod, or the liquor which runnes from it ; as also in a pike ; and computeth that they much exceed the number of men upon the whole earth at one time, though hee computes that there may bee thirteen thousand millions of men upon the whole earth, which is very many. It may bee worth your reading".

At the same time as these events were taking place, Robert Boyle, at Oxford and London, was engaged in carrying out those experiments in chemistry which led him before long to write his Sceptical Chymist. It is not generally known that in this work, which appeared in 1680, and which set the key for the whole spirit of subsequent physico-chemical research, Boyle has a reference to embryology, and, curiously enough, in connection with a point which, although it is easily seen to be of the highest importance, has been quite overlooked by the commentators upon him. One of the main things he was trying to urge was that, until some system could be proposed which would give a means of quantitative estimation of the constituents of a mixture, no further progress would be made. He was asking, in fact, that chemistry should become an exact science, and his demand is only veiled by the unfamiliarity of his language. His preference for the "mechanical or corpuscularian" philosophy was mainly due to his realisation that, unless chemistry was going to start measuring something, it might as well languish in the obscurity to which Harvey would have willingly relegated it. Thus he says, "But I should perchance forgive the Hypothesis I have been all this time examining (that of the alchemists), if, though it reaches but to a very little part of the world, it did at least give us a satisfactory account of those things which 'tis said to teach. But I find not that it gives us any other than a very imperfect information even about mixt bodies themselves; for how will the knowledge of the Tria Prima discover to us the reason why the Loadstone drawes a Needle, and disposes it to respect the Poles, and yet seldom precisely points at them? how will this hypothesis teach us how a Chick is formed

SECT. 3]



in the Egge, or how the seminal principles of mint, pompions, and other vegetables, can fashion Water into various plants, each of them endow'd with its peculiar and determinate shape and with divers specifick and discriminating Qualities? How does this hypothesis shew us, how much Salt, how much Sulphur, how much Mercury must be taken to make a Chick or a Pompion? and if we know that, what principle is it, that manages these ingredients and contrives, for instance, such liquors as the White and Yolke of an Egge into such a variety of textures as is requisite to fashion the Bones, Arteries, Veines, Nerves, Tendons, Feathers, Blood and other parts of a Chick; and not only to fashion each Limbe, but to connect them altogether, after that manner which is most congruous to the perfection of the Animal which is to consist of them? For to say that some more fine and subtile part of either or all the Hypostatical Principles is the Director in all the business and the Architect of all this elaborate structure, is to give one occasion to demand again, what proportion and way of mixture of the Tria Prima afforded this Architectonick Spirit, and what Agent made so skilful and happy a mixture?" Boyle's instance of the magnetic needle pointing nearly, not exactly, at the north, and his use of the expressions "how much, how many, proportion, way of mixture", indicate that he was moving towards a quantitative chemistry, and by express implication a quantitative embryology. Elsewhere he says that he thinks the Tria Prima will hardly explain a tenth part of the phenomena which the "Leucippian" or atomistic hypothesis is competent to deal with. Thus, although Boyle made few experiments or observations on embryos, he occupies a very important position in the history of embryology. During the last two decades of this century, the Oxford Philosophical Society were occupied on a good many occasions with problems relating to embryology. It is extremely interesting to note, in connection with what we have just seen in Boyle, that John Standard of Merton College reported on February 10, 1685, "the following obbs. concerning ye weight of ye severall parts of Henn's eggs ; done with a pair of scales which turned with \ a grain.

ozs. dr.

A henn's egg weighed 2

The skin weighed

The shell The yolk The white





Loss in weighing



ozs. dr. scr. grns.

Another raw egg of the same sort ... 2 i 2 13


The former egg boiled

Lost in boiling

The skin

The shell

The yolk

The white

2 I I 19

.2 I I 18

• - - - 15


- I 2 19

- 5 - 7 I 2 - 13

Loss in weighing 5

Another early quantitative observation was that of Claude Perrault who found about 1680 that developing ostrich eggs lost one-ninth of their weight in five weeks. The Oxford Philosophical Society, however, preferred as a rule to consider more unusual things, such as "the egges of a parrot hatched in a woeman's bosome, a hen egg figur'd like a bottle, a hen egg that at the big ende had a fleshie excrescence, another hen-eg, monstrous, a suppos'd cocks egg, and the eggs of a puffin, an elligug, and a razor-bill". Mention of these different kinds of eggs reminds us that the systematic collection and classification of eggs had been begun some years before by Sir Thomas Browne (as may be seen in John Evelyn) and by John Tradescant. About this time R. Waller made some noteworthy observations on the "spawn of frogs and the production of Todpoles therefrom", extending the work begun by Swammerdam not long before. Mauriceau now gave a description of the phenomenon of sterile foetal atrophy. The century fittingly closes with Michael Ettmiiller's ponderous treatise, in which all the embryological work of the seventeenth century is summarised with considerable accuracy. He supported the moribund menstruation theory of embryogeny with the argument that animals do not menstruate because they are more prolific than men, and therefore all their blood is required for generation. Garmann's Oologia curiosa, which appeared in 1691, is worth mention also, as a review of the knowledge of the time. But that his work was what the booksellers' catalogues describe as "curious" is shown by the following chapter-headings: De ovo mystico, rnpthico, magico, mechanico, medico, spagyrico, magyrico, pharmaceutico.

3-9. The Theories of Foetal Nutrition

During the course of the seventeenth, and the first quarter of the eighteenth, century, many theories were propounded concerning foetal nutrition. It is convenient to classify them.


I. That the embryo was nourished by the menstrual blood.

Beckher, 1633.

Plempius, 1644. (He did not deny that the umbilical cord was

functional, but insisted that the blood passing through it was


In 1 65 1 Harvey's work was published. Sennertus, 1654. Seger, 1660. van Linde, 1672. F. Sylvius, 1680. Cyprianus, 1700.

II. That the embryo was nourished by its mouth. {a) By the amniotic liquid.

(A) In addition to the umbilical blood. Harvey, 1651. W. Needham, 1667. de Graaf, 1677.

C. Bartholinus, 1679. van Diemerbroeck, 1685. Ortlob, 1697.

D. Tauvry, 1700. Linsing, 1701. PauH, 1707. Barthold, 1717.

S. Middlebeek, 17 19. Teichmeyer, 17 19. Gibson, 1726.

(B) Alone; the umbilical blood being regarded as unnecessary or of minor importance, Moellenbroeck, 1672. Cosmopolita, 1686. Everardus, 1686. P. Stalpartius, 1687. Bierling, 1690.

Case, 1696. (Case thought the embryo arose entirely out of the amniotic liquid like a precipitate from a clear solution.) Berger, 1702.

These persons referred as their principal experimental basis to cases in which embryos had been born without umbilical cords, e.g. of those of: Rommelius, 1675 (in Velsch). Valentinius, 1 7 1 1 .


(b) By the uterine milk or succum lacteo-chylosum.

Mercklin, 1679. Drelincurtius, 1685. Bohnius, 1686. Zacchias, 1688. Tauvry, 1694. Franc, 1722. Dionis, 1724.

III. That the embryo was nourished through the umbiHcal cord only.

{a) By foetal blood (the circulations distinct).

Arantius, 1595.

Harvey, 1651.

W. Needham, 1667.

F. Hoffmann, 1681. (He proved the point by injection long before Hunter, who is stated by Cole to have been the first to demonstrate this.)

Ruysch, 1 70 1.

Snelle, 1705.

Falconnet, 171 1.

It is to be noted that Bierling, P. Stalpartius, Berger, Barthold, and Charleton, who supported the discontinuity theory of the circulations, were all upholders of the theory of foetal nourishment per os, so that their reasons for doing so were not those on account of which we agree with Hoffmann and Needham at the present time.

{b) By maternal blood (the circulations continuous).

Laurentius, 1600.

de Marchette, 1656.

Rallius, 1669.

Muraltus, 1672.

Blasius, 1677.

Veslingius, 1677.

Hamel, 1700.

de Craan, 1703.

Lang, 1704.

van Home, 1707.

Freind, 171 1. (Freind's Emmenologia deserves a special mention. He proved by a calculation that the amount of blood passing through the umbilical cord would be sufficient for the needs of the embryo. This is a parallel to Harvey's famous calculation about the circulation of the blood. He also quotes some experiments of


Rayger and Gayant, who injected a blue dye into the foetal circulation and found it again in the maternal. Therefore he regards it as continuous.)

Mery, 171 1. (Mery combated Falconnet's view of the separate circulations. He said that he had not himself tried Falconnet's experiment, but that some students had, and could not repeat it.)

Aubert, 1 7 1 1 . (Narrative of a case in which the umbilical cord had not been tied at the maternal end and the mother had nearly bled to death through it.)

Nenterus, 17 14.

Wedel, 1 71 7.

Bellinger, 171 7. (Bellinger believed that the maternal blood was transformed by the embryonic thymus gland into proper nourishment for itself, after which it was secreted into the mouth by the salivary ducts and so went to form meconium without the necessity for deglutination. Heister's comments on this extraordinary theory are worth reading. Perhaps Bellinger was indebted to Tauvry for his idea of the importance of the thymus gland. Tauvry had drawn attention in 1700 to its diminution after birth.)

de Smidt, 17 18.

Dionis, 1724.

(c) By menstrual blood.

Plempius, 1644.

(d) By uterine milk.

Ent, 1687.

Camerarius, 17 14. {Opinio conciliatrix!)

F. Hoffmann, 1718.

{e) By the amniotic fluid.

Vicarius, 1700. Goelicke, 1723.

IV. That the embryo was nourished by pores in its skin.

Deusingius, 1660. Nitzsch, 1 67 1. Stockhamer, 1682.

This was suggested on the ground that in the earlier stages of development there is no umbilical cord. In 1684 St Romain argued against it on the ground that, if it were true, the embryo would dissolve in the amniotic liquid.


During this period also there were continued disputes about the origin of the amniotic liquid, van Diemerbroeck and Verheyen considered that it could not be the sweat of the embryo, for the embryo was always much too small to account for it, and, moreover, Tertre had described cases where the secundines had been formed with the membranes but in the absence of the embryo. Dionis affirmed that, whatever it was, it could not be urine, for urine will not keep good for nine days, a fortiori not for nine months. Drelincurtius put forward a theory that the embryo secreted it from its eyes and mouth by crying and salivating, while Bohn and Blancard derived it from the foetal breasts. Lang, Berger and Gofey criticised this notion without bringing forward anything constructive, and Gofey was in his turn annihilated by D. Hoffmann, who with Nenter and Konig supported the modern view, namely, that it was a transudation from the maternal blood-vessels in the decidua. The question was complicated further by the alleged discovery by Bidloo in 1 685 of glands in the umbilical cord, and by Vieussens in 1 705 of glands on the amniotic membrane. J. M. Hoffmann and Nicholas Hoboken supported the view that these were the important structures. There the problem was left during the eighteenth century, various writers supporting different opinions from time to time, and it is still under discussion (see Section 22).

Very early in the eighteenth century (1708) there appeared a work by G. E. Stahl, van Helmont's most famous follower, which struck the keynote of the whole century. Stahl's Theoria Medica Vera, divided as it was into Physiological and Pathological sections, belonged in essence to the a priori school of Descartes and Gassendi. It differed from them profoundly, of course, for, instead of trying to explain all biological phenomena, including embryonic development, from mechanical first principles, it started out from first principles of a vitalistic order, and, having combined all the archaei into one informing soul, it sought to show how the facts could be perfectly well explained on this basis. But the spiritual kinship of Stahl with Descartes and Gassendi is due to an atmosphere which can only be called doctrinaire, and which was common to them all. Like the methodist school of Hellenistic medicine, they subordinated the data to a preconceived theory, during which process any awkward facts were liable to be rather submerged than subordinated.


In 1722 Antoine Maitre-Jan published his book on the embryology of the chick, the only one on this subject between Malpighi and Haller. It was an admirable treatise, illustrated with many drawings which, though not very beautiful, were as accurate as could be expected at the time. Perhaps its most remarkable characteristic is its almost complete freedom from all theory — Maitre-Jan says hardly a word about generation in general, and is far from putting forward a "system" in the usual eighteenth-century manner. He contents himself with the recital of the known facts, including those added by his own observations. He gives no references, and writes in an extremely modern and unaffected style.

The only traces of theoretical presupposition which can be found in him are Cartesian, for he speaks of the activity of ferments in blood-formation. He is an epigenesist, and long before Brooks, he gives the right explanation of Malpighi's error, affirming that the hot Italian summer was responsible for some development in Malpighi's eggs before Malpighi examined them. Maitre-Jan's book must have been accessible both to Buffon and Haller, so it is difficult to see why they should have perpetuated Malpighi's mistake till nearly the end of the century.

In technique, Maitre-Jan was pre-eminent. He was the first embryologist to make practical use of Boyle's suggestion regarding "distilled spirits of vinegar" for hardening the embryo so that it could be better dissected. He also used "weak spirits of vitriol"; after treating blastoderms with it, he said, "I saw with pleasure an infinity of little capillary vessels which had not appeared to be there before". He made a few chemical experiments also, noting that vinegar would coagulate egg-white, and estimating quantitatively the difference in oil-content of different yolks — though for this he gives no figures.

His theory he relegated to an appendix entitled Objections sur la generation des animaux par de petits vers. There were sixteen of them, but the most cogent one was that, as little worms had been found under the microscope in pond-water, vinegar, and all kinds of liquids, there was no reason to suppose that those in the semen were in any essential way connected with generation. For his time, this argument was an excellent one, and was open to no demur save on the ground of filtration experiments which had not yet been made (see p. 215).


About this time there was some controversy over the circulation of blood, the foramen ovale, etc., in the embryo. From 1700 to 1710, Tauvry and Mery were engaged in a polemic on this subject, and the latter also corresponded with Duverney, Silvestre and Buissiere in a controversy which recalls that of Laurentius and Petreus a hundred years before. Nicholls wrote later on the same subject. Daniel Tauvry was interesting, however, for other reasons. He was an epigenesist, and wrote vigorously against the view that the soul constructed during embryogeny a suitable home for itself.

Nine years later two books appeared, which form very definite landmarks in the history of embryology. One was Martin Schurig's Embryologia, and the other the Elementa Chymiae of Hermann Boerhaave.

The former, however, gave to the world no new experiments or observations ; it was the first of what we should now call the typical "review" kind of publication. Schurig saw that he was living at the end of a great scientific movement following the Renaissance, and set himself accordingly for many years to compile large treatises on definite and restricted subjects, taking care to give all references with meticulous accuracy, and to omit no significant or insignificant work. His Spermatologia was the first to appear (in 1720), and it was followed in 1723 by Sialologia (on the saliva), Chylologia (1725), Muliebria (1729), Parthenologia (1729), Gynaecologia (1731) and Haematologia (1744). His Embryologia was the last but one of the series. In it he treated compendiously of all the theories which had been advanced about embryology during the immediately preceding two centuries, and his chapters on foetal nutrition and foetal respiration throw a flood of light on to the "intellectual climate" in which Harvey and Mayow worked, providing, as it were, the perishable background of their immortal thoughts. Schurig's bibliography is a very striking part of his book, extending to sixteen pages, and including five hundred and sixty references; it was the first attempt of its kind.

3-10. Boerhaave, Hamberger, Mazin

Hermann Boerhaave was a more prominent figure, a Professor at Leyden for many years, and renowned for his encyclopaedic learning on all subjects remotely connected with medicine. His Elementa Chymiae, which became the standard chemical book of the whole period, demonstrates throughout the exceedingly wide outlook of its


author, and contains in the second volume what must be regarded as the first detailed account of chemical embryology. I reproduce here the relevant passages in full because of their great interest. It will be noted that they are cast in the form of lecture addresses, as if they had been taken down direct from the lectures of the Professor, a fact which gives them a peculiar charm when it is remembered how many great men must have listened to them, among them Albrecht von Haller and Julien de la Mettrie. In considering what follows, it should be noted that Boerhaave's interest is biological all the time, and that he does not treat the liquids of the egg, as nearly all the chemists before him had done, as substances of curious properties indeed, but quite remote from any question relating to the development of the embryo. Another interesting point is that he deals only with the white, and hardly mentions the yolk; this is perhaps to be explained by the Aristotelian theory that the embryo was formed out of the white, and only nourished by the yolk {ex alb fieri, ex luteo nutriri), a theory which was still alive, in spite of Harvey, in the first half of the eighteenth century. If this was what was at the bottom of Boerhaave's mind, then it is obvious that the egg-white would be to him the liquid inhabited more particularly by the plastic force. This, then, is what he has to say about the biochemistry of the egg.

Op. Chem. in Animalia. [Processus log.] The albumen of a fresh egg is not acid, nor alkaline, nor does it contain a fermented spirit. The white of a fresh egg, separated from the shell, the membranes, and the yolk, I enclose in clean glass vessels, and into each of these I pour different acids, and shake them up, mixing them, and no sign of ebullition appears however I treat them. Therefore I lay these vessels aside. Now in these other two vessels I have two fresh portions of albumen, and I mix with them in one case alkaline salt and in the other volatile alkali. You see they are quiet without any sign of effervescence. Now behold a remarkable thing, in this tall cylindrical vessel is half an ounce of the albumen of an egg and two drams of spirits of nitre, in this other vessel is half an ounce of egg-white, together with four and a half ounces of oil of tartar per deliquium both heated up to 92 degrees. Pray observe and behold, with one movement I pour the alkaline albumen into the acid albumen, with what fury they boil up, into what space they rarefy the mass, so that they stream out of the vessel although it is ten pints in size [decupli capace] . They have scarcely changed their colour. But when the effervescence has abated how suddenly they return to the limits of space occupied before. But now if more egg-white is heated to 100 degrees in a retort [cucurbita] an insipid water containing


no spirit is given off. If egg-white is applied to the naked eye or naked nerve it does not give the smallest sense of pain, and scarcely affects the smell; nothing more inert and more insipid can be put on the tongue. It appears mucous and viscid to the touch, not at all penetrable. Hence in the fresh white of an egg there is no alkali or acid, or both together. It is indeed a thick, sticky, inert, and insipid liquor, yet from this truly vital liquid at a heat of 93 degrees within the space of 2 1 days the chick grows in the incubated egg from a tiny mass hardly weighing a hundredth of a grain into the perfect body of an animal, weighing an ounce or more. We have learnt therefore of a liquid distinct from all others, from which by inscrutable causes fibres, membranes, vessels, entrails, muscles, bones, cartilages, and all the other parts, tendons, ligaments, the beak, the claws, the feathers, and all the humours can be produced — and yet in this liquid we find softness, inertia, absence of acid, alkali, and spirit, and no tendency to effervesce. Indeed, if there were the slightest effervescence in it, it would certainly break the eggshell, therefore we see from how slow and inactive a mass all the solid and fluid parts of the chick are constructed. And yet this itself is rendered absolutely useless for forming the chick by greater heat. It scarcely bears 100 degrees with good effect but at a less temperature never brings forth a chick, for under 80 degrees will not suffice. But by a heat kept between these limits, there is brought about so marvellous an attenuation of the mucous inactivity that it can exhale a great part through the shell of the egg and the two membranes, the yolk and chalazae alone remaining along with the amniotic sac. For the yolk, the uterine placenta of the chick, takes little part in the nourishment. Meanwhile Malpighius has shown that this albumen is not a liquid of a homogeneous kind, as the blood-serum flowing through the vital vessels is, but that it is a structure composed of numerous membrane-like and distinct small saccules, filled with a liquid of their own, in the same way as in the vitreous humour of the eye.

[Processus 1 1 1 .] Exploration of the egg-white with alcohol. In this transparent vessel is the albumen of an egg, and into it, as you perceive, I gently pour the purest alcohol, so that it descends down the sides of the vessel and reaches the albumen. I do this deliberately and with such solicitude that you may see the surface of the albumen which, touching the alcohol, holds it up, being immediately coagulated, while the lower part remains liquid and transparent. As I now gently shake them together, it appears evident that wherever the alcohol touches the albumen a concretion is formed. Behold now, while I shake them up thoroughly together, all the egg-white is coagulated. If alcohol previously warmed is employed in this experiment, the same result is brought about but more rapidly. It appears therefore that the purest vegetable spirits immediately coagulate the plastic and nutrient material.

[Processus 112.] The fresh albumen of an egg is broken up by distillation. These fresh eggs have been cooked in pure water till they became hard. I now take the shining white, separating off all the other things, and break it up into small pieces. I put these, as you see, into a clean glass retort


[cucurbita] and I duly cover it by fitting on an alembic and add a receiver. By the rules of the (chemical) art I place the whole retort in a bath of water and I apply to it successive degrees of fire until the whole bath is boiling. No vaporous streaks [^strid] of spirits are given off but simple water in dewy drops and this in incredible quantity, more than nine-tenths. I continue so with patience until by the heat of boiling water no more drops of this humour are given off. Then this water shows no trace of oil, salt, or spirit ; it is perfectly transparent and tasteless, except that it eventually grows rather sour. It is odourless, save that towards the end it gives off a slight smell of burning. It shows absolutely no sign of the presence of any alkali, when I test it in every way, as you can see for yourselves ; nor does it reveal any trace of acid, when tried how you will. Here you see pounds of this water, but in the bottom of the now open retort see, I beg of you, how little substance remains. Behold, there are fragments contracted into a very small space in comparison with the former quantity. They are endowed with a golden yellow colour, especially where they have touched the glass, but yet they are transparent after the manner of coloured glass. When I take them out I find them very light, very hard, quite fragile, and breaking apart with a crack, smelling slightly of empyreuma, with a taste rather bitter from the fire, and without any flavour of alkali or acid. This is the first part of the analysis. Now I take these remaining fragments in a glass retort [retortam] in such a way that two-thirds remain over. I put the retort into a stove of sand, first arranging a large receiver. Then thoroughly luting all the joints I distil by successive grades of fire and finally by the highest which I call suppressionis. There ascends a spirit, running in streaks [^striatim] fat and oily, and at the same time, volatile salts of solid form everywhere on the walls of the vessel, rather plentiful in proportion to the dried fragments but small in proportion to the whole albumen before the water had been removed from it. Finally an oil appears besides the light golden material mixed with the first, black, thick, and pitchy. When by the extreme force of the fire this oil is finally driven forth, then the earth in the bottom, closely united with its most tenacious oil, swells up and is rarefied and rises right up to the neck of the retort so that had the retort been overfull it would have entered into the neck and clogged it up, even causing it to burst, with danger to the bystanders. The operation is to be continued till no more comes out. That first spirit, oily and fatty, is clearly alkaline by every test, as you may tell from the way it effervesces when acid is poured on it. If we rectify it we resolve it into an alkaline volatile salt, an oil, and inert foetid water. The salt fixed to the walls is completely alkaline, sharp, fiery, oily, and volatile; and the final oil is specially sharp, caustic, and foetid. The black earth which remains in the retort is shiny, light, thin, and fragile, foetid from the final empyreumatic oil, and soft because of it. If then it is burnt on an open fire, it leaves a little fixed earth which is white, insipid, tasteless, and odourless, from which scarcely any salt can be extracted, but only a very heavy dusty powder \^pollinein\.

Cf. the dry distillation of egg-white by Pictet & Cramer in 1919.


[Processus 113.] The fresh albumen of an egg will putrefy. Sound eggs kept at 70° for some days will become foetid and stink. . . .We have learnt then that this is the nature of the material which will shortly be changed into the structure, form, and all the parts of the animal body. Repose and a certain degree of heat produce that effect in that material. We observe therefore the spontaneous corruption and change of the material, and what is extremely remarkable, if an impregnated egg is warmed in an oven [in hypocaustis] to a heat of 92 degrees it employs these attenuated parts changed by such a heat to nourish, increase, and complete the chick for 21 days. But in this chick nothing alkaline, foetid, or putrid is found, hence observe, O doctors [medici] , the remarkable manifestations of nature ^by repose and a certain degree of heat a thick substance becomes thin, a viscous substance becomes liquid, an odourless substance becomes foetid, an insipid substance becomes sour and extremely acrid and bitter to the taste, a soothing substance becomes caustic, a non-alkali becomes alkaline, a latent oil becomes sweet and putrid. Let these results be compared with the observations of Marcellus Malpighius on the incubated egg, and we shall observe things which shall surprise us. I took care to investigate only the albumen of the egg first of all, separating the other parts off where possible, for the albumen alone forms the whole of the material which proceeds to feed [in pabulum] the embryo. The other constituents of the egg only assist in changing the albumen, so that when it is changed, it miay be applied to forming the structure of the chick.

Boerhaave's treatment of these subjects has only to be compared with that of Joachim Beccher, who wrote in 1 703, to show how thoroughly modern in outlook it is. Beccher's Physica Subterranea contains a whole section devoted to the growth of the embryo, but it is extremely confused and very alchemical in its details. The advance made in the thirty years between Beccher and Boerhaave was immense, but, if the biochemistry of development advanced so fast, its biophysics was not far behind, as is shown by the work of G. E. Hamberger and J. B. Mazin.

Hamberger's most important contributions, contained in his Physiologia medica of 1 75 1 , were his quantitative observations on the watercontent of the embryo and its growth-rate, in which he had no forerunners, Hamberger showed "that there are much less solid parts in the foetus than in the adult. The cortical substance of the brain of an embryo loses 8694 parts in 10,000 on drying but in the adult it only loses 8096 and that of the cerebellum from 81 parts is reduced to 12. The maxillary glands of the embryo lose out of 10,000 parts 8469, the liver 8047, the pancreas 7863, the arteries 8278 and even the cartilages lose four-fifths of their weight, decreasing from 10,000 to 8149I ". The


corresponding figures for the adult were: liver 7192, and heart 7836. These figures do not widely diflfer fi:-om those obtained in recent times.

J. B, Mazin published his Conjecturae physico-medico-hydrostaticae de respiratione foetus in 1737 and his Tractatus medico-mechanica in 1742. In the first of these works Mazin supports what is essentially Mayow's theory of embryonic respiration, without, however, mentioning Mayow more than once. It had not been popular since 1700, though Pitcairn had defended it. Mazin put the liquids of eggs under an air-pump, and observing that air could be extracted from them affirmed that the air was hidden in them and that the embryo could therefore respire. He spoke of "aerial particles" in the amniotic liquid, and discussed the respiration of fishes in connection with this. The specific gravity of the embryo also interested him, and he did a great deal of calculation and experiment on it. His most interesting passage, perhaps, is that in which he mentions the "Eolipile" of the Alexandrians, the primitive form of the steam-engine, and says that just as the heat of the fire makes the water boil, so the heat of the viscera makes the amniotic liquid boil, giving off respirable vapours. The time-relations of this analogy are interesting, for in 1705 Thomas Newcomen had succeeded in making a steam-engine which worked with considerable precision, and the question of steampower was widely discussed. Possibly Mazin was acquainted with the Marquis of Worcester's Century of the Names and Scantlings oj Inventions, which had been published in 1663, and which had contained an aeolipile or "water-commanding machine". England was the centre of this movement and other countries employed Englishmen as engineers; Humphrey Potter, for instance, erected a steamengine for pumping at a Hungarian mine in 1 720.

As for the discovery of oxygen, it was near at hand, and Scheele in 1 773 and Priestley in 1 774 were soon to supply the knowledge without which Mazin could not proceed further.

In his second book, Mazin reported many quantitative observations on the specific gravity of the embryo. He found that it diminished as development proceeded, being to the amniotic liquid as 282 to 274 in the fourth month and as 504 to 494 in the fifth month.

Another instance of the way in which experimental physical questions now began to come in is afforded by the work of Joseph Onymos, whose De Matura Foetu of 1 745 spoke of the specific gravity of the embryo at different stages of development.


These writers, together with Haller himself, and J. C. Heffter who handled problems of embryonic rate of growth contribute to one of the best, because most quantitative, aspects of eighteenthcentury embryology.

3*11. Albrecht von Haller and his Contemporaries

Boerhaave's greatest pupil was Albrecht von Haller. Like O. W. Holmes, at Harvard, Haller occupied a "settee" rather than a "chair", at Gottingen, and taught not only physiology but also medicine and surgery, botany, anatomy and pharmacology. Nor did he merely deal with so many subjects superficially; in each case he published what amounted to the best and most complete text-book up to then written. Haller was made Professor in 1736, and for many years worked at Gottingen, devoting much of his time to embryological researches, which, with those of his opponent Wolff, stand out as the greatest between Malpighi and von Baer. In 1 750 he published a series of dissertations and short papers on all kinds of physiological subjects, which would have been the direct ancestors of the modern compilations of groups of experts, had they been more systematically arranged. The volume on generation repays some study. The contributions relevant to the present discussion had been written at various times during the previous seventy years, and may be summarised as follows :

IV. Christopher Sturmius, De plantarum animaliumque generatione. (First published 1687.) In this paper Sturmius argues on behalf of the preformation theory "which in our times does not lack supporters", quoting Perrault, Harvey and Descartes. He contents himself with countering arguments which had been urged against it, as, {a) spontaneous generation, {b) annual recurrence of plants, {c) insect metamorphosis, {d) generation without copulation. V. Rudolf Jacob Camerarius, Specimen experimentorum physiologicotherapeuticorum circa generationem hominis et animalium. The most interesting thing about this is that Camerarius mentions the observations of D. Seiller, a sculptor, who had ascertained that the body is five times the size of the head in the embryo but seven and a half times the size of it in the adult. This is in the direct line between Leonardo and Scammon.

SECT. 3]



XV. Philip Gravel, De Super Joetatione. (First published 1738.)

XVIII. Adam Brendel, De embryone in ovulo ante conceptum praeexistante. (First published 1703.) Brendel "stands for the Graafian hypothesis. Unfortunately, he was also a preformationist and believed that every limb, organ, and function existed not potentially but actually in the unfertilised Qgg before its passage down the Fallopian tube.

XXII. Camillus Falconnet, Non est fetui sanguis maternus alimento. (First published 171 1.) This is the first of the French contributions to the book; they are all very markedly shorter than the German ones and much less heavily ornamented with irrelevant quotations. Falconnet is concerned to prove that the maternal and foetal circulations are separate, and he describes in an admirably concise manner an experiment in which he bled a female dog to death, after which, opening the uterus, he discovered that the embryonic blood-vessels were full of blood although those of the mother had none in at all. Arantius was therefore justified. Falconnet was soon confirmed by Nunn.

XXIII. Jean de Diest's Sui Sanguinis solus opifex fetus est (first published 1735) was written to prove a similar point. He refers to the experiment of Falconnet and the injections of F. Hoffmann, and criticises Cowper's experiment in which mercury had been injected into the umbilical vessels and found in the maternal circulation, on the grounds that mercury is so "tenuous and voluble" that it might pass where blood could not pass normally. He also objects to the view that the foetus is nourished by the amniotic liquid.

XXIV. Francis David Herissant, Secundinae fetui pulmonis praestant officia, et sanguine materno fetum non alitur. (First published in 1 741.) An excellent paper, in which the respiratory function of the placenta is proved by the observation that the foetal blood-vessel leading to the placenta is always full of dark venous blood, while that leading away fi-om the placenta is light and arterial [floridiori coccineoque colore, ut ipsemet observavi]. Herissant adduces also the cases of acephalic monsters, such as that of Brady, which could not possibly have drunk up any amniotic fluid, and yet were fully formed


in all other respects. He concludes that the umbilical cord serves for respiration and nutrition.

XXV. After these three French workers, there is a great drop to Johannes Zeller, whose Infanticidas non absolvit nee a tortura liberal pulmonum infantis in aqua subsidentia (first published

1 691) is a long-winded discussion of the floating lung test in forensic medicine. His memory deserves a word of obloquy for his vigorous insistence upon torture and death for infanticide even during puerperal insanity. Perhaps it was Zeller who called forth the noble answer of de la Mettrie to this inhumanity in his Man a Machine.

XXVI. Zeller's De Vila Humana ex June pendenle (first published

1692) is no better, though at the time, perhaps because of its striking title, it was famous. It deals with the ligation of the umbilical cord at birth.

This completes the list of the papers published by Haller in his 1750 collection. He retired from the Gottingen chair three years later, and in 1757 the first volume of his Elemenla Physiologiae was published, probably the greatest text-book of physiology ever written. It appeared only by slow degrees, so that it was not until 1766 that the embryological section was available. This volume contains a discussion of a mass of literature, most of which had arisen during the preceding twenty-five years, for, although many of the names mentioned by Haller occur also in Schurig, yet many are quite new.

Haller himself published in 1 767 a volume of his collected papers on embryology, most of which were concerned with the developing heart of the chick, which he worked out very thoroughly, in collaboration with Kuhlemann. (Kuhlemann had already done for the sheep what Harvey had done for the doe.) He made a beginning with the quantitative description of embryogeny, and one of his tables showing the changing lengths of the bones is reproduced herewith (Fig. 10). He was a convinced preformationist, a fact which was largely due to his researches on the hen's egg, where he observed that the yolk had a much more intimate connection with the embryo than had previously been supposed. Since the whole yolk was part of the embryo, as it were, the preformation theory seemed to him to fit the facts better than epigenesis.





Decimus Undeci MUS





Quart us




SEXTys Decimus Septimus Decimus Octavus Decimus

NdNUs .Vigesi l| MUS VlGESlMUS











1 8- prox.










+4. ptox.




















o o 10

1 1



43. prox

39** 73 74 70





o o




3 0. prox.

29 48 53 52



i"4rWo 8000

I38?4 27000

72301 8o^24.prox. 110^92 itf<?379 5.00149



1017575 1098712 13242^0






5832 loiCi

29791 305^8


85284 125600

257505 320047


44<^^75 483736

571787 ^76381

Fig. 10. Facsimile of a table in A. von Haller's Elementa Physiologiae of 1766, containing some of his observations on the growth in length and weight of embryonic bones in the chick.


Haller went further than Schurig, in that he usually gave an opinion of his own after summarising those of other people, but his views were by no means always enlightened, and the atmosphere of Buffon is, on the whole, more congenial to us than that of Haller. Haller, for example, believed that the amniotic liquid had nutritious properties, and that the nutrition of the embryo in mammalia was accomplished first of^ all per os and afterwards per umbilicum. He denied that the placenta had any respiratory function, and, indeed, his whole teaching on respiration was retrograde. He mentions, however, an experiment of Nicolas Lemery's, in which it had been found that indigo would penetrate the shell of a developing hen's tgg from the outside. Consequently, air might do so too, and Vallisneri had shown that, if an egg was placed in boiled water under an air-pump, the air inside would rush out through the shell and appear in the form of bubbles.

Haller was much more progressive in holding the origin of the amniotic liquid (according to him a subject of extraordinary difficulty — " solutionem non promittam'") to be a transudation from the maternal blood-vessels. He followed Noortwyck in asserting the separateness of the maternal and foetal circulations in mammalia. He opposed the existence of eggs in vivipara — "We may conclude from all this", he said, "that the ovarian vesicles are not eggs and that they do not contain the rudiments of the new animal". But he accepted it in the restricted sense that the embryonic membranes resembled an egg, thus: "If we call an egg a hollow membranous pocket full of a humour in which the embryo swims, we may admit the opinion of the older authors who derive all animals from eggs with the exception of the tiny simple animals of which we have already spoken. It was in this sense that Aristotle and Empedocles before him, said that even trees were oviparous. This has also been confirmed by the experiments of Harvey on insects, fishes, birds, and quadrupeds".

Haller's most original work was in connection with the growthrate of the embryo ; here he struck out, for once, into entirely new country. "The growth of the embryo in the uterus of the mother is almost unbelieveably rapid. We do not know what its size is at the moment of its formation, but it is certainly so small that it cannot be seen even with the aid of the best microscopes, and it reaches in nine months the weight of ten or twelve pounds. In order to clear


up this speculation, let us examine the growth of the chick in the egg. We cannot in this case either measure its size at the moment when the egg is put to incubate but it cannot be more than j^ in. long, for if it were, it would be visible, and yet 25 days later it is 4 ins, long. Its relation is therefore as 64 to 64 millions or i to i million. This growth takes place in a singular manner, it is very rapid in the beginning and continually diminishes in speed. The growth on the first day is from i to gi^^ and what Swammerdam calls a worm grows in one day from one-twentieth or one-thirtieth of a grain to seven grains, i.e. it increases its weight by 140 or 240 times. On the second day the growth of the chick is from i to 5, on the third day, from i to not quite 4, on the fifth day from i to something less than 3. Then from the sixth to the twelfth day, the growth each day is hardly from 4 to 5, and on the twenty-first day it is about from 5 to 6. After the chick has hatched, it grows each day for the first 40 days at an approximately constant rate, from 20 to 2 1 on each day. The increase of the first twenty-four hours is therefore in relation to that of the last twenty-four hours as 546I to 5 or 145 to i. Now as the total increase in weight in the egg is to that of the whole growth period (up to the adult) as 2 to 24 ozs., all the post-embryonic growth is as i to 12, i.e. it is to the growth of one day alone early in incubation as i to 7|.. . .The growth of man, like that of the chick, decreases in rapidity as it advances. Let us suppose that a man, at the instant of conception, weighs a hundred-thousandth of a grain and that a one-month old embryo weighs 30 grains; then the man will have acquired in that time more than 300,000 times the weight that he had to begin with. But if a foetus of the second month weighs 3 ozs. as it approximately does, he will only now have acquired 48 times the weight he had at the beginning of the period. This is a prodigious decrease in speed, and at the end of the ninth month he will not weigh more than about 105 ozs., which is not more than an average increase of 15 per month. A child three years old is about half the size of an adult. If then the adult weighs 2250 ozs. the three-year old child only weighs 281 ozs., which is an eighth of the adult weight. Now from birth to 3 years he will grow from 105 to 281 or as 5 to 14, but in the following 22 years he will only accumulate 2250 ozs. or eight times what he had at 3 years. The growth of a man will therefore be in the first month of intrauterine life as I to 300,000, in the second as i to 48, in each of the

N E I 13


others as i to 15. In the first 3 years of extra-uterine life his growth will be from 164 to 281 and in the succeeding 22 years from 281 to 384, and the growth of the first month to the last will be as 300,000 to ^% or 136,800,000 to 28, or 4,885,717 to i. The whole growth of man will consequently be as 108,000,000 to i."

In spite of the rather unfamiliar language in which these facts are described, and the theory of the growth of the heart which Haller subsequently put forth to explain them, they remain fundamental to embryology. Their quantitative tone is indeed remarkably modern. In my opinion, when all the voluminous writings of Haller are carefully searched through, nothing more progressive and valuable than these figures can be found. Haller and Hamberger stand thus between Leonardo on the one hand and Minot and Brody on the other. That they stood so much alone is only another indication of the extraordinary reluctance with which the men of past generations assented to the truth contained in Robert Mayer's immortal words, "Eine einzige Zahl hat mehr wahren und bleibenden Wert als eine kostbare Bibliothek von Hypothesen".

Of development as a whole, Haller spoke thus, " In the body of the animal therefore, no part is made before any other part, but all are formed at the same time. If certain authors have said that the animal begins to be formed by the backbone, by the brain, or by the heart, if Galen taught that it was the liver which was first formed, if others have said that it was the belly and the head, or the spinal marrow with the brain, adding that these parts make others in turn, I think that all these authors only meant that the heart and the brain or whatever organ it was, were visible when none of the other parts yet were, and that certain parts of the embryonic body are well enough developed in the first few days to be seen while others are not so until the latter part of development; and others again not till after birth, such as the beard in man, the antlers in the stag, the breasts and the second set of teeth. If Harvey thought he descried an epigenetic development, it was because he saw first a little cloud, then the rudiments of the head, with the eyes bigger than the whole body, and little by little the viscera being formed. If one compares his description with mine, one will see that his description of the development of the deer corresponds exactly with mine of the development of the chick. If, more than twenty years ago, before I had made many observations upon eggs and the females of quadrupeds I employed this reasoning to prove


that there is a great difference between the foetus and the perfect animal, and if I said that in the animal at the moment of conception one does not find the same parts as in the perfect animal, I have realised abundantly since then that all I said against preformation really went to support it". The reasons for this change of opinion become no clearer as Haller's writings are more assiduously read, and, as Dareste says, why he should have made it, will always remain a mystery.

The emboitement aspect of preformation presented no difficulties to Haller. "It follows", he said, speaking of the generation of Volvox, "that the ovary of an ancestress will contain not only her daughter, but also her granddaughter, her great-granddaughter and her greatgreat-granddaughter, and if it is once proved that an ovary can contain many generations, there is no absurdity in saying that it contains them all,"

The following passage is interesting. "We must proceed to say what is the efficient cause of the beautiful machine which we call an animal. First of all let us not attribute it to chance, as Ofrai [is this Julien Offi-ay de la Mettrie? Haller had a habit of using Christian names, e,g, Turberville for J. T. Needham] would have us do, for although he pretends that all animals come from earth, he is not attached to the ancient opinion, and nobody now believes what Aelian says, namely that frogs are born from mud. . . , Vallisneri has found the fathers and mothers of the little worms in galls, a quest of which Redi despaired, and Redi in his turn has made with exactitude and precision those experiments which Bonannus, Triumphet, and Honoratus Faber had only sketched out imperfectly. Moreover, no seed, no clover. . . . This was the received opinion but in our century a proscribed notion has been revivified and some great men have pretended that there are little animals which are engendered by an equivocal generation, without father and mother, and that all the viscera and all the parts of these animals do not exist together, but that the nobler parts are formed first by epigenesis and that then the others are formed little by little afterwards." This is an admirable illustration of how spontaneous generation and epigenesis were bound up together, "M. Needham", Haller goes on to say, "does not admit an equivocal generation but he does admit epigenesis, and a corporeal non-intelligent force, which constructs a body from a tiny little germ furnishing the necessary matter for it. He says that there are only



the primitive germs which were made at the original creation and that germs organised Hke animals do by no means pre-exist, for if they did, molae uterinae, encysted tumours, and the like, could not come into being." Haller then goes on to describe Needham's experiments with meat broths, etc., and objects to his "system", largely on the ground that "blind forces without any intelligence could hardly be able to form animals for ends foreseen and ready to take their places in the scheme of beings". He considers that Needham's theories are completely disproved by experiments such as those of Spallanzani, though, curiously enough, he does not quote the latter author in this connection. I shall return to this later.

"Nobody", he goes on to say, "has upheld epigenesis more than M. Wolff, who has undertaken an examination to demonstrate that plants and animals are formed without a mould out of matter by a certain constant force which he calls 'essential' [in his Theoria Generationis] .... I have indeed seen many of the phenomena which he describes, and it is certain that the heart seems to be formed out of a congealed humour and that the whole animal appears to have the same consistency. But it does not follow that because this primitive glue which is to take on the shape of the animal does not appear to possess its structure and all its parts, it has not effectively got them. I have often given greater solidity to this jelly by the use merely of spirits of wine and by this means I saw that what had appeared to me to be a homogeneous jelly was composed of fibres, vessels, and viscera. Now surely nobody will say that the vis essentialis of the spirit of wine gave an organic structure to an unformed matter, on the contrary it is rather in the removal of transparency and the accession of greater firmness to the extremities, as well as the making of a more obvious boundary to the contour of a viscus that one could see the structure of a cellular tissue, which was ready to be formed but which the transparency had previously hidden and the wetness not allowed to be circumscribed by lines. . . . Finally, to cut a long story short, why does this vis essentialis, which is one only, form always and in the same places the parts of an animal which are so different, and always upon the same model, if inorganic matter is susceptible of changes and is capable of taking all sorts of forms? Why should the material coming from a hen always give rise to a chicken, and that from a peacock give rise to a peacock? To these questions no answer is given." This was the


case because Wolff was not a theorist, but rather an experimentalist; his writings are marked by their abstention from the discussion of speculative points. The above passage is very interesting. It reminds us of the great difficulties with which the embryologists of this epoch had to contend. Serial section cutting was unknown, the staining of thin layers and reconstruction were unheard of; even the hardening of the soft embryonic tissues was only just discovered, as is indicated by Haller above. Hertwig has excellently discussed the advances in embryological technique which took place during this and the following century. It is true that dyes were beginning to be used, as some instances already given demonstrate, and as is seen from the use of madder in the staining of bones, which began about this time, and was later much used by the Hunters. Heertodt's Crocologia is important in this connection. Heertodt, by injecting saffron into the maternal circulation, found it afterwards in the amniotic fluid, and his experiment was cited by Haller in support of that theory of the origin of the liquid. But the most important advance in technique was the progress in artificial incubation. The art, though lost throughout the Middle Ages and the seventeenth century, was now to be revived.

During this period much work was done on it. As far back as 1 600, de Serres had mentioned some experiments of this nature, but they were not successful. "The chicks", he said, "were usually born deformed, defective or having too many legs, wings, or heads, nature being inimitable by art." Birch, in his History of the Royal Society, also refers to it. "Sir Christopher Heydon [a relative of Digby's Sir John?] together with Drebell, long since in the Minories hatched several hundred eggs but it had this effect, that most of the chickens produced that way were lame and defective in some part or other." Antonelli states that similar trials were made at the court of the Grand-duke Ferdinand II at Florence about 1644, Thomas Bartholinus gives a like account with reference to the contemporary court of King Christian IV of Denmark, and Poggendorff and Antinori relate that the Accademia d. Cimento, inspired by Paolo del Buono, made trial of artificial incubation between 1651 and 1667.

But the most famous of all the attempts to make artificial as successful as natural incubation, were those of Reaumur, whose book De I' art defaire eclore les poulets of 1749 achieved a wide renown. He devotes many chapters to a detailed description of incubators of very


various kinds : but he nowhere gives any indication of his percentage hatch. It was probably low. He speaks also of the ^^funestes effets'^ of the vapours of the dung on the developing embryos, without, however, furnishing any foundation for an exact teratology. In the second volume he describes those experiments on the preservation of eggs by varnishing them, which caught the imagination of Maupertuis and were held up to an immortal but by no means deserved ridicule by Voltaire in his Akakia. For the details of this amusing but irrelevant issue see Miall and Lytton Strachey.

After Reaumur, there were numerous continuations of the kind of work which he had done, in particular by Thevenot, La Boulaye, Nelli, Porta and Cedernhielm. Much the most interesting of these was the work of Beguelin, who attempted to incubate eggs with part of the shell removed so as to form a round window. He was not, however, successful in the carrying out of this very modern idea. Probably the most peculiar investigation made in this field at this time was that of Achard, who is mentioned in a passage of Bonnet's. "Reaumur did not suspect in 1749", says Bonnet, "that someday one would try to substitute the action of the electric fluid for his borrowed heat. This beautiful invention was reserved for M. Achard of the Prussian Academy who excels as an experimentalist. He has not so far succeeded in actually hatching a chick by means of so new a process, but he has had one develop up to the eighth day, when an unfortunate accident deranged his electrical apparatus." Bonnet goes on to say that this substitution of electricity for heat gives him hope that by electrical means an artificial fertilisation will one day become possible.

The references to these experiments and to those of many minor investigators will be found in Haller. By the beginning of the nineteenth century a great mass of literature had developed on the subject, and it had become possible to hatch out eggs more or less successfully from furnaces, though the losses were still great. Early in the nineteenth century Bonnemain and Jouard referred to the large number of monsters produced, and in 1809 Paris wrote, "During the period that I was at College, the late Sir Busick Harwood, the ingenious Professor of Anatomy in the University of Cambridge, frequently attempted to develope eggs by the heat of his hotbed, but he only raised monsters, a result which he attributed to the unsteady application of the heat".



This is the most convenient place to mention theological embryology once again. Its place in the eighteenth century was small, and in the nineteenth, with the recognition that whatever the soul is, it is not a phenomenon, it altogether disappeared from serious general discussion. F. E. Cangiamilla's Embryologia Sacra, however, ran through several editions between 1700 and 1775. Cangiamilla {Panorm. Eccl. Can. Theol. et in toto Sicil. regno contra haereticam pravitatem Inquisitore provinciali) deals very frilly with the time of animation, quoting a host of writers such as St Gelasius, St Anselm, Hugh of St Victor and Pico della Mirandola. His mind retains a quite mediaeval conformation, as the following curious passage illustrates : '^ Quot non foetus abortivos ex ignorantia obstetricum et matrum excipit lafrina, quorum anima, si Baptismate non fraudaretur, Deum in aeternam videret, esset decentius tumulandum! " His instructions for the baptism of monsters are also very odd. But theological embryology probably reached its climax in the report of the Doctors of Divinity at the Sorbonne on March 30, 1733, in which intra-uterine baptism by means of a syringe was solemnly recommended. This is included in Deventer's book, and has been referred to by Sterne and Spencer. For other aspects of these tracts of thought see Nicholls and his anonymous antagonist. But Cangiamilla and his colleagues — Gerike, Kaltschmied, etc. — are only of decorative importance to our present theme, and for fuller information regarding them, reference must be made to the treatise of Witovski. It is interesting to note that as late as 1913, 182 days was fixed as "perfection-time", whatever that may be, by Moriani.

3*12. Ovism and Animalculism

We must now return to the beginning of the century in order to pick up the thread of the main trend of thought. By 1720 the theory of preformation was thoroughly established, not only on the erroneous grounds put forward by Malpighi and Swammerdam, but on the experiments of Andry, Hartsoeker, Dalenpatius and Gautier, who all asserted that they had seen exceedingly minute forms of men, with arms, heads, and legs complete, inside the spermatozoa under the microscope. Gautier went so far as to say that he had seen a microscopic horse in the semen of a horse (he gave a plate of it) and a similar animalcule with very large ears in the semen of a donkey; finally, he described minute cocks in the semen of a cock.


Haller remarks gently that he has searched for these phenomena in vain. Vallisneri asserted the same kind of thing about the mammahan ovum, though he admitted that, in spite of long searching, he had never seen one. Besides the main distinction between prefer mationists and epigenesists, then, there arose a division among the former group, so that the ovists regarded all embryos as being produced from smaller embryos in the unfertilised eggs, while the animalculists regarded all embryos as being produced from the smaller embryos provided by the male in his spermatozoa. The animalculists thus afforded a singular example of a return to the ancient theory mentioned by Aeschylus in the Oresteia (see p. 65). Their most conspicuous example was Nicholas Andry, who pictured each c^gg as being arranged like the Cavorite sphere in which H. G. Wells' explorers made their way to the moon, i.e. with one trap-door. The spermatozoa, like so many minute men, all tried to occupy an egg, but as there were far fewer eggs than spermatozoa, there were, when all was over, only a few happy animalcules who had been lucky enough to find an empty egg, climb in, and lock the door behind them.

The whole controversy was intimately bound up with the question of spontaneous generation, for, whatever the case might be in the higher animals, if it were true that the lower ones could arise de novo out of slime, mud, or meat infusion, for instance, then their parts at least must have been made by epigenesis, and not in any other way, for it could hardly be held that a homogeneous infusion had any structure of that kind. And if epigenesis could occur in the lower animals, then the thin end of the wedge had been driven in, and it might occur among the higher ones as well. It was in this way that the spontaneous generation controversy came to have a peculiar importance for embryology in the eighteenth century. Driesch has essayed to make the generalisation that all the supporters of epigenesis were vitalistic in their tendencies, while those who adhered to the preformation theory were not. But there are too many exceptions to this rule to make it of any use. In so far as there is truth in it.

Fig. 1 1 . Hartsoeker's drawing of a human spermatozoon.


it doubtless arose from the fact that, in epigenesis, a continual production of new organs and new relationships between organs already formed would seem to require an immanent formative force of some kind, such as the vis essentialis of Wolff; while, on the preformation hypothesis, where embryogeny was little more than a swelling up of parts already there, it could be explained as simply as nutrition. The failure of the "short-cut" mechanical philosophers such as Gassendi and Descartes thus led to preformationism just as much as to epigenesis. A remark of Cheyne's throws much light on this question, for in 17 15 he wrote, unconsciously following Gassendi's line of thought, "If animals and vegetables cannot be produced from matter and motion (and I have clearly proved that they cannot), they must of necessity have existed from all eternity". Preformationism was thus the only resource if the universal jurisdiction of the mechanical theory of the world was to be retained. Stahl and, later, Wolff, saw no point in retaining it, and carefully joined together what Descartes had, with equal care, put asunder.

The original discoveries of de Graaf and Stensen were extended by Tauvry in 1690 to the tortoise, and by Lorenzini in 1678 to the Torpedo', so that the eighteenth century began with an excellent basis for ovistic preformationism. The greatest names associated with this school were Swammerdam, Malpighi, Bonnet, v. Haller, Winslow, Vallisneri, Ruysch and Spallanzani. But there were many others, some of whom did valuable work, such as Bianchi, Sterre, Teichmeyer, Weygand, Perrault, Vercelloni, Vidussi, Bussiere, Fizes and Coschwitz. The treatises of Imbert and Plonquet were written from this point of view, as was the bright little dialogue of de Houpeville. J. B. du Hamel asserted that he could see the chick embryo in the Ggg before fertilisation, and Jacobaeus made a like affirmation in the case of the frog.

On the other side, that of animalculistic preformationism, the contestants were fewer. Their greatest names were Leeuwenhoek, Hartsoeker, Leibnitz and the cardinal de Pohgnac. In England the physicians Keil and Cheque supported this position, in France Geofroi and the obstetrician la Motte, in Germany Withof and Ludwig, and in Belgium Lieutaud. De Superville wrote in favour of it in the Philosophical Transactions of the Royal Society, and an anonymous Swedish work of some fame supported it. To the argument of Vallisneri that the existence of so many animalcules must be an


illusion, since Nature could hardly be so prodigal, the animalculists retorted by instancing such observations as that of Baster, who had taken the trouble to count the eggs of a crab and had found that they amounted to 12,444. James Cooke later elaborated a theory of a world of the unborn to which the spermatozoa could retire between each attempt to find a uterus in which they could develop — this avoided Vallisneri's argument. "All those other attending Animalcula, except that single one that is then conceived, evaporate away, and return back into the Atmosphere again, whence it is very likely they immediately proceeded; into the open Air, I say, the common Receptacle of all such disengaged minute sublunary bodies; and do there circulate about with other Semina, where, perhaps, they do not absolutely die, but live a latent life, in an insensible or dormant state, like Swallows in Winter, lying quite still like a stopped Watch when let down, till they are received afresh into some other Male body of the proper kind, to be again set on Motion, and ejected again in Coition as before, to run a fresh chance for a lucky Conception ; for it is very hard to conceive that Nature is so idly luxurious of Seeds thus only to destroy them, and to make Myriads of them subservient to but a single one." But Cooke's attractive hypothesis, published in 1762, came too late, as Punnett says, to save the animalculists.

On the experimental side. Garden and Bourguet came forward with descriptions of little men inside the animalcules, thus confirming the work of Gautier and Hartsoeker. It is fair to add, however, that Garden held quite enlightened views of the mutual necessity of egg and spermatozoon. La Motte maintained that the egg (which he identified with the Graafian follicle) was too big to go down the Fallopian tube, and Sbaragli, another writer on the animalculist side, agreed with him.

Leeuwenhoek, it must be admitted, indulged in assertions no less fantastic than those of his followers. He said there were spermatic animalcules of both sexes, as one could see by a slight difference near their tails, that they copulated, that the females became pregnant and gave birth to little animalcules, that young and feeble ones could be seen, that they shed their skins, and, finally, that some had been observed with two heads. Haller, who made good use, on the whole, of his strong vein of scepticism, characterised all these remarks as "only conjectures". (See Fig. 12.)

SECT. 3]



As for the supporters of epigenesis, they were few, but they included Descartes, de Maupertuis, Antoine Maitre-Jan and John Turberville Needham. Von Haller affords some evidence against the identification of epigenesis with vitaHsm and preformation with mechanism, for he says, "Various authors have taught that the parts of the human body are formed by a mechanism depending on general laws (i.e. laws not simply of biological jurisdiction) or by the virtue of some ferment, or by rest and cold making crusts out of the different juices, or in other ways. All these (mechanical) systems have some resemblance to that of M. Wolff". Haller also speaks always of Wolff's vis essentialis as "blind". Minor writers on the epigenetic side were Tauvry, Welsh, Dartiguelongue, Bouger, Drelincurtius and Mazin. After 1 750 C. F. Wolff brought an abiding victory to their opinion.

Some maintained a quite independent position, such as Buffon, who welded together an epigenetic theory

of fertilisation with a ^^S- 12. Dalenpatius' drawings of human spermatozoa.

preformationist theory of embryogeny. Pascal (not the great Jansenist) put forward the chemical view that fertilisation consisted of a combination between the acid semen of the male and the "lixivious " semen of the female, no doubt because in chemistry acids were regarded as male and alkalies female. Claude Perrault and Connor also suggested that the formation of the embryo was a fermentation set up in the egg by the spermatic animalcule. In this they were following the example of van Helmont, who had originally suggested such a theory. In 1 763 Jacobi discovered how to fertilise fish eggs with milt; a practical matter which had a good deal of influence on biological theory. Launai alone still held to the Aristotelian conception of form and matter.

There is no need here to do more than glance at the spontaneous generation controversy itself, for it has always been well known in the history of biology, especially in connection with the subsequent


work of Pasteur. J. T. Needham's books, New Microscopical Discoveries of 1745 and Observations upon the generation, composition, and decomposition of animal and vegetable substances of 1 750, exercised a considerable influence. They were written after the French fashion (Needham had been educated at Douai) very concisely, and with some brilliance of style, and it is hardly true to say, as Radl does, that their experimental foundation was meagre. That it was inadequate was proved definitely as events turned out by Spallanzani. De Kruif 's account of the controversy is false and misleading, especially in its estimate of Needham who is much more truly described in the words of Louis Pasteur (see also Prescott).

Needham's case rested upon the statement that, if meat broth was placed in a sealed vessel and heated to a high temperature so that all life was destroyed in it, it would yet be found to be swarming some days later with microscopical animals. All depended, therefore, upon the sureness with which the vessel had been sealed and the efficacy of the heat employed to kill all the animalcules initially present, and, in the ensuing controversy, Needham lost to Spallanzani entirely on a question of technique. It may be remarked here, without irrelevance, that the problem is still unsolved; for all that was proved by the experiments of Spallanzani was that animals the size of rotifers and protozoa do not originate spontaneously from broth, and all that was proved by those of Pasteur was that organisms the size of bacteria do not originate de novo in that way. The knowledge which we have acquired in recent years of filter-passing organisms, such as the mosaic disease of the tobacco-plant, and phenomena such as the bacteriophage of Twort and d'Herelle, has reopened the whole matter, so that of the region between, for example, the semi-living particles of the bacteriophage (lO"^^ gram) and the larger sized colloidal aggregates (io~^^ gram) we know absolutely nothing. The dogmatism with which the biologists of the early twentieth century asserted the statement omne vivum ex vivo was therefore, like most dogmatisms, ill-timed.

But to dwell further on this would be a digression. The important point was that Spallanzani's victory was a victory not only for those who disbelieved in spontaneous generation, but also for those who believed in the preformation theory of embryogeny. By 1 786, indeed, that viewpoint was so orthodox that Senebier, in his introduction to an edition of Spallanzani's book on the generation


of animals and plants, could treat the epigenesists as no better than atheists.

Spallanzani's views on embryology were largely drawn from his study of the development of the frog's egg. Here he went far beyond Bosius, but, in spite of many careful observations, he thought he saw the embryo already present in the unfertilised ova. This led him to claim that amphibia ought to be numbered among viviparous animals. His principal step forward was his recognition of the semen as the actual agent in fertilisation on definite experimental grounds — the narrative of his artificial insemination of a bitch is too famous to quote; he said it gave him more intellectual satisfaction than any other experiment he had ever done. This demonstration finally disposed of the aura seminalis which Harvey had found himself obliged to adopt on the grounds of his dissections on does. Curiously enough Spallanzani never convinced himself that the spermatozoa themselves were the active agents.

3-13. Preformation and Epigenesis

Of all the preformationists Charles Bonnet was the most theoretical. He was an adherent of that way of thinking mainly on the theoretical ground that the organs of the body were linked together in so intimate a manner that it was not possible to suppose there could ever be a moment when one or two of them were absent from the ranks. "One needs", he said, "no Morgagni, no Haller, no Albinus to see that all the constituent parts of the body are so directly, so variously, so manifoldly, intertwined as regards their functions, that their relationship is so tight and so indivisible, that they must have originated all together at one and the same time. The artery implies the vein, their operation implies the nerves which in their turn imply the brain and that by consequence the heart, and every single condition a whole row of other conditions." Bonnet compared epigenesis to crystal-growth in which particles are added to the original mass independently of the plan or scheme of the whole, i.e. in opposition to the growth of an organism, in which particles are added on only at certain places and certain times under the guidance of "forces de rapport". Przibram has recently discussed the question of how far such a comparison is admissible, but, in Bonnet's time at any rate, it became very famous. Bonnet made reference to Haller's discovery of the intimate relationship between embryo and yolk as evidence


for his theory. The embryo begins, according to him, as an exceedingly fine net on the surface of the yolk, fertilisation makes part of it beat and this becomes the heart, which, sending blood into all the vessels, expands the net. The net or web catches the food particles in its pores, and Bonnet supposed that, if it were possible to abstract all the food particles at one operation from the adult animal, it would shrivel and shrink up into the original invisible web from which it originated.

Bonnet was no more afraid of the emboitement principle than was Haller; indeed, he called it "one of the greatest triumphs of rational over sensual conviction". Many of his arguments were reproductions of Haller's, and he says in his preface that he had written his book some time before Haller's papers on the chick appeared, but then, finding his own views confirmed by the more experimentally founded ones of Haller, he determined to publish what he had set down. Thus in one place he says, " I shall be told, no doubt, that the observations on the development of the chick in the tgg and the doe in the maternal uterus make it appear that the parts of an organised body are formed one after another. In the chick for instance it has been observed that during the early part of incubation the heart seems to be outside the animal and has a very diflferent form to what it will have. But the feebleness of this objection is easy to apprehend. Some people wish to judge of the time when the parts of an organised body begin to exist by the time when they become visible to us. They do not reflect that minuteness and transparency alone can make these parts invisible to us although they really exist all the time".

Bonnet was therefore what might be called an " organicistic preformationist", for his objection to epigenesis lay in the fact that it apparently did not allow for the integration of the organism as a whole. His mistake was that he assumed the capacities of the adult organism to be present all through foetal life, whereas the truth is that they grow and differentiate in exactly the same way as the physical structure itself does. Bonnet's philosophical position, which has been analysed by Whitman, seriously contradicts the generalisation of Driesch that all the epigenesists were vitalists and all the preformationists mechanists. For Bonnet an epigenetic and a mechanical theory were one and the same; he hardly distinguished, as Radl says, between Descartes and Harvey; and it was just the neo-vitalistic


idea of the organism as a whole that he could not fit in with epigenesis. Needham and Wolff were undoubtedly epigenesist-vitalists, and Bonnet was undoubtedly a preformationist-vitalist, but Maupertuis was equally clearly an epigenesist-mechanist.

G. L. Leclerc, Comte de Buffon, the most independent figure in the controversy, stood alone as much because of his erroneous experiments as because of his originality of mind. As has so often been observed, Buffon was not really an experimentalist at all: he was a writer, and preferred other people to do his experiments for him. The volume on generation in his Histoire Naturelle begins with a very long historical account of the work that had been done in the previous centuries on embryology. At the beginning of the section on reproduction in general he said, "The first and most simple manner of reproduction is to assemble in one body an infinite number of similar organic bodies and to compose the substance in such a manner that every part shall contain a germ or embryo of the same species and which might become a whole of the same kind with that of which it constitutes a part". Such an idea resembles the ancient atomistic speculations, and is explicated by W. Smellie, the obstetrician, who translated Buffon into English, as follows: "The intelligent reader will perceive that this sentence, though not very obvious, contains the principle upon which the whole theory of generation adopted by the author is founded. It means no more than that the bodies of animals and of vegetables are composed of an infinite number of organic particles, perfectly similar, both in figure and substance, to the whole animal or plant of which they are the constituent parts ". This conception explains Buffon's curious attitude to the preformation question. An embryo was preformed in its germ because all the parts of the germ were each a model of the animal as a whole, but it was also formed by epigenesis because, the sexual organs being first formed, all the rest arose entirely by a succession of new origins. Buffon's "organic, living, particles" bear some resemblance to the "biogen molecules" which later generations were to discuss, and he says that an exactly similar but simpler structure is present in dead matter.

In his discussion of former theories he resolutely rejects the emboitement aspect of preformationism, giving various calculations to show its impossibility and maintaining that "every hypothesis which admits an infinite progression ought to be rejected not only as false


but as destitute of every vestige of probability. As both the vermicular and ovular systems suppose such a progression, they should be excluded for ever from philosophy". He completely destroys the theory which the ovists and animalculists had set up in order to explain resemblance to parents, namely, that, although the foetus might originate either from egg or spermatic animalcule originally, it was moulded into the form of its parents by the influence of the maternal organism during pregnancy. This field, which was more than once disturbed by the contestants during the course of the century, received systematic attention from time to time by medical writers. There was a memorable dispute on this point between Turner and Blondel, whose polemics, written in an exceedingly witty manner, are still very pleasant and amusing to read. Blondel was the sceptic and Turner the defender of the numerous extraordinary stories which passed for evidence on this subject. It is interesting to note that Turner believed in the continuity of foetal and maternal blood-vessels. Krause and Ens later supported the opinions of Turner, while Okes, in a Cambridge disputation, argued against them.

Buflfon's sixth chapter, in which he relates the progress of his own experiments, is unfortunate, in that his main result was to discover spermatozoa in the liquor folliculi of ovaries of female animals. The explanation of how he came to make such an enormous mistake has never been satisfactorily given, and it was not long before the truth of the observation was questioned by Ledermuller. It led him naturally to the assertion that the ovaries of mammalia were not eggproducing organs but animalcule-producing organs, and to the view that the beginning of embryonic development lay in the fusion of the male with the female spermatic animalcules — a curious revival of Epicureanism. But it is to be observed that he does not mean one male animalcule with one female animalcule, but rather all with all, in a kind of pangenesis. "All the organic particles", he says, "which were detached from the head of the animal will arrange themselves in a similar order in the head of the foetus. Those which proceeded from the backbone will dispose themselves in an order corresponding to the structure and position of the vertebrae". And so on for all the organs. The fact that for the organs common to both sexes a double set of animalcules will thus be provided does not give Buffon any difficulty and is fully admitted by him. Accordingly he could only agree to the aphorism omne vivum ex ovo in the sense of


Harvey, namely, as referring to the egg-shaped chorion of vivipara, and definitely not in the sense of de Graaf and Stensen, namely, in the modern sense. "Eggs", he says, "instead of being common to all females, are only instruments employed by Nature for supplying the place of uteri in those animals which are deprived of this organ. Instead of being active and essential to the first impregnation, eggs are only passive and accidental parts, destined for the nourishment of the foetus already formed in a particular part of this matrix by the mixture of the male and female semen." Biology at this period was still labouring under the disadvantage of being without the celltheory, and therefore unable to distinguish between an egg and an egg-cell.

In spite of his leanings towards epigenesis, Buffon repeats precisely the error of Malpighi. "I formerly detected", he says, "the errors of those who maintained that the heart or the blood was first formed. The whole is formed at the same time. We learn from actual observation that the chicken exists in the egg before incubation. The head, the backbone, and even the appendages which form the placenta are all distinguishable. I have opened a great number of eggs both before and after incubation and I am convinced from the evidence of my own eyes that the whole chicken exists in the middle of the cicatrice the moment the egg issues from the body of the hen. The heat communicated to it by incubation expands the parts only. But we have never been able to determine with certainty what parts of the foetus are first fixed, at the moment of its formation." The experiment of taking a look at the cicatrices of eggs on their way down the parental oviduct is so obvious that Buffon must have thought of it, and it would be really interesting to know what factor in the intellectual climate it was that made him regard such an observation as not worth attempting. His observations on the embryo itself were good and, in some ways, new; thus he noticed that the blood first appears on the "placenta" or blastoderm, and for the first few days seems hardly to enter the body of the embryo. He gave an extremely good account of the whole developmental process in the chick and in man, and his opinions on the use of the amniotic liquid and the functions of the umbilical cord were very advanced.

J. T. Needham, however, spoke very clearly in favour of epigenesis, though he himself did no embryological experiments. His Idee sommaire of 1776, written against Voltaire, who had called him

N E I 14


a Jesuit and who had drawn materialistic inferences from his writings, contained the following passage: "The numerous absurdities which exist in the opinion ofpre-existent germs together with the impossibility of explaining on that ground the birth of monsters and hybrids, made me embrace the ancient system of epigenesis, which is that of Aristotle, Hippocrates, and all the ancient philosophers, as well as of Bacon and a great number of savants among the neoteriques. My observations also led me directly to the same result". Needham's embryology is mostly contained in his Observations nouvelles sur la Generation of 1750. He was explicitly a Leibnitzian and postulated a vegetative force in every monad.

Needham was not the only thoroughgoing epigenesist of this period. Maupertuis, whose Venus Physique was published anonymously in 1746, came out very clearly on the side of epigenesis. "I know too well", he said, "the faults of all the systems which I have been describing, to adopt any one of them, and I find too much obscurity in the whole matter to wish to form one of my own. I have but a few vague thoughts which I propose rather as thoughts to be examined than as opinions to be received, and I shall neither be surprised nor think myself aggrieved if they are rejected. It seems to me that both the system of eggs and that of spermatic animalcules are incompatible with the manner in which Harvey actually saw the embryo to be formed. And one or the other of these systems seems to me still more surely destroyed by the resemblance of the child, now to the father and now to the mother, and by hybrid animals which are born from two different species. ... In this obscurity in which we find ourselves on the manner in which the foetus is formed from the mixture of two liquors, we find certain facts which are perhaps a better analogy than what happens in the brain. When one mixes silver and spirits of nitre with mercury and water, the particles of these substances come together themselves to form a vegetation so like a tree that it has been impossible to refuse it the name." This was the Arbor Dianae, which played a great part in these embryological controversies of the eighteenth century. It has a great interest for us, for it was perhaps the first occasion on which a non-living phenomenon had been appealed to as an illustration of what went on in the living body. It is true that Descartes long before had said that the movements of the living body were carried out by mechanisms like clocks or watches, and that they resembled


the statues in certain gardens which could be made to perform unexpected functions by the pressure of a manipulator's foot on a pedal, but these instances were all artificially constructed mechanical devices, whereas the Arbor Dianae was a natural phenomenon quite unexplained by the chemists of the time, and the lineal forerunner of Lillie's artificial nerve, and Rhumbler's drop of chloroform. We know now that its formation is a simpler process than anything which occurs in the developing embryo, but the course of research has made it undeniably clear that the same forces which operate in the formation of the Arbor Dianae are at work also in the developing embryo. To this extent Maupertuis is abundantly justified, and Driesch's comments on him are not in agreement with the facts.

"Doubtless many other productions of a like kind will be found", Maupertuis goes on, "if they are looked for or perhaps if they are looked for less. And although they seem to be less organised than the body of most animals, may they not depend on the same mechanics and on similar laws? Will the ordinary laws of motion suffice, or must we have recourse to new forces? These forces, incomprehensible as they are, appear to have penetrated even into the Academy of Sciences at Paris, that institution where so many opinions are weighed and so few admitted." Maupertuis goes on to speak of the contemporary deliberations on the subject of attraction. "Ghymistry", he says, "has felt the necessity of adopting this conception and attractive force is nowadays admitted by the most famous chymists who have carried the use of it far beyond the point which the astronomers had reached. If this force exists in nature, why should it not take part in the formation of animals?" Maupertuis was thus an epigenesist and a mechanist at the same time. His opinions have an extremely modern ring, and his only retrograde step was in suggesting that the spermatic animals had nothing else to do except to mix the two seeds by swimming about in them. But that legacy of ovism was common all through the eighteenth century, and thirty years later Alexander Hamilton could say, "From the discovery of Animalcula in semine masculino by Leeuwenhock's Glasses, a new Theory was adopted which is not yet entirely exploded".

But the real middle point and fulcrum of the whole period lay in the controversy between von Haller and Caspar Friedrich Wolflf, the former at Gottingen and the latter at St Petersburg in the Academy of the Empress Catherine. Kirchhoflf has described this polemic.



Wolff's Theoria generationis, which was a defence of epigenesis on theoretical and philosophical grounds, written in a very formal, logical, and unreadable manner, appeared when he was only twentysix years old, in 1759. Leibnitz, as Radl points out, had borrowed from the earlier preformationists the conception of a unit increasing in bulk in order to become another kind of unit; but Wolff, following Needham, borrowed from Leibnitz the idea of a monad developing into an organism by means of its own inherent force, and to this he joined the Stahlian notion of a generative supra-physical force in nature. On the practical side, Wolff's work was indeed of the highest importance. If the embryo pre-exists, he argued, if all the organs are actually present at the very earliest stages and only invisible to us even with the highest powers of our microscopes, then we ought to see them fully formed, as soon as we see them at all. In other words, at the moment at which any given organ comes into view, it ought to have the form and shape, though not the size, of the same organ when fully completed in the embryo at birth. On the other hand, if this is not the way in which development goes on, then one ought to be able to see with the microscope one shape changing into another shape, and, in fact, a series of appearances, each one different from that which had immediately preceded it, or, in other words, a series of advancing adaptations of the various parts of the primitive embryonic mass. WoliT chose as his first test case the blood-vessels of the blastoderm in the chick, for he saw that at one moment this apparatus was in existence, while the moment before it had not been. His microscopical researches led him to the conclusion that the homogeneous surface of the blastoderm partially liquefies and transforms itself at these points into a mass of islands of solid matter, separated by empty spaces filled with a colourless liquid but afterwards with a red liquid, the blood. Finally, these spaces are covered with membranes and become vessels. Consequently it was obvious that the vessels had not been previously formed, but had arisen by epigenesis.

Haller replied to this new experimental foundation for epigenesis without delay, for he was working on the development of the chick at the same time, and held closely to the opposite theory. We have already seen what his one and only argument against Wolff was. He used it time after time in all its possible variations, maintaining stoutly that the chick embryo was so fluid in the early stages that Wolff had no right to deny the presence of a given structure simply because


he could not see it. Haller's explanation of Wolff's results was that the blood-vessels had been there all the time but that they had not become visible until the moment at which Wolff saw the islands forming. "After I had written the above", said Haller, "M. Wolff made new objections against the demonstration. Instructed by new researches, he denies absolutely that the yolk-membranes, which he makes two in number, exist before incubation. He pretends that they are new and that they are born at the beginning of incubation, and consequently that the continuity of their vessels with the embryo does not in the least prove that in the body of the mother the yolk received vessels from the foetus. I have compared the observations of this great man with my own and I have found that the yolk never has more than one pulpy and soft membrane, part of which is what I have called the umbilical area, and that the fine exterior membrane does not belong to the yolk but to the inner part of the umbilical membrane. ... I do not believe that any new vessels arise at all, but that the blood which enters them makes them more obvious because of the colour which it gives them, and so by the augmentation of their volume, they become longer."

Wolff replied by another extensive piece of work, which he called De Formatione Intestinorum, and which appeared in one of the publications of the Russian Academy for 1768. It ruined preformationism. In it he demonstrated that the intestine is formed in the chick by the folding back of a sheet of tissue which is detached from the ventral surface of the embryo, and that the folds produce a gutter which in course of time transforms itself into a closed tube. The intestine, therefore, could not possibly be said to be preformed, and from this as starting-point, Wolff went on to propose an epigenetic theory which applied the same process to all organs. It is interesting to note that the facts brought forward by Wolff have never been contradicted, but have been used as a foundation to which numberless morphological embryologists have added facts discovered by themselves. It is noteworthy that, although Wolff's second general principle, that of increasing solidification during embryonic development, led to no immediate results, it has been abundantiy confirmed since then (see Fig. 221). His observations on the derivation of the parts of the early embryo from "leaf-like" layers were even more important, and acteA as a very potent influence in the work of Pander and von Baer.

It happened, however, that Haller had much the greater in


fluence in the biological world at the time, so that Wolff's conceptions did not immediately yield fruit in any general advance. Looking back over the second half of the seventeenth and the first two-thirds of the eighteenth century, it is remarkable how little theoretical progress was made in view of the abundance of new facts which were discovered. Punnett, in an interesting paper, has vividly brought this out. "The controversy between the Ovists and Animalculists had lasted just a century", he says, "and it is not uninteresting to reflect that the general attitude of science towards the problem of generation was in 1775 niuch what it had been in 1675. When the period opened, almost all students of biology and medicine were Preformationists and Ovists; at its close they were for the most part Ovists and Preformationists." Ovism sprang in the first instance from de Graaf's discovery of the mammalian egg, which gave a new and precise meaning to Harvey's aphorism. Preformationism, already old as a theory, acquired an apparent factual basis in the work of Malpighi and Swammerdam, and allied itself naturally with ovism. With Leeuwenhoek and his spermatozoa, animalculism came upon the field. The main outlines of the battle which went on between the two viewpoints have already been drawn, but it is worth remembering that there were independent minds who were impressed by the obvious facts of heredity and found it difficult to call one sex essential rather than the other. Among these Needham and Maupertuis might be counted, and among the lesser men, James Handley with his Mechanical Essays on the Animal Oeconomy of 1730 ought to receive a mention. Though fond of theological arguments he upheld the common-sense attitude against ovists and animalculists alike — "We dissent in some things", he said, "both from Leeuwenhoeck and Harvey. . . . Both the semen and ova (notwithstanding all that can be said) we believe to be a causa sine qua non in every Generation". But what finally killed animalculism was the discovery in so many places of small motile living beings, flagellates, protozoa, large vibrios. It was difficult to maintain in the face of this new evidence that the spermatozoa were essential elements in generation, though the seminal fluid itself might very well be, as of course was Spallanzani's opinion. The preformation theory was what was holding up further progress, and when Wolff's arguments prevailed in the very last years of the eighteenth century, the way was open for the recognition of the true value of the spermatozoa.


The otherwise unknown physician d'Aumont, who wrote the article on "Generation" in Diderot's famous Encyclopaedia, brought this out in an interesting way, for himself an ovist, he summarised the arguments, which, in 1757, were destroying the animalculist position, and reducing rapidly the number of its adherents.

1. Nature would never be so prolific as to produce such millions of spermatic animalcules, each one with its soul, unnecessarily.

2. The spermatic animalcules of all animals are the same size, no matter how large the animal is: how, therefore, can they be involved in its generation?

3. They are never found in the uterus after coitus, but only in the sperm (?).

4. How do they reproduce their kind?

5. What evidence is there that they are any different from the animalcules (of similar shape, etc.) which are to be found in hay infusion, scrapings from the teeth, etc. ? Nobody supposes that these have any relation to reproduction.

3-14. The Close of the Eighteenth Century

The last forty years of the century were not marked by any great movement in a fruitful direction for morphological embryology, an iconographic wave of some merit due to Albinus, W. Hunter, Tarin, Senffj Rosenmuller, Danz and Soemmering excepted ; and it was not until 181 2 that J. F. Meckel the younger translated Wolff's papers into German. This was one of the principal influences upon Pander and von Baer. In his introduction, Meckel describes how Wolff's work had been disregarded, and points out that Oken, writing in 1806, had apparently never even heard of it. In the very early years of the nineteenth century morphological embryology received a great impetus, however. One of the most interesting figures of the new period was de Lezerec, a Breton, whose father had been in the Russian naval service. The son, as a Russian naval cadet, no doubt stimulated by the writings of Wolff, who had lived at St Petersburg, used to incubate eggs on board ship. He eventually left the sea, studied medicine at Jena, and wrote an excellent dissertation on the embryology of the chick in 1 808, which Stieda has recently brought to light. He then went to Paris, and, taking a medical appointment at Guadeloupe, was lost to science. Very much more important was the work of Pander in 181 7 and von Baer in 1828,


but it belongs to the present period, and I shall not treat it historically. For data on von Baer, see Kirste, Addison and Stieda. It is interesting to note, however, that the recapitulation theory, which was first clearly formulated by von Baer, was already taking shape in various minds during the closing years of the eighteenth century. Lewes has thus described the thesis of Goethe's Morphologie, written in 1795: "The more imperfect a being is the more do its individual parts resemble each other and the more do these parts resemble the whole. The more perfect a being is the more dissimilar are its parts. In the former case the parts are more or less a repetition of the whole, in the latter case they are totally unlike the whole. The more the parts resemble each other the less subordination is there of one to the other : and subordination is the mark of high grade of organisation".

William and John Hunter belong also to the end of the century. The former, in his book on the anatomy of the gravid uterus, proved finally and completely the truth of the view that the maternal and foetal circulations are distinct. His injections left no shadow of doubt about the matter, and the way was clearly opened up for the study of the properties of the capillary endothelial membranes separating the bloods, a study which is still vigorously proceeding, especially in its physico-chemical aspect (see Section 21). There was a quarrel between the brothers over the priority of this demonstration. John Hunter's Essays and Observations also contain material important for embryology. His drawings of the chick in the &gg were very beautiful, and are still in the archives of the Royal College of Surgeons. He adopted Mayow's theory of the office of the air-space, and anticipated von Baer's theory of recapitulation much as did Goethe. "If we were capable of following the progress of increase of the number of parts of the most perfect animal as they were first formed in succession, from the very first to its state of full perfection, we should probably be able to compare it with some one of the incomplete animals themselves, of every order of animals in the creation, being at no stage different from some of the inferior orders. Or, in other words, if we were to take a series of animals, from the more imperfect to the perfect, we should probably find an imperfect animal corresponding with some stage of the most perfect." It is impossible not to reflect on the curious course which was taken by the essence of the idea of recapitulation in the history of embryology. As Aristotle first formulated it, it was as much bodily as mental, but all his sue


cessors until the eighteenth century a.d. treated it as a psychological rather than a physiological or morphological theory, and lost themselves in speculations about the vegetative, sensitive, and rational souls. Yet the other aspect of the theory was only asleep, and was destined to be of the greatest value as soon as investigators began to direct their attention more to the material than to the spiritual aspect of the developing being.

Hunter did not absolutely reject preformationism, but regarded it as holding good for some species in the animal kingdom ; he therefore attached no philosophical importance to it.

Although Wolff's work did not lead to the immediate morphological advances which might have been expected, it was in many ways fruitful. It produced J. F. Blumenbach's Uber den Bildungstrieb of 1789, a work which elaborated the Wolffian vis essentialis into the nisus formativus, a directing morphogenetic force peculiar to living bodies. It is interesting to note that Blumenbach passed through an exactly opposite succession of opinions to that of Haller, i.e. he was first attracted by preformationism, but, being convinced by Wolff's work, abandoned it in favour of epigenesis. Blumenbach compares his nisus formativus with the force of gravity, regarding them as exactly similar conceptions and using them simply as definitions of a force whose constant effects are recognised in everyday experience. Blumenbach says that his nisus formativus differs from Wolff's vis essentialis because it actively does the shaping and does not merely add suitable material from time to time to a heap of material which is already engaged in shaping itself. Wolff was still alive at this time, but he did not make any comment on Blumenbach, though he might very well have said that Blumenbach had misunderstood him, and that their forces were really alike in every particular. Both Blumenbach and Wolff were mentioned by Kant in the Critique of Judgement where he adopted the epigenetic theory in his discussion of embryogeny.

A word must be said at this point about the opinions of the eighteenth century on foetal nutrition. At the beginning of it, there was, as has been shown, a welter of conflicting theories; and though, later on, writers on this subject were fewer, the progress made was no more rapid. In 1802 Lobstein was supporting the view (which had been defended by Boerhaave) that the amniotic liquid nourished the embryo per os, although Themel had shown forty years before






that this could be at most the very slightest source of material, from a study of acephalic monsters. These workers had obviously learnt nothing from Herissant and Brady, who had been over precisely the same ground fifty years before. On the other hand. Goods and Osiander reported the birth of embryos without umbilical cords, so that the solution of this question became, in the first year of the nineteenth century, balanced, as it were, between the relative credibility of two kinds of prodigy. Nourishment per os was defended by Kessel, Hannes and Grambs, and was attacked by Vogel, Bernhard, Glaser, Hannhard and Reichard. The idea lingered on right into the modern period, and as late as 1 886 von Ott, who was much puzzled about placental permeability, decided that a great part in foetal nutrition must be played by the amniotic liquid. WeidHch, a student of his, fed a calf on amniotic liquid for some days, and as it seemed to get on all right, he reported the amniotic liquid to have nutritive properties. The appeal to monsters was still resorted to at the end of the nineteenth century, for Opitz, in order to negative von Ott's conclusions, drew attention to a specimen in the Chemnitz Polyklinik in which the oesophagus of a well-nourished normal infant was closed at the upper third without the development of the body having been in any way restricted. The fuller possibilities of biochemistry itself have sometimes been exploited in favour of the ancient theory of nourishment />^r os\ thus Kottnitz in 1889 collected some data about the presence of peptones and protein in the human amniotic liquid with this object in view. That the foetus swallows the liquid which surrounds it towards the end of gestation in all amniota, can hardly be disputed, and as there are known to be active proteolytic enzymes in the intestinal tract, no doubt some of the protein which it contains is digested — but to maintain that any significant part is played in foetal nutrition by this process has become steadily more and more impossible since 1600.

But to return to the eighteenth century; all was not repetition; occasionally somebody brought forward a few facts. Thus the deglutition of the amniotic Hquid was discussed by Flemyng in 1 755 in a paper under the title " Some observations proving that the foetus is in part nourished by the amniotic liquor". "I believe", he said, "that very few, if any at all, will maintain now-a-days with Claudius de la Courvee and Stalpartvan-der-Wiel, that the whole of its nourishment is conveyed by the mouth." But he himself had found white


hairs in the meconium of a calf embryo with a white hide. Both Aides and Swammerdam had found the same thing, but Aides did not think it of any significance, and Swammerdam merely remarked that the calf must lick itself in utero.

More interesting was W. Watson's "Some accounts of the foetus in utero being differently affected by the Small Pox". This was the earliest investigation of the permeability of the placenta to pathological agents. "That the foetus", said Watson, "does not always partake of the Infection from its Mother, or the Mother from the Foetus, is the subject of this paper." Two of his cases, he said, "evince that the Child before its Birth, though closely defended from the external Air, and enveloped by Fluids and Membranes of its own, is not secure from the variolous Infection, though its Mother has had the Distemper before. They demonstrate also the very great Subtility of the variolous Effluvia". But other cases "are the very reverse of the former, where though from Inoculation the most minute portion of Lint moisten'd with the variolous Matter and applied to the slightly wounded Skin, is generally sufficient to propagate this Distemper; yet here we see the whole Mass of the Mother's Blood, circulating during the Distemper through the Child, was not sufficient to produce it. . . . From these Histories it appears that the Child before its Birth ought to be consider'd as a separate, distinct Organization; and that though wholly nourish'd by the Mother's Fluids, with regard to the Small Pox, it is liable to be affected in a very different Manner and at a very different Time from its Mother". Doubtless the modern explanation of Watson's discordant results would be that! in one case there were placental lesions, destroying the perfect barrier between the circulations, and in others there were not.

In the last year of the century (but the seventh of the Republic) Citizens Leveille & Parmentier contributed an interesting paper to I the Journal de Physique in which they observed the increase in size] of the avian yolk on incubation and spoke of a current of water yolkwards (see Fig. 225).

3* 15. The Beginning of the Nineteenth Century

At the beginning of the new century a fresh influence came in! with the work of Lamarck, though it did not have such a great effect on his contemporaries as on later generations. Its relations with] biochemistry are so remote that there is no need to deal in any detail


with it here, but Lamarck's opinions on embryology may perhaps be given in the words of Cuvier, written in 1836.

"In 1802 he pubUshed his researches on living bodies, containing a physiology peculiar to himself, in the same way that his researches on the principal facts of physics contained a chemistry of that character. In his opinion the egg contains nothing prepared for life before being fecundated, and the embryo of the chick becomes susceptible of vital motion only by the action of the seminal vapour; but, if we admit that there exists in the universe a fluid analogous to this vapour, and capable of acting upon matter placed in favourable circumstances, as in the case of the embryon, which it organises and fits for the enjoyment of life, we will then be able to form an idea of spontaneous generations. Heat alone is perhaps the agent employed by nature to produce these incipient organizations, or it may act in concert with electricity. M. de Lamarck did not believe that a bird, a horse, nor even an insect, could directly form themselves in this manner; but, in regard to the most simple living bodies, such as occupy the extremity of the scale in the different kingdoms, he perceived no difficulty; for a monad or a polypus are, in his opinion, a thousand times more easily formed than the embryo of a chick. But how do beings of a more complicated structure, such as spontaneous generation could never produce, derive their existence? Nothing, according to him, is more easy to be conceived. If the orgasm, excited by this organizing fluid, be prolonged, it will augment the consistency of the containing parts, and render them susceptible of reacting on the moving fluids which they contain, and an irritability will be produced, which will consequently be possessed of feeling. The first efforts of a being thus beginning to develope itself must tend to procure it the means of subsistence and to form for itself a nutritive organ. Hence the existence of an alimentary canal. Other wants and desires, produced by circumstances, will lead to other efforts, which will produce other organs : for, according to a hypothesis inseparable from the rest, it is not the organs, that is to say, the nature and form of the parts, which give rise to habits and faculties ; but it is the latter which in process of time give birth to the organs. It is the desire and the attempt to swim that produces membranes in the feet of aquatic birds; wading in the water, and at the same time the desire to avoid getting wet, has lengthened the legs of such as frequent the sides of rivers; and it is the desire of flying


that has converted the arms of all birds into wings, and their hairs and scales into feathers. In advancing these illustrations, we have used the words of our author, that we may not be suspected either of adding to his sentiments or detracting any thing from them."

If the latter part of the eighteenth century did not produce the move forward in the morphological direction which might have been expected from the work of Wolff, a remarkable amount of work was accomplished on the chemical side. This mass of work did not spring from any one source, it was not due to a great discovery on the part of one man, but rather it came about that, as the technique of chemistry itself improved, a number of otherwise undistinguished investigators, such as Dehne, Macquer and Bostock, applied physicochemical methods to the embryo, though it is true that among the names are those of certain great chemists, such as Scheele and Fourcroy. The results of this movement were summarised in the work of J. F.John, whose Chemische Tabellen des Tierreichs appeared in 1814. With this date I propose to bring my historical assessment to an end. The work that was done in physico-chemical embryology after 181 4 will be considered in the appropriate sections dealing with the problems of the present time; for Gobley, as an example, who gave the name to the substance still called vitellin, was working only a dozen years after the date of the publication of John's Tabellen.

In this translation of the Tables, I have made one alteration only. John groups together a number of data which are contained in von Haller's Elementa Physiologiae, and attributes them to that great man. But actually they were obtained by earlier investigators and only came to John through the medium of Haller and Fourcroy — I have therefore allotted them to their true originators.


Substance or liquid

investigated Composition

Amniotic liquid (man) It contains a substance which can be precipitated with tincture of gall, phosphate of lime and muriatic salts „ It is salt

„ It is sweet

,, It coagulates on boiling













Barbati &



SECT. 3]




Substance or liquid investigated

Amniotic liquid (man)

Cheesy material, given off into the amniotic liquid by the body of the foetus (man)

Embryonic tissue-juice (man)

Amniotic liquid (cow)

Amniotic and allantoic liquids (cow)


It is miscible with water It is coagulable by tincture of gall It is coagulable by alcohol It is coagulable by alumina It is coagulable by spirits of nitre Free mineral alkali, water, albuminous substance, common salt Much water, very little common salt, fire-stable alkali, phosphoric acid, some earth, and oxyde of iron

Much water, a lymphatic coagulum, common salt, salmiac, a trace of phosphate of lime Sp. g. 1-005. Albuminous matter, soda, muriate of soda, phosphate of lime, the rest is water

Animal slime, and a characteristic fatty material, or rather an albuminous material tending to fat, carbonate of lime

It contains hydrofluoric acid

Water, much sulphate of soda, phosphate of lime and talc, an animal substance soluble in water, insoluble in spirits of wine, and not forming a combination with tannic acid, a crystalline amniotic acid

The liquid of the allantois is very different quantitatively in the different periods of pregnancy, as also in the qualitative aspect of its composition. First it is crystalline and colourless, then it gets yellowish, and finally a dark reddish-brown. But it remains watery all the time and never has the property possessed by the amniotic liquid, of becoming at last quite slimy even to the point of showing fibres in it. During the last months the hippomanes appear in it, these are soft and yet tough. The quantity of this liquid is much greater at the end than at the beginning. Alcohol precipitates from it a very large amount of a reddish substance; sulphate of baryta, tartaric acid, and carbonate of lime give a large precipitate. These reagents do not change the amniotic fluid at all. 1000 gm. Uq. allant. gave 20-25 gm. solid residue, 1000 gm. liq. amnii gave lo-i i gm. solid residue













Gmelin &



van den Bosch



Vauquelin & Buniva

Vauquelin & Buniva


Buniva & Vauquelin







Substance or liquid investigated

Blood of embryo (man)

Blood of embryo (rabbit)

Foetal urine (man) Meconium (man)

Meconium (cow)

Eggs (wild birds) Air-space


Shell-membranes Egg-white


Shell-membranes Shell

Composition Investigator

Soda, much serum, and some leathery Fourcroy

fibrous threads, which made up

only ^ grain out of 3 gros 6 grains

of cruor . They were jelly-like in consistency. No phosphoric acid. It

differed from the blood of an adult

(i) in not giving a red flush when

shaken up with air, (2) in not clotting in air, (3) in the fibres being

more jelly-like Does not coagulate in the cold but Fourcroy

gives rise to a red serum tending

towards brown. It was not as solid

as usual except when heated, then

it went grey though the supernatant

liquor was red It is odourless and colourless and of Fourcroy

a slimy nature Water f , ^r> spirituous extract similar Bay en

to gall, a black residue dissolving

partially in water to give a yellow

colour. He holds it to be a milky

excrement Contains true gall-like substances

Date 1790


Does not contain air of different composition from atmospheric air

Phosphate of lime, animal glue, and some combustible substance which escapes with a sulphurous smell from shells when they are softened in acid. Ferrous particles. Sometimes some common salt. An egg, which weighed 2 ozs. 2 scruples 15 grains, had white which weighed 10 qentchen 2 scruples, yolk ^ oz. \ scruple, and shell and membranes 2 drachms 5 grains

An animal material insoluble in acids

6 qentcfwn 2 scruples 7 grains lost practically 6 qentchen in drying, it contains no caustic salts, the ash is an earthy insipid dust

Albuminous matter, water, muriate of soda, phosphate of lime, and sulphur

Albuminous matter, oil, yellow pig- Adet ment

From 60 eggs, 5^ ozs. oil Dehne

Albuminous matter with much Adet oxygen

Carbonate of lime, phosphate of lime, Adet and very oxydised albuminous matter

Buniva & — Vauquelin

Hehl 1 796

von Wasserberg 1 780

von Wasserberg 1 780 von Wasserberg 1 780


SECT. 3]



EXCERPTS FROM J. F. JOHN'S CHEMISCHE TABELLEN OF 1814 {cent.) Substance or liquid


investigated Eggs (domestic hen) Shell

Investigator Date


A fine earth and a gelatinous material True lime, containing perhaps phosphoric acid \ oz. of pulverised clean shell, digested with spirits of wine, gave i \ grains of an extract which smelt and tasted rancid. The same amount of shell gave i scruple of a yellow watery extract which tasted salt Carbonate and phosphate of lime, traces of a jelly, which can be used as gum. Phosphoric acid can be had from the ash Carbonate and phosphate of lime, bitter earth and iron, a jelly which can be used as gum Carbonate of Hme 72 parts, phosphate of lime 2, jelly 3, water and loss 23 Carbonate of lime 89-6 parts, phosphate of lime 5-7, animal substance 4-7, traces of sulphur. As a hen lays 130 eggs in six months and as an egg weighs on an average 58-117 grams, 7486-226 grams of solid must be used for egg-production in that time, i.e. since the shells would weigh 64-685 gm., 7333-793 gm., 14 pounds 15 ounces 7 gros 8 grains. The secretion of the lime is probably accomplished by means of the kidneys Carbonate and phosphate of lime, and jelly Very much carbonate of lime, very little phosphate. Traces of phosphate of iron, earthy carbonates, rnuriates, albuminous and gelatinous substance to hold it together. I cannot find any uric acid in it, as Vauquelin says is there, nor is he right in saying that the sulphur is in the shell — it is in the membranes only, and under the form of sulphuric acid

Consist of an animal material Have the properties of the fibrous

part of blood A jelly-like material, soluble in hot

water An animal substance with traces of

phosphate of lime, carbonate of

lime, muriates, and a sulphurous

body An albuminous substance containing

traces of sulphur and soluble in

caustic potash

Macquer Leonhardi




1 781



Merat-Gaillot —

Vauquelin 1 799

Fourcroy John


Macquer Jordan





Vauquelin —




Substance or liquid investigated





Egg (Snipe, Tringa vanellus) Shell


An agglutinative substance insoluble in water, apparently like dried tragacanth gum

A white lymphatic transparent sticky slimy material

Soda, albuminous matter, water, sulphur

Water, albuminous matter, with some free alkali, phosphate of lime, muriate of soda, and sulphur

Contains benzoic acid

Water 80 parts, uncoagulable substance 4-5 parts, albuminous matter 15-5 parts, traces of soda, sulphuretted hydrogen gas, and benzoic acid

Contains sulphur

Water, albuminous matter, a little jelly, soda, sulphate of soda, muriate of soda, phosphate of lime, oxyde of iron (?)

An oxydised albuminous substance Apparently an albuminous substance

Consists of a lymphatic material and

a fatty oil Water, oil, albuminous matter, jelly Water, oil, albuminous matter, jelly, phosphates of lime and soda, with other salts Water, oil, albuminous matter Water, a mild oil, albuminous matter, a colouring matter which is perhaps iron Water, a yellow mild oil, traces of free (phosphoric?) acid, a small amount of a reddish-brown material, not fatty, and soluble in ether and warm alcohol, a jelly-like substance, a great deal of a modified albuminous substance, and sulphur

Egg (lizard, Lacerta viridis) Yolk


Egg (fish, salmon)

Is composed of the same constituents as that of the hen, but the dark green pigment and the dark brown splashes are probably oxyde of iron

A yellow oil, an albuminous material, and salts

Diflfers from that of fowls in being granular and greasy when hardened by boiling

420 grains contained of pure dry albuminous matter 26 grains, of a viscous oil 18 grains, insoluble albuminous matter 102 grains, mu

nvestigator John





Proust Bostock

Scheele John

Fourcroy —

John —

Macquer 1781

Thomson —

Hatchett —

Jordan —

Fourcroy —




John John




Substance or liquid

investigated Composition Investigator Date

riate of soda and sulphuric alkali 28 grains, jelly, phosphate of lime, and oxyde of iron 2 grains, water 242 grains

Egg (fish, Cyprinus barbiis) Contains a substance dangerous for Crevelt —

man, the nature of which is unknown

Egg (insect, Locusta viridissima, and migratoris)

Shell An animal combustible substance John —

and phosphate of lime

Contents Albuminous matter, a yellow fluid John —

fatty oil, a little jelly and a characteristic substance, acid, phosphates, and sulphuric alkali

The most interesting of the investigators in this table is Dzondi, whose work in 1806 was the first in which definite chemical characteristics were systematically followed throughout embryonic development. It is surprising that so long a time should have elapsed between Walter Needham and John Dzondi: no less than 139 years.

After 1 8 14 events were to move so rapidly in the world of science that it would not be possible to follow all the embryological work that was done, and at the same time maintain the proper proportion between the historical part of this book and the other parts. The eighteenth century was the period during which the chemical side of embryology began to differentiate and split itself off from the rest. After 1 8 14 it pursued a course of its own, the individual tracks of which I shall mention under their appropriate heads. But another century had yet to pass before the value of the physico-chemical approach to embryology could become generally recognised, and we are ourselves only at the very beginning of this new period.

A certain contrast may appear between the critical treatment which I have given to the investigators whose work I have been discussing, and the saying of William Harvey's — "all did well", which stands prefixed to this Part of the book. Yet history without criticism is a contradiction in terms, and the praise and dispraise, which I have tried to allot as accurately and justly as I could, is, as it were, technical, rather than spiritual. All the workers who have been mentioned, and others besides them who left no special marks on their time, are worthy of our respect and of our fullest praise, for they preferred wisdom before riches and, according to their several abilities and generations, diligently sought out truth.



All things began in order, so shall they end, and

so shall they begin again, according to the ordainer

of order and the mystical mathematicks of the city

of heaven.

Sir Thomas Browne.


There have already been certain reviews of work in chemical embryology as a whole, among which those of Paechtner and Schulz are the most valuable. The former dealt almost exclusively with the chemistry of the egg from a static viewpoint, and only devoted a short section to the metabolism of the embryo during its development, while the latter, though dealing specifically with embryonic metabolism, gave hardly more space to it than Paechtner. In both cases the discussion was little more than a catalogue of references, and in neither case was the literature anything like complete, including, indeed, less than a tenth of the relevant citations.

The first review of chemical embryology was written by Grafe in 19 10, but, though he outlined several valuable ideas, it is now of small importance. Good information may, however, be found in Aron's monograph on the chemistry of growth and on the mammalian side there are Harding and Murlin. Other, less satisfactory, reviews are by Cazzaniga and Steudel. Finally there is, of course, an immense amount of work which can be found in no review, for investigators have followed the counsel of Godlevski (1910): "Unsere Kenntnisse hinsichtlich der chemischen Zusammensetztung der Eier noch lange nicht ausreichend sind, so waren weitere Forschungen auf diesem Gebiete auch aus dem Grunde sehr erwunscht weil sie den Ausgangspunkt fiir die Physiologic des embryonalen Stoffwechsels welcher bisher gleichfalls nur sehr wenig untersucht wurde, bilden mussen".

Every effort has been made to give an accurate and complete presentation of the data in the Tables of this book and of the experimental conditions under which they were obtained, but investigators should always consult in addition, whenever possible, the relevant original memoirs referred to in the Bibliography.



I -I. Introduction

In giving an account of the present state of our knowledge about the chemical constitution of the egg-cell and the food-material which is accumulated around it or inside it, I shall not follow a strictly logical order of exposition, according to the phyla of systematic biology. I have judged it best to begin with the egg of the hen, for not only is it the most familiar and the best known of all eggs, but it is also the one which has been most thoroughly investigated biochemically.

It should be remembered that the two main morphological divisions of the egg, (a) the egg-cell itself and {b) its coverings, appear in protean modifications throughout the animal kingdom. The former may be a simple cell with its ooplasm, nucleus, nucleolus, etc., as in the echinoderms, and no covering at all save its cell-membrane, or at the other extreme it may be swollen up with food-material or yolk to the prodigious proportions of the avian egg-cell. The membrane again may be a thin coat of investing cells such as the tunicate egg possesses, or it may be the jelly of the amphibian egg, or, again, it may be the complex arrangement of egg-white, chalazae, shellmembranes, and shell, which is present in the bird's egg. All imaginable degrees of richness in yolk are present in the egg-cells of animals, and upon this fact depend the various kinds of cleavage which they show: alecithic eggs, on the one hand, such as those of most invertebrates, having a holoblastic form of development in which the whole egg participates in cleavage; and yolk-rich eggs, on the other hand, such as those of most vertebrates, having a meroblastic development, only a localised part of the egg undergoing cleavage, the rest remaining as a sac full of yolk until it is finally absorbed.

1-2. General Characteristics of the Avian Egg

After the historical introduction which has been given, it should be unnecessary to remark on the general arrangement of the bird's egg. We have with Harvey referred to it as an exposed, and, as it were, detached uterus, and with Fabricius ab Aquapendente we have


enumerated the parts of the typical avian ovum. Fig. 13, however, shows the general disposition of parts diagrammatically.

First, as to size and shape. The size and shape of the egg were shown by Curtis in 191 1 and by Surface in 191 2 to be due partly to the structure of the oviduct, which very probably may be considered an inherited character, as was claimed by Newton. D'Arcy Thompson's discussion of the mechanics of egg-formation in birds, in his Growth

Fig. 13. Diagrammatic representation of the hen's egg. The chalazae were called by Tredern Ligamenta albuminis. Bartelmez gives a discussion of the factors governing the angle which the embryonic axis makes with the axis of the egg as a whole. The yolk is not a perfect sphere but lengthened along the main axis. The egg-white is divisible into three layers which increase in density from without inwards. The chalazae, as Berthold was the first to find, are not present in reptilian eggs.

and Form, will be famiHar, but some biologists, such as Horwood, have taken exception to his conclusions about the physical influences which shape the egg. Ernst's well-known experiment was the startingpoint of these discussions ; she caused hens to lay on a surface of wet sand and charcoal, and so, observing the process, found the blunt end to be blackened. This was in agreement with many other observers, such as V. Nathusius; Landois; Jasse; Konig-Warthausen and Erdmann; and d'Arcy Thompson accordingly described the hen's egg as moving down the oviduct blunt end forwards, the pointed end owing its form to the peristaltic compression of the oviduct. Unfor


tunately all observers agree (Purkinje; von Baer; Coste; Kiitter; Taschenberg; Wickmann and Patterson for the hen, Blount and Patterson for the pigeon, Kiitter for the hawk, and Wickmann for the canary) that the pointed end passes first down the oviduct. It appears that the egg must turn right round in the act of being laid, and Bartelmez, indeed, has seen this occur. Curtis has shown that the shape of the egg depends to some extent upon its size and this biometric observation was afterwards confirmed by Pearl & Curtis. Many abnormalities have been reported in eggs. They need merely be mentioned here with their authorities, thus :





Eggs containing masses of tissue^ more or less organised.

von Nathusius.



Dwarf eggs.

Pearl & Curtis.



Ovum in ovo.






Pearl & Curtis.



Roberts & Card.


Double and triple-yolked eggs.


Parker. '




Inadequate shell.

Riddle & King.

Dwarf or absent yolk (ovum centennium^).

Mercier. Szuman. Bugnion. Gelabert.

^ See Sir Thos. Browne, Pseudodoxia Epidemica, Bk iii, ch. 7, "Of the basilisk". The eggs of Chelonia also, according to Deraniyagala, are sometimes laid without yolks.


It is interesting in this connection that Riddle has traced the occasional production of eggs with deficiency of white and shell but not of yolk, to a lack of the thymus hormone which he has called "Thymovidine". Feeding with desiccated thymus removed completely these effects. "The whole of the data", he said, "seem to demonstrate the presence in the thymus of a substance having a highly specific action on the oviduct of birds — and presumably on that of all those vertebrate animals which secrete egg-envelopes." The syndrome involved eggs with normal yolks but hardly any shell or albumen, frequent reduction of normally paired ovulations to single ovulations, diminished fertility, and restricted hatchability of the eggs. "Though not necessary to the life of the individual", said Riddle, "thymovidine would seem to be essential to the perpetuation of those vertebrate species whose eggs are protected by egg-envelopes. Such animals were the ancestors of mammals and thus mammals could hardly have come into existence without the thymus." These considerations are of much interest in view of other speculations on the evolutionary aspect of chemical embryology, e.g. Section 6-6. They also suggest that the mammalian thymus is now a vestigial organ.

The air-space, the shell and the white of the normal egg need no special remark at present, but the yolk is a more complicated structure. Around a central core of "white" or "milky" yolk the yellow yolk is secreted in the ovary of the hen in concentric layers, which form the appearance of "haloes" in the finished egg, and which show up especially clearly when the hen is fed on Sudan III or some other nontoxic dye which has a selective staining action on fat. The white yolk in the centre is continued in a flask-like shape (the latebra) up to the surface of the yolk underneath the germinal disc, and is then continued in a very thin layer all round the exterior of the yolk underneath the vitelline membrane. The white yolk is thus the first nourishment of the embryo. It is not certain whether there are also layers of white yolk between the concentric layers of yellow yolk, for they have never been analysed chemically, and Balbiani maintains that they only differ from the yellow layers by having less yellow pigment. The differences between the true white yolk and the yellow yolk are, as will be seen later, far more profound. Balfour & Foster, in their Elements of Embryology of 1877, described the yellow yolk as consisting histologically of spheres of from 25 to loo/x in diameter, filled with numerous minute highly refractive granules and



[PT. Ill

very susceptible to crushing and rough treatment. After boihng, the spheres assume a polyhedral form. The granules seen within them must consist of protein, for they are not soluble in ether or alcohol. On the other handjthe white yolk elements are vesicles smaller than the globules of the yellow yolk, being about 4 to 75 /n across, with a highly refractive body, often as small as i [x, in the interior of each. These vesicles are sometimes collected together into much larger vesicles. They observed also underneath the blastoderm or the germinal disc a number of large vacuoles filled with fluid — large enough, in fact, to be seen with the naked eye. The histology of yolk has been reviewed by Dubuisson, and at one time many papers were published on it, e.g. those of Virchow. They cannot be considered in detail here.

1-3. The Proportion of Parts in the Avian Egg

Of the weight of the whole egg, the shell takes up about 10 per cent., the albuminous white 50 per cent, and the yolk 30 per cent, in round numbers. These relationships have been determined by a multitude of investigators, whose results are drawn up in Table i .

Table i . Distribution of the parts in the egg,

Italic figures represent dry weight only.


egg weights










and date

Hen, Polish i



van Hamel-Roos ('.890)

,, Polish ii




,, Holland (Zwol.)




„ Holland (Tiel)









Miinster Ag. Sta. (1900)





Drechsler (1896)










Plimmer (1921)


I i-i








Langworthy (190 1-2)





















Lebbin (1900)





Welmanns (1903)






Segin (1906)






Liihrig (1904)







Hen (various breeds)





von Czadek (191 7)





Rose (1850)





Hen (various breeds)




Carpiaux (1903)




Lehmann (1850)




Prout (1859)




Poleck (1850)



Stained by Kossa's method for the detection of calcium phosphate. The considerable variations in the vitelline globules may be noted. Magnification, 6xD: prepared and microphotographed by Dr V. Marza.



SECT. l]



Table i {cont.)


egg weights










and date

Hen, Leghorn



Murray (1925)

Hen (various breeds)





Iljin (1917)









van Meurs (1923)












Voit (1877)

Nidicolous birds




Tarchanovf (1884)

Starling ...












Canary ...




Thrush ...















Nidifugous birds

















Turkey ...













17-78 (grains)




Glikin (1908)






Davy (1863)





















3 J

Golden-crested wren





) )








































(gm-) 57-57




Hartung (1902)





Voit (1881)



Fere (1896)









Pott & Preyer (1882)






Rozanov (1926)






Hepburn & Katz (1927)











Baudrimont & de St Ange (1846)

Dwarf hen ...




Sacc (1847)



Pott (1879)





Atwater & Bryant ( 1 906)


■f- All Tarchanov's figures exclude the shell weight.



[PT. Ill

Table i


Weight of








tents (gm.)





and date


• 57-12





Friese (1923)


• 137-38






781 1






• 92-93


















• 25-40












. 27-03





Blackbird ...
















Weight of








ents (gm.)




and date

Plover ( Vanellus crist






Bauer (1893-5)

Hen {Gallus domestict







Guinea-fowl {Meleag

ris gallopavo)





Swallow (Hirundo ru.






Partridge [Perdrix ci






Sparrow {Passer dom



Thrush (Turdus

? ) ...





Duck (Anas) [doubl






The above data were all obtained without any ad hoc investigation of the probable errors involved in weighing eggs and parts of eggs. An elaborate study by M. R. Curtis in 1 9 1 1 gave the following results on Gallus domesticus :

Actual weight in gm. %

56-04 100

33-22 59-26

16-31 29-14

6-28 ii-i8

0-23 0-42

Whole egg Albumen


Shell and membranes Error

But though this is the case with the egg in its natural state, the solid matter is concentrated much more in the -yolk than in the white, so that, as the analyses of Poleck and Iljin, for instance, show, for dry weight the conditions are exactly reversed. The egg-white may, indeed, be regarded as the principal reservoir of water for the embryo which develops on dry land, and this is a point which will be discussed later (see Section 6-6). The eggs of different breeds of hen vary to some extent in the relative weights of shell, white and yolk; but, although it is difficult to lay down any general rule, these variations do not greatly exceed the variations due to factors connected with the individual hen. Iljin's lightest shells make up about 7 per cent, of the G,gg weight and the heaviest not more than 11-5 per cent.


It is certain that there are constant differences between the eggs of different breeds, but as a whole these are quite outweighed by individual differences, and only appear when extended statistical studies are undertaken. The eggs of other birds, however, do not fall within these limits. Langworthy, for example, has shown that, in the duck's egg, the shell may account for as much as 14 per cent, of the whole weight. A similar result was found for the turkey and the goose, while the guinea-fowl's egg has a shell of nearly 1 7 per cent, of the whole weight. The wide series of Friese, shown in Table i, seems to indicate that the larger the egg the more shell it has to have : thus the canary's egg weighing just under 2 gm. has 4 per cent, while the goose's egg which weighs 137 gm. has 14 per cent. Heinroth, and Groebbels & Mobert, among others, have collected a great many data of this kind for all varieties of bird, but their papers must be referred to for the figures. Thus the fertilised embryo starts its development on the surface of a mass of food only slightly diluted with water, and surrounded by a further and much wetter supply. This is reflected well by the work of Bellini, who found that the yolk of the hen's egg was seven times as viscous as the white at the beginning of development. (Alb. 3-4 units, yolk 28-5 units.)

A good deal of work has been done on the variability of the weights of the parts of the egg within a given species of fowl. Thus Jull found that egg weight is the least variable factor, albumen weight slightly more variable than egg weight, yolk weight considerably more variable than albumen weight, and shell weight the most variable. It would seem, therefore, as if a compensatory process takes place during egg-production, the largest yolks having the smallest whites, since the weights of the entire eggs do not vary as much as the weights of the components. On the other hand, the smaller eggs contain the highest percentage of albumen and shell and the lowest percentage of yolk. Jull also studied closely the seasonal variations, which may be quite considerable, finding that the component parts of the egg contribute in different degrees at different times of the year towards the total egg weight. The question as to which part of the egg is mainly responsible for large or small eggs is still debated, for Curtis concluded from his observations that it is the egg-white, while Atwood found many indications contrary to this. Statistical studies on the egg of the tern have been made by Rowan, Parker & Bell; Rowan, Wolff, Sulman, Pearson, Isaacs,



[PT. Ill

o V

CO „


« CO

c -n CO -C CO-X3 •s ^eo 3 CO Kj

(V^ 3 . .c?i ^ C 1


"£ ^q o ^^C5 g o 3

CL, ^ U CM ^ < O pa

S ^



o 05 y <^





O «3 Thin 9 r^

.t! to CO CO r^ <i

^ I















a i

o « 

.§ 3 -a

'S o o

i c f3

C « i o botJ

[i, O CO " O CO

to « CO Crs N 1^

« cn 05 o Oi^ ts o Co o O) 0500 oico o 05 oj 9 cp

<o lO c<-)Co e« lOif^Tj^crjc* fso o> r^<o cn 7^ ^p

• 3 co^ri " CO CT>

V m CO t^t^ ■*

O !£> « CO CO "

&3 " " " " "

f~- Tf< OiQS O^COkiCO 0^';*'C\| LOOjf^ t-v<0 ^ CO


CO to o oi

■;*" r^ iT)^ CO lOCp ^ ts ^ irjCO irj '^ d-j s 7^ f^ N e» coTfcoei " I -^,'^ 6 ,0 CD 6^ eo.crj m ,0 ■^,f^.,ir) n ■^



••::•:::::::::•:::: : : : : : "^ « ;i:;rr.; iiii^tiii*

u o -H c

c oc>c.„.u.o„i2-.S >"a.c7^

ffi qjuohK q o h o (!< s hi

SECT. l]





'""'<« 2,

en O ^ C8 ^-' bo

g e'C


- c 'S,


<^_^ en

^ t-" ^y^

s 3 s

  • i CO CO

C CO „

IS .— .^_^co -—

s s H Sea —





■<*<o> oco ■*6

o o



6 "

CO 10

O O O 1-' - CO

CO Tt< O " CO CJ f^

io " 6 6 CO 1^ CO

r^co CO CO CO tri r^

o w

« "CO CD


« - 6 io

1 I 1 1 1



10 1 C< CO K


i-i 1 W -^ O)

1 1 T^ 9^ 1 1 ■^ co^o

►H 1-C k.




lO lO 1 CO CI "O >*,



1 CO " CO CO


C u . . . .

(u _Q I- S 
 : ^ : :

SfZ 2 cii o

rt^i!'-' ^-5^2 O^

"CC ■^ O ^ -^ -^ • = g^ g^.SpS^ |g g g

K E >H 2; P5 o ph ffi


o bcpd

o > C o o C 3 O flj o 3 « 

_|'SEQS „~ c" ^ „ .,

^ PS 1; « ., r.


- o y c!







Elderton & Tildesley; and by Watson, Watson, Pearson, Karn, Irwin & Pearson.

The division of birds into the two classes of nidicolous (those which hatch as squabs) and nidifugous (those which hatch as downy, feathered and active chicks) has been shown to extend to the composition of their eggs by several investigators. Davy found that the eggs of the nidicolous birds had thinner and more fragile shells, which took up a less proportion of the weight of the whole egg than the shells of nidifugous birds. Thus the wren's egg-shell weighs only 5 per cent, of the whole egg weight, while the hen's weighs lo per cent. Da\y's figures show very clearly that the main reservoir of solid is the yolk and the main reservoir of water is the egg-white.

Tarchanov carried the matter further, and observed that the yolks of nidicolous birds always formed a smaller proportion of the total amount of material inside the egg than in nidifugous birds. Thus, for the former class the egg-white accounts for about 78 per cent, of the egg and the yolk for 22 per cent., while, in the latter class, the egg-white accounts for about 55 per cent, and the yolk for 45. All these differences are probably related to the shorter incubationtime of the nidicolous eggs; and, as will be seen later, there are not wanting indications that the yolk of these is less tightly packed with food-material and more rich in phosphatides.

1*4. The Chemical Constitution of the Avian Egg as a whole

The composition of the egg as a whole is further considered in Table 2, where it is noticeable that the analyses of water and ash have not been significantly improved upon between 1863 (the date of the first analysis, Payen's) and the present time. The later figures for protein and fat are, however, much the more reliable. It should be observed that there is an approximately equal quantity of fat and protein at the disposal of the embryo, though the former is, of course, in the yolk, and the latter is preponderantly in the egg-white. This protein-fat equality is by no means the rule in all eggs, and, as we shall see later, the eggs of fishes depart widely from it. There appear to be only small differences between the eggs of different kinds of birds in protein content. At one time it was thought that the duck's egg was particularly rich in fat, on the authority of Commaille's analyses, but Liihrig has since then brought it into line with all the others. It does seem, however, to have a considerably higher per





centage of mineral substances than the rest. The dry-weight figures merely demonstrate again the approximate equality of the protein and fat.

Before we proceed to consider the parts of the tgg in separation, the question of individual and racial differences must be taken up

Table 3. Individual differences between hen's eggs, Malcolm's figures (1902). Averages of individual hens.

Breed unknown

Italian hens (fed on maize and barley)

From one hen

From one hen

Iljin's figures (19 17).

Houdan... Orpington Plymouth Rock Rhode Island ..

itty acids



















I 52






















30- 1 2







% dry weight


Lecithin P











59- 1 6



Table 4. Race differences in hen's eggs.

Leveque & Ponscarme's figures.


% of egg d.

"y weigh



dry weight



of whole



N in



i> 111





in yolk


Andalusia ...














Bressane ...







Coucou de Rennes














Dorking ...







Faverolles ...














La Fleche ...























O ■"

WD "^ _ _,_ _^__ _.__ ,_^ ^_^__ __.__

J2 050 -^CTl m(N COOtD-r)"COOtDCO {S«30 I^

2 eD-*inTt<r^io<r>co co^p ^nto 'O'X) m coco in

2t; CTio o-*' COM cni^ or~ cor^ tj-od co ct> <~~ ci "SS in^<N<o ocococo co«coo oeoip >oo co ccdo"'-"'-' Of" 6" "'-I "CI o<6 66c<"

a a Tt>o com o<co into oci p~-o cocoot~4j(00 lOCT) coio i-cio -^lO oco mmtD maii-i

r^Mcoc^cooci 00)66 ctjo 6" 6<oicO" " cococococococow cococtcocococotxcoco

i- o o o o o


■53 oit^tico (r)f~-t6t6 trjLO ■ij-ti) (i> f^ t6

i-i coco i^ CI 0^0 V a><£) lo CT) t£> ■*

rt (X) C£) CO •^ CT> K


6 6 6 6 mm mm

o CO r^ o

CTi r~- (T) ■*

« en cicb m -^ ■* ■*

- CO CO c< CO CI to 05C£) oieom'^'cici o ocor^

-fi CO CO CO CO CTlCO CO CT> CO O OiCO CD p r^ f^ O CO

<6666 66 66 6« 66 6" 66«6

ot; o-*ciTi-coococi ocicoci o^n coocor^ -O 2 f^ CI coco ►"t^mcTico'^cici cocna)coCT;co

o o ■- "

o « o " o


t^ m a^ o^

0000 6 6 6 6

00 00 00 00 00

co-tomcomco"- otiD cicitoei i.C7^'CT>cicoc7)cotpoco(£!'*coor^_

u'v ci" cici Tfci 616 "6 i-" '*Tf<eo65-^ci

d m ■^ ■^ m




r-to en



^too ctjci ocicococimcirt" ■J3 mcDi^o mci yh'Lnmcocir~tt( in in 'J^ t^ mc£> ■^ m mt£i «6 m

oi en CO CO CI CO CI 1^ CO 7*" 9 CO mt£> in ■* r^ m

« o CO Im CO CI m

m-^cocotocotrsmcico r^eo" «  en-* oiCT! coco mci too to o^mo inf^ 4fmtDf^!i>(o mf^^o ■^cnti

^en-*toco cor-^eomoio ci'^'P';-' ci^Jcoi"7, 66 r~f^«"Hi cimcocovoto 6<J>v6 r^" tj< Sh cococici •^coeocococococo'^eococi-^co

< bo"_c cbCT)COco(i)cb r^m W> mmi£><o Tj-m mm


CO CO CI M r-^ r^ CO ' tJ< r^ 1- « to CO in 1 m ■* m m -^vo m

1=3 CTico r^ m _^ 6 oi coco

P ^ CI CT)


•*tj< r-co cir~ r^mr-~ei

66 cndici6)cicbci 6




o >

^ti m" ci-^ioto oeoto-cococi'+'enoci co

^ -tj, mco t^"^ o^ m t^ •* oi t^ mco it* ci to r^ r^ to

^•r; f^f^ 6 6 f^cb -^cb 1" '

P^^^-CICI "«" "-i-< '

to CO r^ t^

fiP'^'ij CI in r-«r^ W ^ > coco ^'^

d cir^-*0 r^O mi

to « r^ CO

CO-* CT)05 mm r^i-< •*" co coco ci 6 ■* ineb


S-^ oci mo p-Tj-t^o ■*'-' coo com-*omai djbCdr^mco cimcom mm mco ci r^ m m m m ■^^tointoto inin-*-* into in -*



in ■* to -^to

mmi^f^ -^m -^m mto

t3 G _

« o ^


« c o

en o * CO m CO " m

I p 2


JS'.S 000


c c L0-§



M 3



S w 6 6 6













m m


ri « 





to to


-, n

rf *

w' 3



" CO









  • r-








CO to


V <f

0^ CO



■* m










c 1— 1







u.^ CO Oito

.Q I d^CT5 en

J3 "& >^ o eo to K .SP C f^co t~«  it^v-a^ ■- ■ ■

bo ?








,..j5 !S o CO

bc-^coto r ll3 'C ,2 °^ en

'^ > to m m











<M 11 i-i (N « "

O " O O


coco CO =? ? T

•<*"•*■* •* ■* T^


J3 b£ bo« 

o o o

in o o<

o o o

6 6 6

6 6 6

6 6 o

w "^ « 







50 I

w 5

.SPI > .SP o >

S J < £ !-!<

jjoo-g uJoqSaq

o « 

bc.-a ^ oitri <.o o ■:" IN CO 11 "5 J? to in lo f^ f^ intis (6


CO t^ <M CO r~« eo m

^o o CO in o «r> 00 f^to to f^ CD r^ ti

CO OD CO CO 03 CO 00

lite Yolk ght weight

0503 05 01 '^ 1^ r^(£)

coy3 ^ i^" ':' '^ <p^'-p

O) CO r^tr> CO r^ o< O

<^ m Oi

in'-b in r^ in in -^^

ci f^ci) tr>

(6 r^ to

•* CT) Tj- m coco 1^ Tf ■^ O) o uD c« O) inco

o 05 mr^ ■tf o ^ in

05 in f^

Otp CO


««665e<coc<e( cococoo* cocococo

tJ< CO ■^ in


CO '^ CO coco CO

V bO

O CO i^co CO e* m >tJ< tJ< Ln CO CO 05 r^co

o t~- ine*

N m CO o r^co

Tj^eoin^ ineoiheo

io -^ io 'Ji

^in Af


en 1-4

w 1


inminm mininm

CO o •* in Of into i£)

to into CO in in in in

o in i^

into o

incb r^ in in in




o -a









o « 

U « C C O H


ttj a./


CO NH CO *-•

oi in in ■^ 6 6 6 6

(N 01 C< CO

bl „ >-l » u

ttl 01 1- ~ CI


o r^ r~ r^ j~»

•- to >- CO m ii CO O) ino) o ^ u ci (s

« c( in in

03 1~~ lO O)

to to to f~in in in in

r- Oito J in O r}< j

05 6 6

'Z.^ <^

o o o o

_j CO

C O (U ►C V S •- .


again. Malcolm maintained in 1902 that, although there were undoubtedly differences between the eggs of different breeds of hen, they did not exceed the amount of variation between individual eggs from hens of the same breed. His results are shown in Table 3. Thus, although the feeding was carefully controlled, the eggs from one hen might show a difference of i'i4gm. of fat, while between two eggs from different breeds the difference might be only 0-13 gm. His conclusions were supported in the main by Carpiaux; Leveque & Ponscarme ; von Czadek ; Iljin ; and Willard, Shaw, Hartzell & Hole; who made very long studies of a considerable number of breeds. Some of the figures obtained by von Czadek and Iljin are given in Table 5. Von Czadek studied the Sulmtal, Minorca, Orpington, Rhode Island, Faverolle, and Wyandotte races, together with an Italian and a Rhineland breed. His outside values for egg weight, for instance, were 43 gm. and 75 gm. — a considerable difference — but the former was from the Rhineland hen and the latter from the Minorca variety. The span showed great variations, thus an egg weighing 55 gm. might be a heavy Orpington or a rather light Faverolle or a medium weight Italian. The only breed which stood well out of the range of individual differences was the Minorcas which laid very heavy eggs. Certain instances have shown, however, that remarkable agreement may exist between work done on eggs of widely different breeds. Thus the classical work of Plimmer & Scott on the phosphorus metabolism of the developing chick was confirmed very strikingly by Masai & Fukutomi, who worked in Japan. Here the correspondence was almost numerical. But on the other hand there is evidence that eggs of different breeds differ not only in their gross characteristics, but also as regards more subtle properties; thus Moran has demonstrated that eggs from different breeds of hen vary very greatly in their resistance to cold, so that the viability is different, and Needham, working on the inositol metabolism of the embryo, observed differences between the embryos from Black and White Leghorn hens. Physico-chemical differences between breeds of silkworm eggs are enumerated by Pigorini.

The individual differences between eggs may be equally important. Benjamin has shown that there are numerous variable factors which modify the constitution of the egg. The amount of yolk, egg-white, and water, as well as the thickness of the shell, vary according to the season, diet, age (Riddle) and general condition of the bird


in question. Nor are such comparatively slowly changing factors the only ones which bring about differences between individual eggs ; the time the egg takes to pass down the oviduct, for instance, will materially affect the amount of albumen it contains, and such variable quantities as the blood-sugar level (Riddle) and the level of cholesterinaemia in the parent animal will exercise their effects upon the resulting egg. Again, the length of time elapsing between the laying of the egg and the beginning of incubation will have a marked efTect, for a certain amount of water will evaporate from the egg-contents through the shell, and just how much does so will depend on the humidity of the surrounding atmosphere. The process of water-absorption by the yolk (Greenlee) from the white will also be affected by these conditions, so that the embryo at the initiation of its incubatory development may find a remarkably inconstant set of circumstances in its immediate environment. Moreover, a certain amount of development always takes place in the egg after fertilisation as it passes down the oviduct, so that the embryo has already gastrulated by the time that the egg is laid by the hen. It was the ignorance of these facts which led Malpighi, as we have already seen, to his erroneous conclusions, for if he had known of the phenomenon of "body-heating", as it is called by the poultryfarmer, he would not have put forward the preformation-theory, and the eighteenth century would have been spared the trouble of getting rid of that embryological phlogiston. Thus no two eggs are ever exactly the same age, and as there is reason to believe that enzymic action begins in the yolk, if not in the white, very shortly after fertilisation, this fact makes it additionally difficult to get precise figures for the constitution of the unincubated egg. Then the position of the egg in the clutch (whether first or second) in pigeons may, according to Riddle, make a difference of 9-15 per cent, in yolk weight. It may be concluded that nothing short of the greatest caution must be employed in the material which is used for chemico-embryological researches on the hen's egg. The individual hens should be marked, and the eggs produced by them should be noted, their food should be constant in composition and the breed used should be not only single in any one series of experiments, but also, if possible, genetically pure. It is very greatly to be wished that standard hens could be obtained, such as the standard rats necessary for feeding experiments, and much further work, with a proper statistical backing, is needed


on the range of individual and racial variations in all the properties

of eggs.

The effect of the diet of the hen on the chemical composition of

the egg has been studied by various workers, notably by Terroine

& Belin. Except in certain respects, it showed a remarkable fixity

of composition :

Table 6.

Ordinary Corn and potato Hemp seed mixed ration almost ration

ration free from fats (fatty)

White in % of total weight ... 56-7 54-3 —

Yolk in % of total weight ... 31-3 34-0 33-2

Shell in % of total weight ... 11-4 lo-g — White

Water % 87-8 87-4 87-4

Ash% 0-49 — —


Water % 49-9 50-33 50-99

Ash% 1-48 — —

Total nitrogen % ... ... 2-67 — —

Total fatty acids % 28-4 26-6 2655

Unsaponifiable fraction % ... 1-85 — 2-08

Cholesterol % i-i8 1-58 i-ii

Lecithin P % — 0-425 0-434

Thus, although the character of the substances stored for the use of the embryo can be varied considerably, as will be seen later, the balance of them cannot. But the question is probably rather complicated, for it has been shown by Dam that by feeding hens on a ration rich in cholesterol, the cholesterol content of eggs can be increased from 501 to 615 mgm. per cent, of the wet weight or roughly by 22 per cent, of the original value. In another instance the cholesterol rose from 476 to 560 mgm. per cent. This would not be in disagreement with Terroine & Belin's figures, but it would be a very desirable thing to make a detailed study of the limits of variation of all the constituents of the egg, and to find out exactly how different in chemical composition an egg can be from the normal while retaining its hatchability. Klein regards the cholesterol output of the hen in its eggs as showing a synthesis of that substance in the parent body. Leveque & Ponscarme have stated that it was not possible to show any effect on the eggs in eleven breeds of hen by minor variations in the diet; and this was amply confirmed by Gross.

The ingenious and partially successful attempt of Riddle and Behre & Riddle to make hens preserve their own eggs by feeding them with hexamethylenetetramine, sodium benzoate, and sodium


salicylate, may here be mentioned. Starting out from this practical suggestion the work led to the discovery of a number of specific effects of substances such as quinine on egg size and yolk size. Thus Riddle & Basset found that alcohol markedly reduces yolk size in pigeons, Riddle & Anderson found that quinine reduces egg size, yolk size and albumen size but has no effect on the protein/fat ratio of the egg, while Behre & Riddle found that the diminution of albumen size under quinine bore more on the solids than on the water and involved considerable reduction of the protein.

The elaborate investigations on the egg of the tern, already mentioned, led to a significant correlation between abundance of food and size of egg, and it is certain that the size of the hen's egg is affected by its diet since the work of Atwood. There seems also to be a seasonal fluctuation, the weight of the eggs increasing from July to February and decreasing from March to June. These seasonal fluctuations appeared distinctly in Atwood's data, and explain the results of Curtis and of Fere. Rice, Nixon & Rogers and Riddle found a definite relation between the amount of food consumed and the number of eggs produced, both of these factors varying exactly with the seasonal variation in the egg size. Fluctuations of a regular kind seem even to occur each month, according to Hadley who observed such changes in egg weight and number. According to Curtis the size of the eggs increases as the laying bird matures, in the case of the hen, and Pearson has observed similar variations in the case of the sparrow.

The genetics of egg production have been studied by Pearl and Benjamin.

The relation between the egg weight and the chick weight at hatching has been studied by Halbersleben & Mussehl and by Iljin. The former workers found a quite consistent relation within one breed between the weight of the egg before incubation and the weight of the chick at hatching, the latter averaged 64 per cent, of the former. After thirty-five days of post-natal life, however, the slight advantage possessed by the chicks from the heavier eggs had altogether disappeared. They also noted that, other things being equal, chicks hatched from the more pigmented eggs (browner) weighed slightly more than those hatched from the less pigmented ones. Abnormally large and abnormally small eggs did not hatch as well as those of medium weight. Iljin collected a great many figures but his text contains no statistical analysis.


Stewart & Atwood reported that chicks hatched from pullet eggs were neither so large nor so vigorous as those hatched from the eggs of hens two or three years old. Whether there is here a direct effect on the chick of the age of the hen, or whether the effect is indirect, due to the small size of the egg, may be well questioned.

What relations exist between the chemical constitution of the egg and the percentage "hatchability" are at present obscure, owing perhaps to the comparative crudity of our estimation methods. The work of Pearl & Surface indicated definitely that differences in the hatchability of eggs are determined by or associated with innate differences in the individual hens which laid them, that these differences are probably inherited, and that variations within rather wide limits in certain environmental factors, e.g. the temperature, during incubation, are of secondary importance in determining the death or the hatching of the embryo. Hatchability of embryos would appear then to be, like fecundity, a heritable character. The experiments of Lamson & Card confirmed the conclusions of Pearl & Surface, but although some physico-chemical mechanism is undoubtedly at work, these statistical studies gave no hint as to its nature.

Dunn determined to probe further into it. In his first paper he argued that if hatchability was associated with constitutional vigour, it should show a correlation with such a value as the chick mortality in the first three weeks of post-natal life. Experimentally this was not the case, e.g. post-natal mortality remained the same, although in two instances the pre-natal mortality was on the one hand extremely high (20-39 per cent, hatchability) and on the other hand extremely low (80-100 per cent, hatchability). It therefore seemed likely that mortality before and after hatching is determined by quite different factors. The more specific influences operating in embryonic life must doubtless be looked for in the physico-chemical constitution of the unincubated egg.

Hays & Sumbardo, in a search for such influences, were able to exclude statistically fresh weight, length, diameter, specific gravity, shell thickness, outer and inner shell-membrane thickness, porosity and imbibition of water from 25 per cent, salt solution. Other factors which have been excluded are percentage of protein in the diet of the laying hen (Rosedale), percentage of yolk-pigment (Benjamin), evaporation rate of the egg (Dunn), yolk-fat percentage (Cross), egg


fat percentage (Cross), yolk-protein percentage (Cross), egg-protein percentage (Cross), egg-phosphorus percentage (Cross), chickphosphorus percentage (Cross).

It appears, however, that the constitution of the egg-proteins maybe influenced by the presence of unusual proteins in the diet of the hen, and that this may influence hatchability. Pollard & Carr have reported the results of feeding the following proteins to laying hens : wheat, rye, corn, oats; kaffir, barley, peas, soya, hemp; buckwheat, popcorn, sunflower seed.

The first group of four (all, of course, being fed alone) were very efficient for the production of normal eggs; the second group (of five) permitted the hens to lay eggs but the eggs were hardly hatchable at all, while the third group allowed of no eggs. Pollard & Carr studied the egg-proteins in all cases and obtained evidence of tryptophane deficiency in the second group, so that they concluded that a minimum tryptophane content was essential for successful development through hatching. It is unfortunate that their results were never published in full.

The effect of sex on the chemical composition of the egg has been discussed by Riddle. As is well known, in some, probably most, animals, the male produces two kinds of spermatozoa which are not equal in their prospective sex value, i.e. some which will give rise to females and some which will give rise to males. In birds, on the other hand, the dimorphism of the germs exists not in the spermatozoa but in the egg-cells. The female produces two kinds of eggs of unequal prospective sex value. Riddle found that pure wild species of doves and pigeons were ideal material for studies on sex, since very abnormal sex-ratios could easily be obtained from them, and his studies led him to the view that sex was more a matter of metabolic level or rate of protoplasmic activity than anything else. But what concerns us here are the consistent differences which he was able to demonstrate between male and female eggs.

Pigeons generically crossed, when not permitted to lay many eggs, produce only males, but when made to lay many eggs produce first only males, and eventually "under stress of overwork" only females. These facts and their proper conditions having been ascertained previously by extensive statistical investigations, the way lay open for the chemical analysis of the two sorts of pigeon's eggs. 900 analyses were made and more than 12,000 yolks weighed.



[PT. Ill

Fig. 14, taken from Riddle, gives the differences diagrammatically. A glance at it shows that the male-producing egg of the spring contains less stored material than the female-producing egg of the

Sex conbrol and known correlations in pigeons



d cf cT d* cT ^ $ ^^ *? *$ ^9 •.>' ^ •-': "■

9 10 11 12 13 14

N? 1

N° <2

N° 3

N9 4

N? 5

N° 6

N? 7

N9 8

N° 9



N0I. shows comparabive size of eggs of Alba (A) and Orienbalis (0)

Fig. 14.

autumn. The amount of water and ash present, on the other hand, diminishes, and the rise indeed is mainly to be seen in the fat and lipoid fractions and in the calorific value. Table 7 gives the figures for one individual pigeon during 191 2. The differences are not

SECT. l]



large, but they were invariably found. Another series of figures showing the rise in calorific value during the course of the year and the transition fi-om male to female eggs is given in Table 8. Here also the increase is unmistakable. Within one clutch, also, the watercontent of the second egg is lower and the calorific value higher than the first egg, which fits in very well with the fact that under normal


May 26


June 7



July 3 5

15 17 23 25


4 13 25







Nov. Dec.

Table 7. Effect of sex on pigeon's eggs.

Female Turtur orientalis x Streptopelia alba, no. 410 for 191 2.

% wet weight % dry

Analysed Weight t ^ ^ weight

or in- of Pro- Extrac- ale- Calories

cubated yolk Lipoid tein tives Ash Water sol. per egg Sex

An. 2-330 18-32 25-44 5-28 4-85 5701 72-65 7405 —

An. 2-660 17-54 25-63 5-25 2-62 54-82 72-45 8990 —

Inc. — — — — — — — — Male

Inc. — — — — — — — — Male

Inc. — — — — — — — — Male

An. 2-026 16-49 26-00 3-63 2-43 56-05 71-95 6714 —

An. 2-330 19-18 26-55 3-75 1-93 5522 72-27 7881 —

Inc. — — — — — — — — Male

Inc. — — _____ _ Male

An. 2-422 17-82 25-88 3-82 I -80 55-84 72-42 8061 —

An. 2-720 18-88 25-96 3-86 1-81 55-33 72-45 9296 —

Inc. _______ _ Male

Inc. — — — — — — — — Male

Inc. — — — — — — — — Male

Inc. — — — — — — — — Female

Inc. — — — — — — — — Female

Inc. — — — — — — — — Female

An. 2-700 21-40 — — — 55-45 73-17 9323 —

An. 2-715 21-63 — — — 55-39 73-02 9383 —

Table 8. Eggs from the same female Streptopelia risoria (1914).


Weight of yolk Energy in cals.

June 6










I -000


July 14






Aug. 30



Nov. 6












Dec. I



















conditions the first egg laid nearly always gives rise to a male and the second to a female.

In Fig. 14 the line marked "developmental energy" implies that a higher percentage of the male eggs hatch successfully than of the female eggs. The data for length of life show the same curve. The smaller eggs of both clutch and season are the eggs which give positive results in strength and vigour tests, and the larger eggs are those which are liable to display weakness. These facts are in entire accord with the higher metabolic level which Riddle associates with the small male eggs. It is interesting to note that Lawrence & Riddle found consistently higher values for total fat and total phosphorus in the blood of female fowls than in that of male fowls, from which they concluded that the metabolic differences between male and female germs persist in the adult, and all these facts are in agreement with the work of Goerttler and Baker on human and Smith on crustacean blood-fat, and of Benedict & Emmes on sex differences in basal metabolism. But for further discussion of the metabolic theory of sex, the papers of Riddle must be consulted. Interesting data on the hatchability, vigour, etc., of rotifer eggs are contained in the paper of Jennings & Lynch, but these authors made no chemical experiments.

To say, as Riddle does, that there are, as it were, two kinds of eggs in some species, one male-producing, and the other female-producing, may either be taken to mean that there are quantitative differences between them or that their constituent substances are qualitatively chemically different, or, thirdly, that the same substances in the same quantities are differently distributed spatially and temporally. As will be seen later in connection with the lipoids of mammalian egg-cells, the second view finds supporters, and some such opinion is held by Russo. Faure-Fremiet, in the course of his work on the egg of Ascaris megalocephala, to which he applied every conceivable method, examined a very large number of individual eggs in order to find whether they separated at all chemically into two types. His method was to centrifuge them separately, much as McClendon had done with the frog's egg, and then to measure in mm. the thickness of {a) the mitochondria layer, and {b) the fatty layer. Fig. 15 {a) taken from his paper shows the frequency polygon which he constructed on the basis of these results, the ordinate giving the number of eggs measured, and the abscissa the thickness of the mitochondrial layer.

SECT. l]



It is quite evident that there are not two modes on the line joining the points, i.e. that there are not two types of eggs, but only one type. Faure-Fremiet made very similar experiments, determining the glycogen content of the eggs histo-colorimetrically with iodine solutions, and there also the frequency polygon had but one mode (Fig. 15 (^)). But this second case was based on an unsatisfactory method. In the particular instance under investigation, neither mitochondria nor



12 3^5

30 20



J :

<k 5 7 8

ib) Fig. 15 glycogen happens to be an entity which varies as between the two kinds of eggs. Nevertheless, the plan of work was an interesting one, and widely extended researches with it, using accurate chemical methods, would be very desirable.

1-5. The Shell of the Avian Egg

Litde attention has been paid to the shell of the bird's egg from a physiological point of view. The relevant analyses are given in Table 9. There is some difference between the shells of different








o o


• rt £-< tJ-* 1-H

^ OCO « 


S ^


3 --^



m g .

U " 3 . -1 ^-^ u


^ ^



g a


in CO ■*

" I~» CO

6 6 6

in-*iininTj<co,coinc«oeoor^ CT)








CO mco


o o o



CO t^ -*


O .bo



r-» CO tJh o^cd o CO oj f^ tjh a>

CoiOTf<f^-^" COlOCOtj^C^ 01CT)OCnC505CT)CT)005CT)


' Um

e o u

(U O 3



• •^


C C >- D

5 s i!

b ^ 3

O C O c O U 3 « 



. . T5 i-i . . . .

. . g i) . . . .

• • Q-a • • • •

o g ffiOOOOffiffiO

P 2

«1i I

4) o O S it( t^ O « !;<

c !- 'y ,0 c«UOh



3 II

SECT. l]



birds, and it would be interesting, for example, to know why the pheasant's contains such an unusually high percentage, of phosphorus, and why the herring gull's has so high a percentage of organic substance. A certain interest attaches to the determinations of Balland on ostrich eggs, some from a tomb of the Hellenistic period and others from modern ostriches, but the differences he found were probably not very significant, as the analysis of Torrance seems to give values half-way between those of Balland. Neither Balland, Torrance nor Wicke states whether the ostrich used was the North or South African variety, a complication which might make a difference. Wicke believed that the difference in shell-composition between different kinds of birds was almost entirely dependent on their usual foods.

The microscopic structure of the shell was investigated by Nathusius in the 'sixties, and since then little has been added to his work. The shell consists of an outer layer of crystals of calcium carbonate arranged with their long axes perpendicular to the bounding surface (Fig. 16), and an inner layer composed of undifferentiated calcium carbonate (Herzog & Gonell). Kelly; Schmidt; Meigen and Osawa have found that the mineralogical form of the lime is invariably calcite, no aragonite being present in any bird's egg-shell. This has been confirmed with X-ray analysis by Mayneord. The ostrich, Emj>s europaea, is the only doubtful case, for Kelly identified its egg-shell lime as conchite, but Torrance considers it to be calcite. Prenant's review should be consulted for further details regarding this interesting biochemical problem. Only one paper exists dealing with the changes which the shell undergoes histologically during the development of the chick; it will be considered in the section on embryonic respiration, where the data we possess on the question of the NEi 17

-^:- -;>.,«■ j

Fig. 16. a, Outer crystalline layer; b, c, d, amorphous layers; e, mamillae; /, shellmembrane.



3co ^

Ss ll«5 In





I I I I I I I M I iSalsi I I I I I I i|:m 1 I I m fji^lli I I El!" f I I '|i 1 I I I I i|

I I M ill \^f\l I I I I lsf?l II? 1 I I I I I 111 if


' I I I ?.j

1 II ?i

IS I I I I I ; ^'i I 1 I 1 1 '

I °^


'? ll

11 gl§'S!

i^loo 11 loll 5 ol°"'^lllo 00 Oc^OO^SO «■« 


III 1 1 M i 1 1 1 iiii g°Eir^R

ii|l 1 1 l|l 1 1 f I II I I I I II

|l 1 1 II l|p|l I I I I I I I I I I

'Hl.^ltli 1 1 i^ s' 6 s" MINI

I i If I |||S|.| O I g 3 .-3 iiS5^^i£^

■ ^o 6:2 8 s-vgsra

S § SO gOidpiCCK H Q Q O (


1||| I =1

"■0 ^ ^'g

"m "CS"


Leven McLe; Leven Stern McLe;

1 1 1 1

1 £l 1 1

.-s -I



permeability of the shell and its membranes will also be dealt with. There has been some controversy on the subject of whether the eggshell contains any elements of the secreting organ in it, like the decidua of mammals. Von Baer thought not, but the presence of cellular structures has been reported by von Hemsbach; Landois; and Blasius.

The shell-membrane has been studied chemically by Liebermann and Lindwall, who found that it consisted almost entirely of a protein, the percentage composition of which agreed very closely with keratin (Table loa). Krukenberg, alone, on the ground of its reactions, held it to be a mucin. This ovokeratin, which contains four times as much sulphur as the albumen of the egg-white, was found by Morner to include 7 per cent, of cystine, but there are reasons for supposing that this figure is much too low. Nothing is known of the part played by ovokeratin in embryonic metabolism, but, in view of the fact that calcium is transported from the shell to the embryo during the period of ossification of embryonic cartilage, it is not impossible that the sulphur or the cystine of ovokeratin may be made use of in a similar manner to meet the need for sulphur and cystine for the feathers. This will be discussed later under the head of sulphur metabolism (Section I2'7). Morner considered that sulphur must exist in the ovokeratin in other forms besides cystine, for that amino-acid would not account for more than a third of what he found was there. The amino-acid analyses of ovokeratin are placed in Table 1 1 ; they are due to Abderhalden & Ebstein, and to Plimmer & Rosedale, the former by isolation and the latter by the van Slyke nitrogen distribution method. The arginine figure is rather high.

The strength of the shell is clearly an important biological factor : according to Romanov its average thickness is 0*3 11 mm. giving a breaking-strength of 4-46 kilos. The relation between shell-thickness and breaking-strength is a straight line.

The physiological properties of the shell of the bird's egg have been very insufficiently studied. In the last century there was a general impression that the shell possessed a differential permeability and that, while water and other liquids would readily go through the egg-shell and its membranes from outside in, they would not easily pass from the inside to the outside. It is difficult to find how this idea originated; thus Ranke in 1872 attributed it to the younger Meckel, and Ranke's own statement was subsequently copied down


siuaQ ^ ujio^

aioonwoAO :-i3ii3Z



V V t-, S-.

lOM lO t^ t^ CT) J^

'-' <-o 6 6 6 6 ,'-'

(moo) NIHSVD :u3piEqjapqY

CCTlOLOi-iiNar^iNuOO lOCC cc


s3uof jj sujoqso

puB jqosjinBjg :§

qwo.wusABsq 'auioqsQ


NmaxiAOAO M o ^^ o coco m ci

(3uiui3jb) uipaH pue ;-|2e»«cooi6w

jajunjj 3? uspjBqjapqv fn "" "




X3 D.

-3S E £^'




aiDV oiNiaivsAq

aiDv aiNiaiYioad


4. (psjBinSEOo) NHIMiiaiV -OAO : (auusXD) jaujoj/v[ puB pjoi\ib§ buaunoSn]^

" I o CO o « I 6 6 -^

o c< , o

O O CO ^J* c( cj ,1, I CO

CO ■^ CO

c< 6 CO

CI «


CJ 1^ lO

ClO CD 'f





uO CD C^

1 ?-^l 1



' 6 t- ' '


^^(pasTiiBjsiUo) NHwriaiv v q -OAO :u3piBqj3pqy puB g «  jSajj ^ uap]Bqj3pqv 2; =0

t^ CI tJ* — CO

  • (p3sn

-jeisAjo) NraiMnaivoAO luaqjiQ Jg aujoqsQ puB q;jOA\u3AB3q 7g sauof 'sujoqsQ ijqoapnsjg 3g qUOA\U3AB3'7 '3ujoqso

M 01 O CO lO CI CD

>- " O ■* CO " r^ C) r^ CO CI ^

« t}< CO " " ,1^


 : : : :'S o cs : : : : : .S E u S 2 o

V V V „,'rt.yS c ^c3-i33o«CI.3-r;SHMc/,b0vic3 >-"e -^ -i^ n (i:




I I I I I I I I ^1 I I I



O 7J"Ocp o

(pajEinSEOo) NHwnaivoAO


jj buaunoSnjj


3U}3{j 3g UBUideqQ



SjaqsjY ^ auaAsq

c T « 

' rsi

r (3>i^is


UBA) aioonivoAO



(35lXlS UBA)















-oaj aiiH.\\-003

(ajjXis UBA) SNiaiOHd-HiOA

o o o o o , o

CO O OtX> O 7**

eoTj^i. 6 " 6

or-6 "

CO a> ;*•


CO 'f CT)









-a c












^ 1)

t^ CO


Ph sh

C< Tf


T ^



t^ !N



o e<





'in O

01 r-^


M t« 

c< o

« « 


r^ c<




-]pqs) NI1VH3M0A0

(aupsAo) j3ujoj\i

puB uiajsqg

^ uapjBqiapqy


?^ 9

UBA) SNiaioHd-ooa

aaxiw :s3puMoq

^ jauiuijitj

+ o + 11 CO -^coo 1 1

' ' CO o


« o Th« 

LO •+■ 01U3


.s.s H.S s->-g i

>-nil5 j^ O W Dh_3




LNaiMiiaivoAO I i I I I I I I I



I 1 I

I I 1 ?. I I 1 I


(suusXo) uBAT^ng

puB (auBqdojdXai)

1 I I I I I l|l I I I I r I

NHwnaivoAO ^



e-6 i u i •: I 1 1 1 1 1 1 I I I

!r> CO

o ^f


I r.

I I I 1 1^ I I I I




aaxiw :3duiax ■„ I 28 uapieqjapqv


-OAO :s3UJ30H

jg jauiBSuapv

C U OJ OJ o 3 n m


aaxiiM :nfpu3S

to o» a ■* 1^


-IVXOO UH33I -U3AB3T 'aujoqso











osine ... tine ... tidine


'.3 "S

c S


■ptophane identified £ al di-amin


S 6 — t« 








wrongly by Schafer. However, Thunberg in 1902 conclusively demonstrated the error of the belief, and showed experimentally that water would pass through the membranes equally well both ways, though he found that of the two the inner one was the less permeable. In the case of water birds, there is evidence that the shell is absolutely impermeable to water; Loisel, for instance, found that the eggs of the grebe, Podiceps cristatus, and the duck, Anas domesticus, when placed in distilled water absorbed not a trace of it, and gave out no chloride to it over a period of many hours. The eggs of the ordinary hen, on the other hand, increased considerably in weight and allowed some chloride to pass out into the water. At the same time, Loisel found that hen's eggs develop quite normally, at any rate up to the seventh day, even if they are lying in water. This experiment of Loisel's was confirmed by Lippincott & de Puy, who succeeded in hatching chicks from eggs incubated while lying in | in. of distilled water. The differences between these eggs and the controls suggested that the eggs lying in the water not only failed to lose as much water through evaporation as eggs incubated under ordinary conditions, because of the limitation of the evaporating surface, but actually absorbed some water. Trials with rhodamine red and methylene blue demonstrated penetration by these dyes, extending in the former case to vital staining of the embryo. It is known (Rizzo) that the avian egg-shell has many pores (o-86 to 1-44, average 1-23 per sq. mm.).

As regards gases, the only paper is that of Hiifner. Hiifner placed small pieces of egg-shell with their membranes in a diffusiometer, and measured the rate at which gases passed through the obstacle. He found that oxygen diffused through with most difficulty, then nitrogen, then carbon dioxide, and, finally, hydrogen most easily. It may be significant that carbon dioxide would thus appear to be able to escape somewhat quicker than oxygen can enter. But under normal atmospheric pressure the amount diffusing through the whole egg-shell (goose) per second was 2-115 c.c. of oxygen and 0-503 c.c. of carbon dioxide. The diffusion velocity was always proportional to the partial pressure of the gases, and the removal of the inner membrane made no difference at all, suggesting that the principal barrier was the amorphous calcium carbonate layer. It would be very desirable to repeat these observations with more modern methods, and on a greater variety of eggs.


1-6. The Avian Egg-white

The white of the egg is divisible into three portions which have been studied separately by Romanov. The outermost and thinnest layer makes up 39-8 % of the whole and has 1 1'6 % dry solid. The middle layer accounts for 57-2 % and has 12-4% dry solid, and the innermost, thickest, layer is only 3 % and has i4'5 % dry solid*. The chalazae have only once been analysed separately (Liebermann), when the elementary composition of their protein was ascertained. Table 1 2 summarises the results of the investigators who have made general analyses of the white. It is a very watery solution of protein, containing only the most negligible traces of fats and lipoids, but a great many water-soluble substances such as carbohydrate in various forms, protein breakdown-products, choline, inositol, etc. Natural egg-white, according to Rakusin & Flieher, is a saturated solution of ovoalbumen (15-35 P^^ cent.). The water-content does not vary much, but Tarchanov's analyses go to show that the smaller eggs with short incubation-time are wetter than the others. The proteins of the egg-white are believed to be variable in number in the eggs of different birds. In that of the hen, four are known, in that of the crow three, and in that of the dove one only. The egg-white of the hen's egg contains two albumens, ovoalbumen and conalbumen, and two glucoproteins, ovomucoid and ovomucin.

It was at one time thought that there was a fifth, ovoglobulin. Dillner studied it in 1885, and estimated that it made up 0-67 per cent, of the egg-white and 6-4 per cent, of the total protein, but Osborne & Campbell showed that it was simply a mixture of the others in different proportions. This had already been made probable by the results obtained by Corin & Berard, who were able to separate the ovoglobulin into two or three constituent proteins having several different coagulation temperatures (57*5°, 67°, 72°, 76° and 82° C.) and other special characteristics.

Hofmeister was the first to prepare crystalline ovoalbumen, and he published several papers on it. Other workers confirmed his discovery, such as Gabriel ; Harnack ; and Bondzynski & Zoja, but Hopkins & Pincus showed that the albumen so crystallised only accounted for half the protein present in the egg-white. Part of the missing protein was found by Osborne & Campbell to be in the

  • A large number of concentric rings can be seen in egg-white coagulated in situ,

according to Remotti.



[PT. Ill

  • 2 U

> B

c ^

he- O 1) > •O O g :3 §

o M

W a o

tj> CO

Ph O ■ - (u D c u 2 9 o



O ^Z,C^ CD "

, e< ' > . ,„ n

3 " 5-5 S cng ^

>-j >j= ^ to.c e £

Ki cij u .^ bc:^ tS i=!






'JD O — r^ I CO r- I I

o I •* « ' CO



— . O CO LO r^ f^ CO xi LOCO 6 6 6 6 6

o in

o o


^D^D I 4^ f^


? 2 u A ?^ =« 



r^tr> oco(NCOco, "^.cjoc*

0(C)c<oooo 7" -^ p 6666666 6 66

" o 6 6


" oco x^ CO ■* CO




!N CO 0» CO CJ


- - « 


cr> 0!n

CO 0)0 I^ ot

000 CO t^UD

M o -' o 6


rS bn w

C bo Q <fi u OS

(0 D


t, -2 c

3 D W



C3 Q^ ^

S"! K i-i c3 o « 



< 3 "=o o
















r- oi


If I I I I I I I I I I i I I I I i I I ?

OCO" 000000000 0000000 CI N

cocTi^ «a>-<o~ciLOcoyf oito c~,co c~. o o c> o

cococo cocococoo oico coco cocooococococococo

o ^ ^

jn v X . t„ to

u '^


Table 13. Distribution of proteins in avian egg-white




Passeres (singing birds) Turdus iliacus Turdus piliaris ... Anthus pratensis ... Anthus cervinus ... Garrulm infaustus Corvus cornix Corvus nionedula ...

Zygodactyli (2-toed) Dryocopus martins

Accipitres (birds of prey) Strix aluco Falco gyrfalco Falco peregrinus ... Falco aesalon Falco tinnunculus... Astur palumbariiis Buteo lagopus Pandion haliaetiiis

Pullastrae (doves) Columba livia

Gallinae (fowls) Gallus domesticus Meleagris gallopavo

Grallae (marsh birds) Charadrius apricarius Haematopus ostralegus Tringa alpina Phalaropus hyperboreus Totaniis glareola . . . Actitis hypoleiicos Limosa lapponica Numenius arquatiis Numenius phaeopus Fulica atra

Lamellirostres (ducks) Anser segetum Fuligula marila ... Fuligula fuligula Oedemia fusca Clangida glaucion Somateria mollissima Mergus senator . . .

% wet weight


% of total protein in eg5-white






1-78 1-85

2-33 2-09 1-56

1-73 1-58



1-53 1-26

1-55 1-28


2-II 1-25


I "49

i-i6 1-40 1-05 1-03 1-32 1-31 1-26 1-71 1-54

1-57 2-o6

1-45 1-40 1-56

2-00 1-67

OVO albumen 80

80 80

Investigator and date

Osborne & Campbell

(1909), r^

Komori (1920) Needham (1927) Morner (19 12)

— Morner (1912)

12-50 18-25



Table 13 [cont.].

% wet weight


% of total protein in egg-white


ovo- ovo- ovomucoid mucin albumen

Investigator and date

Steganopodes (pelicans) Phalawcrocorax carbo Phalarocrocorax graculus ...

0-2I 0-46

2-o8 — —

Mdrner (191 2)

Longipennes (swallows) Larus cantis Lesiris crepidata ... Sterna macrura ... Sterna hirundo

1-76 1-28 1-36


15-00 — —


Pygopodes (divers)

Podiceps cristatus Colymbus arcticus

2-04 1-88

— — —


form of the two glucoproteins and the other albumen, while part of it was accounted for by the fact that the yield of the crystallising process is not great. Ovomucoid was originally discovered by Neumeister, who called it "pseudopeptone", and first studied by Salkovski and Zanetti.

The investigations of Osborne & Campbell, whose memoir is the best on this subject, give no very definite indication of the proportions in which these proteins make up the protein fraction of egg-white, but they put ovoalbumen at about 80 per cent., and ovomucin at about 7 (see Table 13). Later, Komori estimated that ovomucoid accounted for about 10-5 per cent, of the proteins, and in 1927 I obtained a figure of 7-6 per cent, for the same constituent. Morner, in his extensive study of ovomucoid in numbers of birds' eggs, obtained results from which higher figures emerge on calculation, namely, from 10 to 20 per cent. The only exceptions were the pelicans, which seemed to have very little ovomucoid. The most probable relationship between the proteins is as follows: ovoalbumen 75, ovomucoid 15, ovomucin 7 and conalbumen 3 per cent., but these values are only very approximate, and further work on this point is much to be desired. Leaving out ovomucin, Wu & Ling found that the proportions were as follows (for Gallus domesticus) : ovoalbumen 78-3, ovomucoid 12-3 and conalbumen 9-4 per cent., or 1-34, 0-21 1 and o-i6i gm. per cent, respectively. Certain Russian workers (Worms and Panormov) have described two proteins, anatin and anatidin, in the egg-white of the duck's egg, and three, corvin,


corvinin and corvinidin, in the egg of the crow. It is not certain, however, to which of the well-known proteins of the hen's egg-white these others correspond. Judging from the percentage composition tables in Table lo a, the columbin of the dove's egg corresponds to hen ovoalbumen and to duck anatinin, while duck anatin corresponds to hen ovomucin, but in the absence of definite information the question must be regarded as unsettled, and would repay further investigation.

The minimal molecular weight of ovoalbumen, according to Cohn, Hendry & Prentiss, is 33,800 (Marrack & Hewitt suggest 43,000), and its percentage composition is seen in Table 10 a; the best analyses are probably those of Osborne & Campbell, who give an account of its general properties. It has been further analysed by several workers who have determined the proportions of its constituent amino-acids, and whose results are seen in Table 11. The hydrolyses of Osborne, Jones & Leavenworth; Osborne & Gilbert, and of Abderhalden & Pregl were all done by acid, but those of Hugounenq & Morel and Skraup & Hummelberger were alkaline, the former using baryta. The figures agree accordingly, and all that can be said of them is that for purposes of calculation the amounts of amino-acids must be taken as minimum in each case. Attention may also be drawn to the less complete analyses of Chapman & Petrie and Hugounenq & Galimard and to the analysis of mixed egg-white proteins by Plimmer & Rosedale, using the van Slyke technique. The large amounts of hexone 4Dases found by them contrast with those found by the remaining workers, using direct isolation, and if this is not due simply to difficulties of technique it may lead us to expect a high content of hexone bases in conalbumen and ovomucin when they come to be analysed.

In Table 10 a the results obtained by Gupta on the hydrolysis products of ovoalbumen are given (see also Rudd). It is noticeable in them, as in the analyses of ovoalbumen itself, that they contain a high proportion of sulphur, though not so much as ovomucoid. The spontaneous evolution of hydrogen sulphide by egg-white on standing has long been known, and was made the subject of a paper in 1893 by Rubner, Niemann & Stagnita, who found that 100 gm. of egg-white gave off when boiled with water 10-7 mgm. of HgS. Hausmann later decided that its source must be some labile sulphydryl grouping in the ovoalbumen molecule. In 1922 Harris


observed that raw egg-white was quite non-reactive towards the nitroprusside test for sulphydryl groups, but that immediately upon coagulation by heat it became vividly reactive, and gave an intense purple colour. This change only took place in conditions where denaturation of the protein was involved, and Harris suggested that this treatment might unmask a thiopeptide linkage or some similar arrangement which by hydrolysis or keto-enol transformation would give rise to an active sulphydryl group in the resulting metaprotein molecule. Later, Harris found that only 14 per cent, of the sulphur in ovoalbumen could be accounted for as cystine, so that some unknown sulphur compound must be present in considerable quantity, and an exactly similar finding was later reported by Osato for the egg-membrane protein of the herring. The cystine recoverable from serum albumen, on the other hand, accounted for 86 per cent, of the sulphur there. The possibilities of these facts with relation to the metabolism of the embryo have not yet been explored. Philothion, according to de Rey-Pailhade, exists in the egg-white of the hen but not in that of the duck.

The principal investigation of ovomucoid is that of Morner. He had previously discovered that percaglobulin, a protein extracted from the unripe ovarial fluid of the perch {Perca fluviatilis) would precipitate ovomucoid from its solution. With this reagent he made an examination of a wide variety of birds' eggs, in order to study the distribution of ovomucoid. By direct estimation he found it to be present in all the eggs he studied, but it seemed to exist in two sharply distinguished forms, one which would give a precipitate with percaglobulin, and another which would not. Thus the hen [Gallus domesticus) with i -46 per cent, of ovomucoid gave a highly positive reaction, but the hawk {Astur palumbarius) with i -45 per cent, gave none at all. Preparations of ovomucoid from the two varieties of egg-white (see Table 10 a) did not show up the existence of two obviously different chemical individuals, and it was concluded that the preparations were in each case mixed with a small amount of the other substance. Moreover, of the egg-whites which gave a positive reaction with percaglobulin, some contained ovomucoid precipitable with Esbach's reagent (e.g. Clangula glaucion and Somateria mollissima) and others contained an ovomucoid which could not be so precipitated (e.g. Gallus domesticus and Podiceps cristatus). It is quite uncertain how many of the effects observed by Morner are due to



[PT. Ill

physical and colloidal rather than to chemical differences, and the whole question should be reinvestigated. There seemed to be no special significance in the distribution of the ovomucoid which was precipitable by percaglobulin ; thus it was present in the accipitres, grallae, lamellirostres, longipennes and pygopodes, but not in the passeres, zygodactyli, pullastrae and steganopodes. As for the fowls, it was present in the eggs of the hen and pheasant, but not in those of the guinea-fowl. Morner was inclined to agree with Milesi's view that ovomucoid did not exist as such in the natural egg-white at all.

Table 14. Variations in properties of avian egg-white.

Coagulation point

Consistency and

of the

egg-white m

colour of coagulum

degrees Fahrenheit



Hard white




Pretty firm, bluish



Soft, white, translucent .


Missel thrush

Soft, transparent



Firm white



Soft white



Pretty firm, greenish, transparent


Golden-crested wren

Soft, bluish, semi-trans parent


As has already been observed, Sir Thomas Browne was one of the first to note that the coagulated egg-white of the gull's egg was quite different in consistency and translucency from that of the hen's egg. In 1863 Davy collected some data on these points, which are shown in Table 14, and Tarchanov devoted much time to the question in the 'eighties of the last century. He found that the whites of many kinds of eggs would not coagulate in the ordinary way on boiling, but either remained liquid and transparent or else set to a watery translucent jelly. This he called " tataeiweiss ", and as he went on to examine the distribution of this property he found that it was associated with the hatching quality of the bird in question. Thus all nidifugous birds, whose chicks are born fully feathered ("downy") and soon leave the nest, had eggs with ordinary egg-white, but all nidicolous ones, whose chicks are hatched as "squabs" or naked and weak, and have some development yet to complete, had eggs with uncoagulable or transparent egg-white. Thus the sand-martin, linnet, finch, thrush, canary, crow, dove, rook, nightingale, robin, starling (roughly passeres and pullastrae), all had tataeiweiss] while the hen,


duck, goose, guinea-fowl, partridge and corncrake had ordinary white. This classification agreed roughly with Davy's high and low coagulation points for the egg-white, and corresponded on the whole to Morner's two classes, the former having ovomucoid not precipitable with percaglobulin and the latter having the precipitable variety, but to this there were some exceptions; thus the plover's ovomucoid could be so precipitated, but its egg-white was tata and it yet produced fully-feathered chicks. It was, however, the only exception to Tarchanov's generalisation, (It should be explained that the word tata was derived from the name of Tarchanov's small daughter.) Tarchanov found that tata egg-white was about 3 per cent, richer in water than the other kind, a conclusion which Morner's later analyses did not confirm. He also said that it was alkaline to litmus, but became less so as the tata eggs developed. This agrees with the later classical work of Aggazzotti on the reaction of the eggwhite of the hen's egg during its development. Tarchanov reported that tata egg-white could be made to coagulate at ordinary temperatures by the addition of a little potassium sulphate, and that it would itself coagulate if the temperature was raised well above the boilingpoint of water. It was, he said, more easily digested by enzymes, it putrefied more readily, and during development it changed into a form resembling ordinary egg-white. He made some studies on its secretion by the oviduct of these birds, and was the first to perform the experim.ent of putting a ball (in his case a lump of amber) at the top of the oviduct and seeing it emerge at the bottom with layers of egg-white and a shell secreted around it. The change during development from tata to ordinary egg-white Tarchanov found he could imitate by bubbling carbon dioxide through the original white, after which it would coagulate in the usual way. On the other hand, he found that if he soaked normal hen's eggs in a 10 per cent, solution of alkali the white took on the properties of tata egg-white, and became just like the glassy material in the sand-martin's egg. He suggested some relation between these phenomena and the alkalialbuminate of Lieberkiihn, but did little to determine its chemical relationships. He was unable to get any development in the case of hen's eggs soaked in alkali.

In 1 89 1 Zoth took up the whole question of tataeiweiss once more. He was led to do so on account of some researches which he had been making on the effect produced on serum-clotting by various


concentrations of alkali, and which showed that the clot could vary very greatly in its properties, from opacity to almost perfect transparency, for instance. Tarchanov had decided that the transparent coagulum of the nidicolous egg-whites was not to be identified with that produced by sodium or potassium albuminate, but Zoth succeeded in showing that the differences were not sufficient to distinguish them. Zoth fully confirmed Tarchanov's finding that ordinary egg-white could be made to pass over into nidicolous egg-white by treating it with i o per cent, potash in the cold for ten days, and was able to explain all the differences between tataeiweiss and alkali ovoalbuminate as due to variations in the amount of alkali present, or rather the amount of cation as compared to anion. It is most unfortunate that we have no detailed ash analyses of the egg-whites of nidicolous birds, for, as will later be seen, the egg-white of the hen has rather more total anion than total cation, and this relationship might be expected to be even more strongly marked in the case of nidicolous egg-white, perhaps^ indeed, as much as to counterbalance the excess of cation over anion in the yolk. There can be no doubt, however, that the egg-whites of nidicolous birds are relatively richer in alkali than are those of others, and it is this, combined with their different water and total ash content, which causes the albumen to coagulate differently from those of others. Thus, if 5 c.c. of filtered egg-white from a fresh hen's egg be put in each of three small Erlenmeyer flasks, 2 c.c. of water added to A, 2 c.c. of 0-89 per cent. KOH to B, and 2 c.c. of a mixture of equal parts 0-89 per cent. KOH and 0-66 per cent. NaCl to C, the coagulum in A will be the usual white, thick, solid and opaque mass, while the other two will be transparent like tataeiweiss, slightly opalescent, more or less liquid, and Cmore opalescent than B. It would be interesting to reinvestigate the whole question anew in the light of recent knowledge and technique.

Another curious effect was noted by Melsens and Gautier. Melsens found that, if a stream of carbon dioxide, hydrogen, nitrogen or oxygen, was passed through dilute egg-white, or if it was shaken violently, a precipitate of fibrous membranous shreds was formed. Gautier observed that about i -5 per cent, of the protein was thus changed; he filtered it off and determined its elementary composition, which showed nothing remarkable. He concluded that a protein which he called " ovofibrinogen " existed in the egg-white, and even suggested that an " ovo thrombin " was present to turn it into "ovo


fibrin". He apparently thought that the ovofibrin was incorporated without change into the substance of the embryo. The subject has not received any attention since the time of Gautier, but it is probable that this phenomenon is explained by the work of Young; Dreyer & Hanssen and others, on the high instability of protein solutions.

Peptones were reported by Reichert to exist in fresh egg-white.

Wu & Ling have recently studied the coagulation of ovoalbumen by strong mechanical agitation. The fact that conalbumen is not coagulable by such means gave them a method of estimating it in egg-white. Thus they obtained the following figures for Gallus domesticus egg-white:

Nitrogen Total N (ovoalbumen + conalbumen 4- ovomucoid) 1-71 gm. %

After shaking (conalbumen + ovomucoid) ... 0-372 ,,

After shaking and heating (ovomucoid) ... 0-211 ,,

Coagulation of ovoalbumen by shaking was not separable into two stages (denaturation and agglutination) like that by heat or alcohol. The isoelectric point of the protein was the most favourable for shaking coagulation (/?H 4-8) and the Q^^q of the reaction was i-g. Piettre has published a method for separating the proteins which involves the use of acetone.

The relationships between the avian egg-white proteins have been the subject of some interesting immunological work. The earliest investigators who crystallised ovoalbumen found that perfect freshness was necessary, for at room temperature the crystallisable protein gradually turns into a non-crystallisable one. Bidault & Blaignan found that this process could be arrested by placing the ^gg at 0°. Sorensen & Hoyrup suggested that the protein formed was conalbumen and wished to look upon the latter as a product of ovoalbumen. Hektoen & Cole, however, first showed that though ovoalbumen was distinct from the serum albumen of the hen immunologically, conalbumen was not, and then went on to demonstrate that during the loss of crystallisable ovoalbumen which takes place as the egg ages, there was no corresponding increase in conalbumen. We must therefore look upon the latter as probably identical with the serum albumen of the adult: and perhaps only present in the ^gg owing to the inefficiency of the oviduct.

The analyses of ovomucoid and Eichholz's ovomucin, as well as the fragmentary one of conalbumen, will be found in Tables i o a and 1 1 . Willanen found that ovomucoid was much more susceptible



to hydrolysis by pepsin than by trypsin (see later under enzymes and antitrypsin). For its properties see the papers of Morner and Neumann. Both the glucoproteins have twice as much sulphur as ovoalbumen. Their carbohydrate content has been the subject of a great amount of discussion and experimental work. Berzelius was the first to draw attention to certain similarities between the breakdownproducts of sugars and proteins when acted upon by boiling acids. In 1876 Schiitzenberger asserted that the ovoalbumen molecule contained a carbohydrate group, basing his views on positive results with Trommer's test after total hydrolysis. In later years a number of workers supported the view that the carbohydrate was glucose, using in different cases methods of varying reliability, e.g. Krukenberg in 1885, Hofmeister and Kravkov in 1897, and Blumenthal ; Blumenthal & Mayer and Mayer in 1898 and 1899. Spencer and Morner, however, failed to get any evidence of a carbohydrate group after hydrolysis, and reported their negative results in 1898. Weiss, about the same time, thought he could identify a methyl pentose among the hydrolysis products, but he was never confirmed. Seemann was the first to announce that the carbohydrate was glucosamine, and his discovery was quickly confirmed by Frankel and Langstein. These later workers began to attempt quantitative estimation of the sugar, and their figures are given in Table 15. Pavy, using the then recently discovered osazone technique, made a study of a variety of proteins, and showed, as might be expected, that the yield from ovoalbumen was always greatly less than from ovomucoid. Eichholz obtained glucosazone from ovoalbumen, ovom.ucoid and ovomucin, but not from either serum albumen or casein. On the whole, it is most likely that ovoalbumen contains extremely little glucosamine, and the figures recorded in the literature for this are probably due to contamination with ovomucin. This is the view of Osborne, Jones & Leavenworth, for neither they nor Osborne & Campbell obtained any glucosamine from their very carefully purified ovoalbumen. Komori has prepared from ovomucoid, and Frankel & Jellinek from ovoalbumen, polysaccharidelike substances which they regard as the prosthetic group containing all the glucose. Following this up Levene & Mori have prepared a trisaccharide containing glucosamine and mannose from egg-white. Ovoalbumen contains 0-26% of this substance, coagulated egg-white 1*9%, and ovomucoid 5*1%. According to Levene & Rothen the

SECT. l]



molecule consists of four trisaccharides each containing one molecule of glucosamine and two of mannose.

Table 15. Ovoalbumen and ovomucoid glucosamine content.

Hen {Callus domesticus)

Hofmeister (1898)

Seemann (1898) ...

Langstein (1900)

Willanen (1906) ...

Pavy (1907)

Samuely (1911) ...

Neuberg & Schewket (1912) ...

Zeller (1913)

Needham (1927)

Abderhalden, Bergell & Dorpinghaus

Neuberg (i 901) ...

Blumenthal & Mayer (1900) ...

Izumi (1924)

Tillmans & Philippi (1929) Turtle ( Thalassochelys corticata)

Takahashi (1929)

Bywaters (1909) Seromucoid Pavy (1899) Ovomucoid







3-4 3-7



15-0 34-9

I9-5-22-3 21-7 29-4 24-0 33-7 II-5

7-4 6-2

Glucose % by hydrolysis with

Osseomucoid ... Tendon mucoid Ovoalbumen ... Serum globulin

10 % KOH

I2-0 12-3


33 2-6 2-8

5 % HCl 15-9


lo-B II-6



The free carbohydrate of the egg-white has also received a great amount of attention from an early date. In 1846 Winckler isolated a quantity of a sugar (4 grains) from the egg-white of the hen's egg, and identified it as lactose. Physiologists", he said, "will be able to tell me whether this is of importance for the embryo or whether it was some abnormality." The observation has not since been repeated, and it is in the highest degree unlikely that any lactose was ever in an egg, unless the diet of the hen was a very unusual one. Budge and Aldridge soon were at work on this subject, the former concluding that the carbohydrate was glucose, but suggesting that it might form a disaccharide during development, and the latter making no concession to Winckler. The presence of glucose was afterwards abundantly established by the work of Barreswil ; Lehmann ; Meissner ; Salkovski ; and Pavy. Later many quantitative estimations were made, and these are collected together in Table 16. The older figures for free carbohydrate may be regarded as fairly








• ?^



































5 o


  • - 1^ bo

iS S


- bD

■ bo



^ o bO

1 c


O 8




in N I CO f^

coco I I I I I

^ I U-)Co

CO Lo « >ti

CO I CO O ■-"

05 CO t\ o)

►1 ' M Cr^ C4

CO OsCTi I «- O CO I I^ CT) IT) CO "J^ 04

in o



> rt g

G u O O










bo bo

y bO I) w

o o o o o o o

O f< Crv o, JN, ^ Cl C< i~i -"i -^ --i IN >^


o <o Cr^l^

s « 

s 5



« S P

^ S; i>H '^ .^3 ^ ^ ti. g '^ tts tt, O O CO

" -^ "^ "^ « s . ^

O G ^

Oh ? £



i~ iS it- V









C „ O u fi „ o - O 3 W "

"bO bo



00000^ six IN O O5 tx"5

•S'S § 3

o o ,0 ,0 ,0 00 o

U "^ -« s ^

? 3 3 eg

« ,a,» K c a

S S ^55 ^ CJJ CJ Cj

Ci u 55 1^ [^ tc, fcc, -^ cq Q,





>*i •-<


S2 3 •



■2 ?<> :



C 3i






•^ 0^



• 2


J2 a

^ a -<a.




§ ■'2






^ ^

■« a « 








Chara Haem Tring


trustworthy, but not for combined sugar, in view of the demonstration of Holden that all copper-reducing methods are seriously interfered with by the presence of amino-acids and protein breakdown products. No method at present in use gives satisfactory results in those conditions, but the most reliable is that of Hagedorn & Jensen. No estimations of total carbohydrate in egg-white alone at present exist, but there is a single figure for glycogen due to Sakuragi. Morner found no evidence of fructose, pentoses or maltose.

A curious phenomenon : the fluorescence of egg-white has been reported by van Waegeningh & Heesterman, but it only occurs if the egg is not perfectly fresh, and is therefore probably not physiological.

I '7. The Avian Yolk

The vitelline membrane was investigated by Liebermann in il who found that it consisted almost exclusively of keratin. This he purified, and, having freed it from ash, made an elementary analysis of it, which is shown in Table loa. Some experiments which demonstrate the peculiarities of the vitelline membrane have been devised by Osborne & Kincaid. They found that a fresh yolk floated into distilled water, o-g per cent. NaCl solution, or glycerol, behaved exactly like a red blood corpuscle in that it swelled up and burst in the former, and shrank to a corrugated globe in the latter, while in the isotonic salt solution it remained unchanged. But with other treatment, nothing took place which corresponded to haemolysis. If the yolk was put into lo per cent. NaCl solution, it did not shrink, as had been expected, but swelled up, owing to the penetration of the saline and the consequent osmotic pressure due to the dissolving of the vitellin in the saline. This showed at once the scleroprotein nature of the membrane and its impermeability to vitellin even when in solution. The membrane is also impermeable to phosphatides and fats dissolved in ether, for if a yolk is put into ether it sinks and swells, so that the upper pole is distended by an accumulation of deeply pigmented ether. But until the yolk bursts, as it eventually does, not a trace of pigment or other substance passes out into the ether, and the same results were found with chloroform and carbon disulphide. In alcohol, on the other hand, there is no swelling, for the alcoholic solution of phosphatides and other bodies can pass out


through the keratin membrane. It would be very interesting to make a more extended study of the osmotic properties of the vitelline membrane (see in this connection Section 5*6).

The yolk of the egg was investigated earlier in the modern period than the white. We may pass directly, excluding the curious analysis of the eggs of Struthio casuarius by Holger in 1822, to the papers of Gobley, which appeared from 1846 to 1850, and which, with those of Valenciennes & Fremy from 1854 to 1856, still remain models of embryo-chemical work. "John, a German chemist," said Gobley, "appears to have been the first to occupy himself with serious researches on the yolk of the egg. The chemists who preceded him considered it as made up only of water, albumen, oil, gelatine, and colouring matter; such was the opinion of Macquer, Fourcroy, and Thomson. John concluded from his experiments, which he published in 181 1, that the yolk was composed of water, a sweet yellow oil, traces of a free acid which he thought was phosphoric acid, and a small amount of a brownish red substance, soluble in ether and alcohol. Besides these he found gelatine, sulphur, and a great deal of a modified albuminous substance." Gobley referred also to the work of Prout, of Chevreul, of Berzelius and of Lecanu, who discovered the presence of cholesterol in yolk in 1829.

Gobley himself found in the yolk nearly all the substances which we now know to be there. His own list of them ran as follows :

1 . Water.

2. An albuminous matter, "vitellin",

3. Olein.

4. Margarin.

5. Cholesterin.

6. Margaric acid.

7. Oleic acid.

8. Phosphoglycerilic acid.

9. Lactic acid.

10. Salts such as chloride of sodium, chloride of potassium, chlorhydrate

of ammonium, sulphate of potash, phosphate of lime, and phosphate of magnesia.

11. A yellow colouring matter and a red colouring matter.

12. An organic substance containing nitrogen, but which is not al buminous.

Most of the constituents of egg-yolk may be recognised under this






V se

9 3't=

Z bo


■^ ^



V a

►S '-°

-G „

i^g^l J35 2 5 w

C^ ^^ O Ph ^ Ph C/D 1-1

0~i c " — o

s ^


1 •: 1 1 1 1 1


u bo


a> CO lo CO oi

1 1 M 1^ C4 O

1 ' io 'i' LO(-b io



O S"! S?

L4 U U

c 3r,

CO CO coco CO 6 6 « 6 6

CO O CO CO -^ ^ C< C< C< C4 W CO

O O "

■*Oc<r^coO'-'OOooo'-' M Oco lO" 01 coc<co cox> a in CO ■^ ^cb c) M M ciiT) ■^ CO CO CO

O O yD CI C(



eg 1^


0-2 '^ "Phco " "

^ 0-7^T3 « 


m lo^d t^ lO lO'Xi t^ t^t^ t^ t^to ^

o >p

01 ►H

r^ « O ^

M CTi CO lOCO f^

« 6 COti> 131 "

CO CO CO CO c< c<

w O

CD lO CO ^ lO lo in t^

o CO a> Ti-co lo CO ■* Tj- CO o

O ■* 1^ CO r^cp LOCO ^ CO t^ ^ ■^co r^ 7

M^r^OCOi-CTiLO rhcb CTi f^ f^ f^ CT) ->


o O

'-30331'iU -^aj

QOhOKffi K

O «  . C

> y


CO " O O lO lO CT) en lO C"!

6 6 ci 6 io

N " " c< "

« „' en

O « ^ « 


o E ~'- X^


o f^

u ^— '


s a


1/3 o

c ^ c _

CO 7*" r^ O

1X> ^ C5 CT)

CO ^ O K)

6 6 « 6

in ■* irs jj

o in o


(i ti t£i

O) 6 Cl CD ^ uO Tf ■*


o o o o

O -1 O CD

(i> f^cb f^

lO lO lO lO


3 O



  • *


bog g

C C u O fc^ ^^ aj

i3t)cubDl'-3 a



be 3 u


2 K


o be

E ^


2 -S



old-fashioned terminology. Gobley made many quantitative observations on the various substances, and his figures are given in Table 17, which sums up all the analyses that have been made of the yolk in the eggs of birds. The original discovery of vitellin was made by Dumas & Cahours, but Gobley was the first to make an extended study of it. His elementary analysis is given in Table 10 a. He knew that it contained both sulphur and phosphorus. Gobley was able to isolate oleic and margaric (palmitic) acids from the fat fraction of the yolk, but, unlike Planche twenty-five years before, he got no stearic, and Kodweiss, one year later, reported its presence under the impression that it had not been found before. Gobley, however, was easily able to repeat Lecanu's discovery of the presence of cholesterol, and made a remarkable examination of the lipoids. "These viscous materials", he said, "appear to have been considered by John as not being of a fatty nature at all. They form the most interesting part of the yolk; they contain all the phosphorus which exists there in considerable quantity." He analysed the glycerophosphoric acid which he obtained from the lipoid, which he named "lecithin", and speculated as to the significance which it might have for the growth of the embryo. He also recognised that fatty acids and nitrogen were present in the viscous matter.

Ten years later Valenciennes & Fremy made a further examination of the yolks of a large variety of eggs with special reference to \itellin. It was they who discovered substances very similar to vitellin in the eggs of reptiles and fishes; these they named the ichthulins. As regards the eggs of birds, they contented themselves with confirming the results of the previous in\'estigators, but they regarded vitellin as having practically the same constitution as fibrin, on the grounds of elementary composition only. At the same time, they held it to be a different compound because it would not, like blood fibrin, decompose hydrogen peroxide.

If Table 1 7 is examined, it will be seen that the yolk is much drier than the white in all birds' eggs examined, having only about 50 per cent, of water as against the 85 per cent, of the latter. On the other hand, the percentage of fatty substances and lipoids is much higher, being just about double the amount of protein, whether related to wet weight or to dry. It is noticeable from the analyses of Tarchanov that the yolks of eggs from nidicolous birds having a short incubation time are about 10 per cent, richer in water than yolks from the eggs


of nidifugous birds. This must imply that the greater requirement for nutrient material in the latter case has, as it were, packed the fat tighter into the yolk. Exactly the same relationship is brought out from the figures of Spohn & Riddle, who compared the pigeon which hatches out as a squab with the hen which hatches out as a fully-feathered chick. Spohn & Riddle's analyses are the only complete ones we have for a nidicolous egg, and bear clearly the same relationship, for there is less protein and less fat, relatively, in the pigeon's egg than in the hen's. The ash content and the amount of non-nitrogenous extractive substances seem, however, to be slightly higher in the latter case. Langworthy's figures were all obtained from the eggs of nidifugous birds, and they show a great similarity among themselves. More delicate consideration, of course, reveals differences according to breed in the hen's egg, e.g. the figures of Pennington and his collaborators, but these are of a comparatively minor order.

The most interesting analyses are those of Spohn & Riddle. They compared the egg of the jungle-fowl, which is supposed to have been the evolutionary ancestor of the domestic hen, with averaged figures for hen's eggs of various breeds, and, as is evident, there was a very close agreement. They also analysed the white yolk as distinct from the yellow yolk of the hen's tgg. When the yolk begins to be formed in the ovary of the hen, it is white and not yellow, and not until the critical point in its maturation is reached, when its growth-rate completely changes, does it begin to store lipochrome pigment. This change in growth-rate, which has been observed by other workers as well as Riddle (e.g. Walton), will be dealt with in more detail in the appendix on maturation. Von Hemsbach, in a paper on the milky or white yolk of the birds, in 1851, suggested that the corpus luteum of mammals corresponded to the yellow yolk of birds, and that the mammalian ovum having been shed out of the ovary into the Fallopian tube and uterus, the fats and lipochrome pigment were laid down in the Graafian follicle instead of around the white

  • 'ovum". Von Hemsbach also supported the view already mentioned

that the shells of avian and amphibian eggs corresponded to the decidua of mammals. He laid stress on the work of Zwicky and Gobel, who had investigated the pigments of yolk and corpus luteum, and had thought them to be identical. This subject will be referred to again under the head of pigments.




In the fresh egg, as laid, the white yolk occupies a central position, and is surrounded by concentric layers of yellow yolk. But as a kind of cylindrical prolongation of the white yolk reaches to the surface of the vitellus underneath the blastodisc or germinal spot, the white yolk must be considered the first food of the embryo, and, until its composition was determined, it was not possible to say what sort of nutrient environment the embryo possessed in the very early days of development, although the composition of the yellow yolk would give this for the later period. The histological differences between white and yellow yolk had been known for a long time (see Purkinje ; His ; Leuckart ; Klebs ; Dursy ; Strieker ; and Virchow) but Riddle and Spohn & Riddle were the first to approach the subject chemically. Their figures showed that the white yolk much the more nearly approximated to the contents of invertebrate eggs with holoblastic cleavage, and living undifferentiated tissue generally. Instead of 45 per cent, of water, the white yolk had 86 per cent., instead of 15 per cent, of protein, it had only 4, and instead of 25 per cent, of fat it had only 2. Thus in its water-content, it was much more like {a) eggwhite and {b) the young embryo itself than like ordinary yolk, while instead of having twice as much fat as protein it had twice as much protein as fat. These data are extremely interesting in view of the facts that are known about the sources of energy made use of by the embryo during its development. Although by far the greatest proportion by weight of substance combusted during embryonic life is fat, yet, in the early stages, the embryo undoubtedly gets its energy preponderantly from protein and carbohydrate (see the whole of Section 7). The percentage of non-nitrogenous extractives did not differ much between white and yellow yolk in the experiments of Spohn & Riddle, but it would be very interesting to know the relative amounts of carbohydrate, and analyses to discover this should certainly be done. Again, the yellow yolk contained eight times less ash than the white yolk, a finding which acquires considerable significance from the fact that, if the ratio inorganic substance/organic substance in the embryonic body is plotted, it is seen to descend steadily from the beginning of development (see Fig. 249). Moreover, as Mendeleef has shown, early embryonic cells contain twice as much electrolyte as those of later stages (see Section 5"8). The amount of phosphatide in the yellow yolk, furthermore, was ten times that in the white, a significant difference; for, as Plimmer & Scott have shown,


one of the main functions of the phosphatide is in furnishing phosphorus for the embryonic bones during the period of ossification, a requirement which is not present in the earher stages of the development. The histochemical work of Marza, who compared the white and yellow yolk following the method of Romieu, is in agreement with this, for he found the elements of the yellow yolk to be richer than those of the white. (See Plate X.)

1-8. The Avian Yolk-proteins

As regards the protein, vitellin (Tables 10 a and 11), several interesting points are to be observed. The best elementary analyses of ovovitellin are probably those of Osborne & Campbell. After its discovery by Dumas & Cahours, Gobley, and Valenciennes & Fremy, it was studied by Hoppe-Seyler, and now for the first time with special reference to its position in the classification of the proteins. Virchow had some time before then suggested that the yolk-platelets, familiar to histologists, contained lecithin, and there had been some doubt as to their nature. Valenciennes & Fremy had opposed the view that they were crystals, basing their view on Sennarmont's work, but Radlkofer and Hoppe-Seyler returned to the crystal theory. Hoppe-Seyler believed that vitellin contained no phosphorus, but that what appeared in the analyses was due to contamination with lecithin. This view was supported also by his assistant, Diakonov, who contributed to the Med.Chem. Untersuchungen one of the earliest investigations of phosphatide. But at the same time Miescher obtained from the yolk of the hen's egg a substance containing a great deal of phosphorus, and possessing certain of the properties of a protein. This he believed to be nuclein. "It is interesting ", he said, "in relation to the origin of nuclear substance, that the nutrient yolk contains ready-formed nuclein in significant quantity." At this time, then, the proteins of the yolk were believed to be ovoglobulin (for so Hoppe-Seyler called the vitellin of the earlier workers) and Miescher's nuclein. Miescher himself identified his nuclein as a constituent of the white yolk of the histologists, but he noted that the hen's egg seemed to have no xanthine in it.

Lehmann, Schwarzenbach and others, however, did not agree with this classification, and regarded vitellin as a mixture of albumen and casein. They did so not on the grounds of its containing phosphorus, but because they found that rennin would completely coagulate it from its pure solution. But this attitude did not prevail.


and the word " nucleovitellin " became general, until Kossel in 1886 found that, if vitellin was really a nuclein, it differed from all other such substances by giving no trace of xanthine after acid hydrolysis. On the other hand, true nuclein, he found, was present by the tenth day of development. Hall and Burian & Schur, Bessau and von Fellenberg confirmed this absence of purines from the fresh egg. In more recent times, Sendju and Mendel & Leavenworth have found exceedingly small amounts of true nucleoprotein (2 and i*6 mgm. per cent, respectively wet weight) in the hen's egg (by purine bases), and Plimmer & Scott, and Heubner & Reeb have done the same (by phosphorus analysis) . Shortly after Kossel's work, Milroy found that vitellin gave a biuret test though no Millon, and materially differed in nitrogen and phosphorus content from any of the nucleoproteins, while, at the same time, Miescher admitted that he could not isolate any purine bases from his "nuclein" in the hen's egg. Levene & Alsberg next investigated the manner of breakdown of vitellin, finding the substance they named " paranuclein " after digestion with pepsin, and "avivitellic acid" after hydrolysing with ammonia. The elementary composition of these substances is given in Table 10 a, from which it could be seen that the increasing phosphorus content implied the presence of phosphorus as an important constituent of the original molecule. Six years later Levene & Alsberg ascertained the amino-acid distribution (see Table 11). They pointed out the significance of the high proline figure, in view of the task of haemoglobin synthesis which the young embryo has before it. Abderhalden & Hunter and Hugounenq undertook a like investigation in the same year, from which a striking similarity between the amino-acid distribution in vitellin and casein came to light, especially as regards the high proportion of leucine and glutamic acid. They drew attention to the similarity in physiological requirements as between the "erste Nahrung" of chick and mammal. The historical associations of this discovery have already been referred to (see p. 53). It was at this time that Neuberg, and Blumenthal & Mayer reported the existence of glucosamine in the vitellin molecule, two observations which stood together in isolation, until in 1929 Levene & Mori isolated from egg-yolk the same trisaccharide which they found to be present in ovoalbumen and ovomucoid and which has been referred to above.

It was not until the paper of Bayliss & Plimmer in 1906 that the



/ kj^EHHHH^^^HHI

' '^ "


Stain, haemalum-eosin: magnification, gxA: prepared and microphotographed by Dr V. Marza. The stratification of the yolk into white and yellow is beginning.


nature of vitellin really became clear. They subjected casein and vitellin to the action of trypsin, and studied the time taken under varying conditions for the phosphorus to be split off into soluble form. Ovovitellin, they found, was much more slowly digested than casein, for after 36 days only half of its phosphorus had been made soluble, whereas after 2 or 3 days a large percentage of the casein phosphorus had gone into solution in inorganic form, and most of the rest was present in water-soluble organic combination, i per cent, soda, however, would bring aU the phosphorus of casein into solution in 24 hours. BayUss & Plimmer concluded that ovovitellin and casein were both phosphoproteins, as distinguished from nucleoprotein, where the phosphorus would be present in the prosthetic group and not in the protein itself Plimmer & Scott later found that ovovitellin behaved in the same way to soda. This reaction served to distinguish between phosphoproteins and nucleoproteins, for all the latter, it was found, were stable to alkali and easily split by acids. From the phosphorus distribution in the unincubated hen's egg, Plimmer & Scott concluded that vitelHn accounted for at least 30 per cent, of the phosphorus, and this led them on to their investigation of the changes which take place in the different phosphorus fractions during the development of the embryo.

The distribution of phosphorus-containing compounds in egg-yolk, as Plimmer & Scott found, makes a very different picture from that of any other tissue. Their summary is shown in the accompanying table (18). It would be extremely interesting to investigate the phosphorus distribution in the white yolk, which at present is altogether uncharted.

Table 18. Phosphorus in per cent, of the total phosphorus.









Lecithin P ...





Total water-soluble P





Water-soluble inorganic P





Nucleoprotein P





Phosphoprotein P ...





Total protein P





The third of these fractions i

includes the phosph

orus of all unstable water-soluble corn


Steudel, Ellinghaus & Gottschalk have recently found that vitellin behaves towards pepsin exactly like casein. The rate of increase of titratable COOH groups during the digestion far exceeds that of NH2 groups, reaching a maximum about the fourth hour. The




[PT. Ill

Table 19.

Per cent, of dry weight

Total N



N/P rati

mer's figures.

Ovovitellin ... ovolivetin ... Ovoalbumen ... Casein ...

... 15-29 15-12

••• 15-51 ... 15-30

i-o o-i o-i

15-3/1 151-0/1


Levene & Alsberg's figures.

Avivitellic acid ... 13-13

Swigel & Posternak's figures.


Swigel & Posternak's figures.

Hydrolysis of ovotyrine b^ (% ) Pyruvic H3PO4 acid NH3 Arginine Histidine Lysine

12-00 1-60 4-90 0-62 0-70 0-75


Ovotyrine a^ ...

... 10-87




Ovotyrine ^^ ...

... 11-33




Ovotyrine /Sj ...

... 10-92




Ovotyrine 71 ...





/-serine 7-90

Table 20. Nitrogen distribution.

Plimmer's figures (1908).

Per cent of dry weight

Ovoalbumen Casein

Ovovitellin . Ovolivetin .



15-51 15-30 15-29 15-12




1-52 0-84 0-75


N 0-29 0-22 0-25 0-32



3-30 330 3-84 3-29

Monoamino N 10-58 10-36 10-26 10-76

linkages must break, therefore, between a carboxyl group and proline, tr-yptophane, histidine or arginine.

Weyl ; v. Moraczevski ; and Gross were the first to describe the properties of the egg-yolk proteins, but the standard account is that of Plimmer, who in 1908 identified two yolk-proteins, ovovitellin, and ovolivetin. Ovovitellin, according to his analyses, contained 1*0 per cent, of phosphorus, but ovolivetin only o-i per cent. He was usually able to isolate far more of the former than the latter, but in some experiments the yield seemed to be nearly equal. Livetin was

SECT. l]



soluble in water as well as 10 per cent, salt solution, but it cannot be ovoalbumen or any of the egg-white proteins, for it is not coagulated by ether. Plimmer suggested that possibly livetin was vitellin with the majority of the phosphorus-containing parts split off from it. Tables 19, 20 contain Plimmer's figures for these two proteins.

Table 21.


May & Rose

Folin & Looney

(1922), (%)

(1922), (%)





















Free tryptophane (% of whole egg)

von Fiirth

& Lieben

Ide (1921)


Whole egg-contents






0-437 "otal tryptophane.

(% of proteins)

(% of proteins)

Whole egg-contents








Whole egg-contents


Ovovitellin has been the subject of recent investigations by Swigel & Posternak. They found that it broke up into three polypeptides which they call ovotyrine a^, ^^ and y^. The properties of these are listed in Table 19. It was found that ovotyrine ^ contained all the iron in the original compound, and that it could be split up into ovotyrine /Sj and ovotyrine ^2 the second of which again contained all the iron. All these derivatives were laevorotatory, and showed considerable resemblance to the lactotyrines which the same workers had previously isolated from casein. They identified their ovotyrine ^ with the avivitellic acid of Levene & Alsberg, and they stated that an enzyme was present in the fresh yolk which would, on standing at 37° C. for 10 days, double the yield of preformed ovotyrine jS.


— - u

O 3

5P =0 § a







•ft IN



orjouii % I

XxojpAq % I

DUBajs % I

oijnuiBd % I

oiajo % O


oijAjnq % I

3% I

N% I

•3 -ds I

xspui aApoBija^j | •ou jauqaH ^^

•ou ]A}30V I


anjBA jajsg r~


an^BA ppv ?

•ou uopBogiuodBS


in CO

O) o


aim oo

•ou auipoj

coco o -^ p eo^ oj w Tho^eoco Tt< e« fj " a f^ ^ii> N tJ< 6 f^ o» 'X> (^ t^co lo cotn i^co r^ c< <J) lO

in o coco (ij 6 CO 6 CO t^UD i^


fl S^ ci o o

< -C(V C^tH in

•3 "^ u . "^ <« ?4

^ D o5i ° n£,-£.

iH Pffiffi

2 2^ &• ^


g jj o s;a cn

c c fi 2 c 2 3



^ « « ^ a

^ o o

'^ - - •< « 





s i


e 3 S3 w


v5 «^

s >- p fl

pHC/2y3 3H













O o


O f~~ CO mCO CT) ^ ■*C0 iD

CT:cb w c< 6 '^ •4'6d ^ CO (£1 •*co r^ r^to i~~ i^ CO (j2


^ bo "^

.ii.a a i^z

c« oj y u c &. c

^ ci, u X! :r! u=! *■

a c c c c c V

o o o o o o


g C3 g

c £ « • o V o a

"* c e

o o




Hydrolysis of ovotyrine ^ revealed the presence of large amounts of /-serine, an amino-acid which had not previously been found in ovovitellin (see e.g. Plimmer & Rosedale's analyses). Some pyruvic acid and ammonia being given off as well, Swigel & Posternak calculated that, supposing these arose from breakdown of serine, there would have been sufficient serine present initially to combine with for all the phosphorus. They therefore suggested that the main phosphorus-containing unit of ovovitellin was serine-phosphoric acid.

Cohn, Hendry & Prentiss consider the minimal molecular weight of vitellin to be 192,000, i.e. much higher than ovoalbumen.

Kay & Marshall have also studied the yolk-proteins. They have prepared purer samples of vitellin and livetin than those of any previous worker, and have been able to free the former almost entirely from contamination with ovolecithin. Their vitellin is a true phosphoprotein containing i -3 per cent, of phosphorus and hydrolysed by I per cent, soda, though not by the phosphatase of the kidney. Their livetin is a pseudo-globulin, containing only the slightest traces of phosphorus (less than 0-05 per cent.). The yolk of the fresh egg contains no albumen. Vitellin, hydrolysed with dilute ammonia, gives a vitellinic acid containing about 10 per cent, of phosphorus. Kay has estimated the cystine, tryptophane and tyrosine in vitellin and livetin (Table 11); in the latter protein they are distinctly high in amount, a fact of some importance in embryonic nutrition. The relative amounts of vitellin and livetin in yolk would appear to be of the order of 3-6 to i for the hen and 3-8 to i for the duck, calculating from their nitrogen content. Kay regards livetin as identical with Gross' protein. The yolk of a fresh egg would contain from 600 to 900 mgm.

1*9. The Fat and Carbohydrate of Avian Yolk

The fatty acids of the yolk have been much investigated since the time of Gobley and Kodweiss, but little has been added to our knowledge of them. Paladino found olein, palmitin and stearin to be present. Analytical details are in Table 22.

A large part of the study of phosphatides, under the generic name of lecithin, has been made on that obtained from the yolk of the egg; thus the work of Diaconov, who showed it contained no neurine, Strecker, who discovered the presence of choline, Bergell ; Cousin ; Laves & Grohmann; Laves; Wintgen & Keller; Erlandsen; Stern


& Thierfelder; Frankel & Bolaffio (whose egg-yolk neottin was only a mixture of sphingomyelin and cerebrosides), McLean; Serono & Palozzi; Eppler; Riedel; Wilson, and Trier, who prepared aminoethylalcohol from it, all comes under this heading. In McLean's book will be found a review of it. Certain aspects of it, however, are important here ; for instance, the question of the presence of very unsaturated acids in ovolecithin. McLean in 1909 found stearic and oleic acids in it, but Cousin was able to isolate linolenic and palmitic as well, and Riedel; Hatakeyama; and Levene & Rolf obtained linolic and arachidonic acids. In another paper Levene & Rolf showed that the lecithin, carefully freed from kephalin, contained only palmitic, stearic, and oleic acids : saturated and unsaturated molecules being present in equal proportions. Again, Stephenson in 1 9 1 2 found an acid in the phosphatide fraction from egg-yolk, which had 20 carbon atoms and 6 or 8 unsaturated linkages. Although the proportion of unsaturated acids in egg-yolk is generally agreed to be small, yet it may be of importance for the young embryo if it passes through a period in the early developmental stages before it has the power of desaturating the ordinary fatty acids. Evidence which suggests this will be presented later (Section 1 1 • i ) .

The nitrogenous radicle in ovolecithin is largely choline, but difficulty was at first experienced in obtaining a theoretical yield on hydrolysis; thus Moruzzi got only 77 per cent, in 1908 and McLean only 65 per cent, in 1909. This was accounted for, however, when it was found that amino-ethyl alcohol was also present. The two bases together make up all the nitrogen in the molecule. Erlandsen was the first to question the view that lecithin alone accounted for the phosphatide fraction, but he was not himself able to isolate anything else. Later workers (Levene & West and Stern & Thierfelder), however, found that kephalin is also present in yolk, and it would probably be in the kephalin molecule that the unsaturated fatty acids would be present. Analyses of kephalin from the yolks of fowls are given in Table 10 ^. McLean in 1909 isolated from egg-yolk a third phosphatide which resembled cuorin, but it is very doubtful whether this was a true chemical individual. Sphingomyelin has also been found in egg-yolk by Levene (191 6), and lignoceric as well as hydroxystearic acid was present in it.

All these substances exist in the yolk in close association with the proteins. Hoppe-Seyler it was who first observed that, after


prolonged extraction of the yolk with ether, a considerable proportion of the phosphatides still remained behind, and could be extracted with alcohol. It was thought for a long time that the phosphatides and the vitellin were in chemical combination which was broken by the alcohol, but since the paper of Fischer & Hooker in 1 9 1 6 the general opinion has been that this combination is only physical. Stern & Thierfelder isolated traces of the cerebrosides, phrenosin and kerasin, from egg-yolk in 1907.

The neutral fats and the lipoids of the yolk are variously affected by the nature of the fats in the food of the fowl. Henriques & Hansen, who were the first to investigate this subject, found that, if food containing very unsaturated acids was fed to the laying hens, the neutral fats in the eggs were affected, but not the fatty acid components of the lecithin. Their figures are shown in Table 22. When the food consisted of barley, pea or rice, the iodine number of the neutral fats in the egg varied round about 77, but hemp or linseed sent it up to about no, although no matter what the food was the iodine number of the fatty acids in the phosphatide fraction remained constant at 75 or so. Henriques & Hansen also found that the iodine number of the fluid fatty acids of the neutral fat was normally 107-5, and that the fluid and solid fatty acids of the phosphatide fraction were 151-3 and 98-9 respectively. The former accounted for 64-3 per cent, of the lecithin fatty acids. The experiments of Henriques & Hansen have been repeated and confirmed by Belin and by Terroine & Belin. The last-named workers, together with McCollum, Halpin & Drescher, some years later reported that the lecithin fatty acids would vary, as well as the neutral fatty acids, with the diet of the hen. Their figures, which are given in Table 22, certainly show a variation in the iodine numbers of both fractions. All these workers recognised the presence of unsaturated acids in the yolk fat, and Henriques and Hansen's figures came between the theoretical values for oleic and linolic acids.

Work was continued along these lines by McGlure & Carr. Using pigeons, they found that the fat content of the eggs could only be altered slightly by feeding rations high and low in fat.

% fat in the eggs Cocoanut fat ... ... ... 4-0

Beef tallow ... ... ... ... 6-75

Average of all fat diets ... ... 4-96

Average of all non-fat diets ... 4-81

3 be

■^ ^


be ffi


















C5 •:


(U bO S


^ OUIUIB l^jox

^ uiajojd ajqEinSBOo



|«j^ upjojd 3iqBinSBOD-uo]y[

N F50X


^ ouiuiB aaaj

^ UT3}Ojd

^ uiajojd ajq^inSBOD-uofj

fvj OUIUIE 331^

fy[ asouinqiY



\[ uiajojd 3{qB|nSB03

I^J upjojd-uou 3{qB{n§BOD-uo^

|vj uiajojd 3iq^|n§BOD-uo^



bo !^

•j3 "O










tr, u:,





!r> o


o o o

O ^ IN ir^ lO 1^^

OO ir>OoO,oOO COtXJ ,O^0D IT) O <0 LO CO

' « «" <N «'" -^ kT ;^ «" cT




V •

^ s

bo w

• O

e &

be V

S I:

"O S c « 

o >^

^ ^-^ S^.^


^ g § ^



[PT. Ill

Again, during the fat feeding, tiie saponification value of the eggfat was 176 (166-190) and during the rest of the time 173. The iodine value was in the former case 70-5 and in the latter 70-8.

Table 24. Lipoid in egg-yolk.




/o /o Wet weight Dry weight






















Glikin' s figures.

Total fatty

P,05 in Lecithin




% of in %



% dry

fatty of fatty




acids acids


Pigeon (yolk)



3- 16 35-73




3-88 38-42


Turtledove (yolk)



4-10 46-65


Starling (whole egg) ...


5-67 64-44


Hen (yolk)


— —


Thrush Cat ... Rabbit Guinea-pig

Lecithin in % dry weight at birth or hatching 8-18 5-06 4-91 3-79

Some suggestive investigations on the biological significance of ovolecithin were made b-y Glikin in 1 908, whose figures are shown in Table 24. Choosing the pigeon as a typically nidicolous bird, and the hen as a typically nidifugous one, he was able to show, using a variety of extraction methods, that the yolk of the former was considerably richer in lecithin than the latter, the former containing about 29 per cent, dry weight, and the latter about 17. The further but rather fragmentary observations which he made on the starling and the turtledove confirmed this relationship. It is interesting that Tso informs us that certain small Chinese breeds of hen produce very small eggs (scarcely 40 gm.) and that these contain a much higher percentage of lipoids than ordinary eggs though an equivalent percentage of protein. He concluded that lecithin, one of the most essential yolk-constituents, was specially concentrated in nidicolous yolks and


was associated with the property of early hatching or birth. Thus he compared the thrush (nidicolous) with the guinea-pig, which is born almost ready to eat green food and hardly passes through a lactation stage; in the body of the former 8 gm. per cent, lecithin dry weight was found, in the latter only 4. The new-born cat and rabbit occupied intermediate positions. It is interesting to note that Glikin's figures bear out those of Tarchanov on the question of water-content of yolks from the two types of birds.

Tornani affirmed in 1909 that a difference in lecithin-content was observable between fertilised and unfertilised eggs. But as he gave no figures in support of his contention, it has not been treated with much respect by subsequent workers.

The carbohydrate of the yolk has been the subject of only a very few researches compared with that of the white. The figures which have been obtained are shown in Table 16, and it will be seen that in no case has the total carbohydrate been estimated, and only in one case the glycogen. After Claude Bernard's isolation of glycogen from the yolk, a persistent belief grew up that considerable amounts of this substance were present there ; this was apparently based on the description by Dareste in 1879 of "amyloid bodies" in the yolk which gave microchemically a strong blue colour with iodine. Dastre immediately pointed out that the occurrence of starch there was highly improbable, and that if any glycogen was there it should give a wine-red colour; he himself, however, could find neither. But he did not succeed in suppressing the rumour, for Virchow, and later Schenk, supported Dareste, though nothing has been heard of this yolk-constituent since 1897, and Sakuragi's analysis revealed the presence of only 2-2 mgm. per cent, of glycogen. Bierry, Hazard & Ranc asserted in 191 2 that they could obtain a great increase of carbohydrate after hydrolysing the yolk with hydrofluoric acid under pressure, but this would not imply, as they seemed to think, that glycogen was present, for all kinds of other compounds such as proteins (Gross' protein for instance) might yield glucosamine under such treatment. They identified glucosamine in the hydrolysate. On the other hand, Diamare, who hydrolysed with acetic acid, could only obtain faint traces of combined glucose in the yolk. He dialysed both white and yolk, and in both cases was able to estimate the free sugar, but in the case of the yolk very little combined glucose seemed to be present. Further studies on this


subject should be undertaken, for the methods of Diamare and Bierry ahke were of questionable reliability. Diamare, however, went rather further into the matter than other investigators, and, thinking that the yolk glucose might only be present there owing to an inflow from the white, examined the ovarian eggs, in which he found glucose in much the same proportion as in the yolks of laid eggs. He does not state whether the ovarian eggs were yellow or white, and, as he frequently gives his results in the form of grams of glucose without mentioning the weight of the fresh material, it is impossible to calculate the percentage (see also Tillmans & Philippi) .

We have already seen that cholesterol was identified in the yolk of the hen's egg by very early workers such as Gobley. In 1908 Menozzi and in 191 5 Berg & Angerhausen sho\ved that egg cholesterol was identical with that from milk and bile. It is certainly present in the unincubated yolk both free and in esterified form with fatty acids. Serono and Palozzi investigated a substance from egg-yolk in 191 1 which they called "lutein" but which turned out to be nothing but a mixture of cholesterol esters. Other investigators have estimated the amount of free and combined cholesterol in the unincubated egg, and their figures are given in Table 25.

Table 25. Cholesterol-content of hen^s egg.


per whole egg

Investigator and date

Parke (1866)

Mendel & Leavenworth (1908) Mueller (1915) ... Ellis & Gardner (1909)... Thannhauser & Schaber (1923)

Cappenberg (1909)

Dam (1929)

Schonheimer (1929)

Cholesterol Cholesterol (free) esters

378 215-9 24-2 489 173 54-2

296 — 337 —

Total 248-3

240-1 600-6 227 180

Free in % of total





Many Other substances have been found to be present in the egg at the beginning of development, e.g. choline, alcohol, creatine, creatinine, inositogen, lactic acid, plasmalogen (Stepp, Feulgen & Voit, 1927), etc. These will be mentioned as occasion arises during the succeeding sections of the book. Allantoin is not present (Ackroyd).

The yolk of the hen's egg also contains vitamines, pigments, and a variety of enzymes, but these will be dealt with under their respective sections. As Langworthy has shown, it may also contain very

'§,0° ? CO V C -!i; Ci cfl an

S: o' " ?^M ^ ?

o o>^


en M O

C — O"

5 2 e5 c9 fc,

08 O'c


E 0)

s a " u

-§ S f^a• 0^^ ^ mi

ja :'

"ego c

>^4 ::

d uispnjsj

d aiqnios


d ajqnios J aiqnios

J upjojd -oqdsoijj

d appEqdsoqj

d unPMA

d ajqnjos


jaiqnios -J31BM 3UIe3jo So I

d aiqnios "I

-J31BM pjOX, g f

d siqnps -loqooiB iBjox ^

(uonoB.ijxa ^ jaqja iajje) d r ajqnps-ioqoofY

d aiqnios-jaqja |



d aiqnps n


d aiqnjos r -ja^BM 3jub3jo

d aiqnios | -jajEM iBiox

d siqnios


(uotjoBjaxs 7^

jaqaa iajjE) d j^

ajqnjos-ioqoDiv ip

d aiqnps-jsqjg ],

d lElox

t^ m >^ iri v^o O O

O (U

^ >. -•

  • ^'o "u

^>> as

j: a


Oj: ■Co o-g

^^ ^^

(U u T3 « 

C O 3 ^

5 °

•^ D.

a 1

m (U

o c

J3-Q.2 V C

3 ao

■+ir) 000





various substances derived from the diet of the hen, and these, if they are odorous or possess taste may very easily betray their presence (e.g. the Swedish "Schareneier" described by Hansson). Table 23 gives the figures which are available for the nitrogen and Table 26 for the phosphorus distribution. These summaries bear out on a detailed basis what has already been said.

I -10. The Ash of the Avian Egg

The ash of the yolk and the white of the hen's egg has been investigated by several workers, and a study of it reveals certain interesting features. If Table 27 be examined, it will be seen that, in the yolk as well as the white, potassium has almost invariably been found to be present in greater amount than sodium. This is one of the characteristics of the egg-cell, as will be seen later when the eggs of other animals are considered. The yolk is also marked by the very high percentage of phosphorus, most of which is, in accordance with other evidence, in organic combination. The calcium is also mainly in the yolk, as is the iron, but not the magnesium. If now the amounts of metallic and acidic ion be calculated out in millimols and milliequivalents per cent, wet weight, it is found that in both yolk and white there is an uneven balance, but while in the former case there is much more anion than cation (anion/cation ratio above unity), in the latter case the exact reverse holds, and the anion/cation ratio is somewhat below unity, about 0-55. In the white, therefore, some of the potassium and sodium must be combined with the proteins, as ovoalbumenates, etc.* However, the excess of cation over anion in the white is not so considerable as the excess of anion over cation in the yolk, and, bearing in mind also the much higher percentage of solid in the yolk than in the white, it would be expected that the anion/cation ratio of the whole egg would be greater than unity, and would approach that of the yolk. The facts show that this is, indeed, the case, for the average anion/cation ratio calculated from the results of all observers for the whole egg is 2-3, as against 2-8 for the yolk alone and 0-54 for the white alone. This was first noted by Garpiaux. Forbes, Beegle & Mensching expressed it simply thus :

Cubic centimetres normal solution per 1 00 gm. dry weight egg Total acid ... ... ... 120-28

Total base ... ... ... 39*42

Excess acid over base ... ... 8o'86

  • In both white and yolk, of course, the inorganic ash is basic.

Duck Hen ...


Table 27. Aili of the avian


% of total ash








PO, c












79-1 —






Bi-2 15-2

Mgm./loo gm. wet weight

K Na Mg Ca

Whole 2-59 2-72 0-75 3-28

)-33 048 0-79 3-5

Fe SO, PO,

.,.„. . , Total cation Total anion

Millicquivalents . '■ ^ , * ^ Anion/

K Na Mg Ca Fe SO, PO. CI mols equiv. mob equiv. ratio Investigator and date

3 — 2-59 2-72 1-50 6-56 — — 30-9 — 9-34 ,j.j7 ,0-3 jo-j 231 Polcck (1850)

y - 3-33 0-48 1-58 7-u - - 35-7 - 8-t ,2-39 ,,-9 jj.7 ,.-88 Rose (1850)

65 738 4-5 0-9. 1-66 7-0 — _ 34-95 7.38 9-79 /.,■„ ,9.03 ^.^j 3.00 Bialascewicz (1926)

3-32 50-3 S'll" 173 204

93 — 3-8 604 ]o8 443 8-8 0-34

004 636 302 443

4-6 — o-o8 19-08

S-j/ 9-42 s2-;a

Plimmer & Lowtidnt (1927) Vaughan & Bills (t878) Delezeime & Foumeau (1918)

Carpiaux {1908) Buckner & Martin (

Normandy Bourbonnais

)125 17-5 273 fo

)I30 15-4 23-8 Trace

3-590 23-55 ao-7 0-96

0-720 a7-tiG i'2-i 2-7

0-634 — — 24-B 25

o-Goo 22-1 20-63 1-41 I
















0-52 2-9 1




4-87 While




0-24 o-.e

■ 2



o-Bi 052

0-65 0-40

3-56 5-3 5-11 3-8









7-17 77-0^ 0-C25 Champion & Pellet (1876)

(not very reliable) 7-77 17-84 0-G29


0-9 4-4 9-39 s-s8

S-72 0-58 Poleck (1850)

3-51 4-12 035 Rose(iB5o)

- - - Voit(,877) 4-82 ssf c-55 lljin (1917)

— — — Prout (1822)

— — — — 4-25' —

523 4-l'8 3-79 1-24 8-72 345 '0-9 '-"8 2-2 13-62 —

129-a 52-5

85-5 4-25 Go- 16 3-6 84-6 —

64-7 1-8 48-4 3-7

.■58 6-6 !-48 9-0 2-4 9-6

,2-48 13-81 12-69

Average ... 0-54


- - 139-8

Straub & Hoogcrduyn (1929)

Poleck (1850)

Rose (1850)

Voit (1877)

Carpiaux (1903)

lljin (1917)

Prout (1822)

Straub & Hoogerduyn (1929)

  • Kreis & Studinger.


This is probably the most interesting consideration that emerges from Table 27, but it may also be noted that the ash-content of the white is just about half that of the yolk, a relation which would practically be reduced to equality if the phosphorus in the yolk was not taken into account.

The presence of certain chemical elements of lesser biological importance has been announced from time to time in a group of papers which have some interest, although it is difficult to see, as yet, what their importance is for the development of the embryo. Fluorine has been estimated by Tammann and by Nickles, copper by Dhere, boron by Bertrand & Agulhon, manganese by Bertrand & Medigreceanu, iodine by Bonnanni and by von Fellenberg, lead by Bishop.

These elements appear to be normal constituents of the egg. The iron-content can be artificially increased by feeding iron-rich rations to the hen, and iodine can also be introduced into the egg in this way, as has been done by Bonnanni, Kreis and others, but Hofmann found that though iron and iodine would enter the egg thus, it was impossible to get copper to do so. In just the same way Ricci found it difficult if not impossible to get As or Hg into the hen's egg by feeding subtoxic doses to the hen. The normal copper-content of the hen's egg cannot be varied like its iron-content. The importance of iron in the formation of haemoglobin is obvious, and the little that is known about this process will be discussed in the section on pigments in the embryo. Wassermann made a histochemical examination of the egg-yolk and vitelline membrane for iron, and found a relationship between the embryonic blood-islands and the iron of the yolk.

Some of the other data in Table 28 call for comment, Tammann's 1-13 mgm. per cent, fluorine in the fresh yolk works out at a quantity of 0-2 mgm. per egg, and, as Zaleski found 0-23 per cent, fluorine in the bones of the chick at hatching, o-o8 mgm. fluorine would be required in the egg at the beginning, or less than half of what is actually there. Zaleski's figure, however, is old, and may be too low. It would appear, on the whole, as if the greater part of the fluorine, iodine, copper, zinc, lead, aluminium, silicon and manganese is localised in the yolk, and the greater part of the boron and arsenic in the white. In view of the importance which we now attribute to these less common elements as catalysts in living tissues, this distribution may be found to have considerable significance.




Pi ^



























^ ^

.« "

^._, en «J 3 <^ >-i ir!

^ c «e ^

O^ o



„ a


.23 w

w 5<^

3 -"

_>. « 


"rt CU




o . , bD



SU > m

-d „ -0-3

o o o o

c ^

F « 

S bD bo

C; V3 C3



O CO ^- o o •■

?? "

-. CI " O o r-» o o o o o o o 6 6 6 6 6 6 6

5 =a


^ s

^..— V

c^ S

.-, lo

1-1 v!l CI

--^ f. W

c>! nj'^

Elvehjem Halpin Wolff (19 Lindow,


Fleurent i McHargu Bodanski




.-S-S 1

6 6 oi 6

? I


h K

-03 OH


S 2








^ ^

,. ^








W n




i-4 « 








„, — ,. '- ^



cn r.






CTi ^ ...





o - >-■ »3


S-i '•

J, ci

«^ it: ^ rt





^^ S-^

Oh cq M

6 6

« cq cq>Mm CLiOiiq Q

^ ^ ^ Oi ^ o





o o




(UN-" HP^

^ o -^ = ^0^:2. o _C ^ '^ '-I — '

P ^ c^


6 Oh

•£ bo

?2 o


tj t?




u ^

6 6 - 1



6 6




6 u ^ ■ — ^ "In. S3 2? J3 -^ ^ I I ffi ffi K S 3 S ^ < ■^ B c3 c S —.0.2 ^ V 1;; O bCc ■w o ^ S c w bo JJ b tP. ^■M U 3 4->

-> u c4

c S-5 o ex ,„ c_> n 5" .U_0 t, « 5 .§■« 3 fci ■" CX bo c -3 3o6 THE UNFERTILISED EGG AS A [pt. iii The figures which have been obtained by those investigators who have examined the iron-content of eggs are seen in Table 29. All found a great deal more iron in the yolk than in the white, as might have been expected from the earlier micro-chemical researches of Tirmann and Kobert. This kind of work had been originated by Schmiechovski, and was continued later by Wassermann in the interesting paper already referred to. Schmiechovski found iron histochemically throughout the yolk, but considered that, in the white or milky yolk, it was confined to the megaspheres. Table 29* Iron in hen's eggs. Italic figures indicate dry weight data. FcaOs gm. % wet weight Without iron-rich diet With iron-rich diet , ' ^ r ^ ^ Egg-white Egg-white Egg- plus Egg- plus white Yolk yolk Shell white Yolk yolk Shell Investigator and date •0024 -0088 -0047 -0272 -0040 'oogs -0059 -0272 Loges & Pingel (1901) — — -0046 — — — -0040 — Kreis (1900) •0057 -026 •0165 — — — — ^ Lebbin (1900) ■03 -05 -03 _____ •001 12 -00995 '00425 — — — — — Hartung (Mar. 1902) •00087 -01106 -00451 — — — — — ,, (May 1902) — — — — -00208 -01621 -00729 — ,, (June 1902) Trace -0108 — — — — — — Bunge (1892) — -0121 -0018 — — '0175 -0032 — Hofmann (1901) — -;r -0057 — — — — — Boussingault (1850) — -063 -og§ — — — — — Leveque & von Tschermak (191 3) None 'OI43 — — None '0143 — — Elvehjem, Kemmerer, Hart, & Halpin (1929) Wassermann, using both the ammonium sulphide and the Berlin blue methods, decided that it was present in both kinds of yolk, but that it was not confined to those special elements in the white part. In fact, it was very much more abundant in the white than in the yellow part. This finding has never been corroborated by chemical analysis, but, if it is, it will have considerable importance, in view of the time at which haemoglobin is most vigorously manufactured by the embryo. For a further discussion of these questions see the section on pigments. i-ii. General Characteristics of Non- Avian Eggs With this we may conclude the discussion of what is known about the typically terrestrial egg, that of the bird. Now SECT. I] PHYSICO-CHEMICAL SYSTEM 307 aquatic species far outnumber the terrestrial ones; as Spenser put it: O ! What an endlesse Worke have I in hand To count the sea's abundant progeny, Whose fruitfull seede farre passeth those on land, And also those which wonne in th' azure sky: For much more eath to tell the starres on hy, Albe they endlesse seeme in estimation, Than to recount the sea's posterity. So fertile be the flouds in generation, So huge their numbers, and so numberlesse their nation. It might therefore be supposed that a much greater space would have to be devoted to their eggs than what has already been taken up, but this is not the case, for the bird's egg has been so convenient a material for research that the knowledge we have of it outweighs that of the eggs of all other animals put together. Indeed, the data about the eggs of other groups are very fragmentary, so that much caution has to be used in making comparisons, and general relationships are much more difficult to enunciate. Van Beneden's classical memoir may be recommended as an account of the morphology of the eggs which are to be mentioned, and it is hardly necessary to refer to Balfour's book on comparative embryology. If Table 30 is examined, and compared with Table 2, it will at once be seen that the percentage composition of eggs of other classes of animals differs markedly and in very definite ways from the egg of the hen. The case of reptiles may first be taken, as being less remote than others. The reptilian egg seems to be distinctly drier than that of the bird, by about 20 per cent., and much more variable in its fat/protein ratio. For, while in all birds' eggs that have been investigated, the amount of protein, whether related to dry or to wet weight, is about the same as that of fat, the reptilian egg shows big variations from this rule. In the eggs of the tortoise and lizard, for example, there is twice as much protein as fat, while in those of the grass-snake, studied by Galimard, there is three times as much fat as protein. This fact will be mentioned again later (see p. 313). Considerably more is known however about those of amphibia, which have also been found to contain a great deal more protein than fat. Thus, instead of the 40 per cent, protein and the 40 per cent, fat which make up the dry substance of the hen's egg, Faure N " 3-— -_-N ■5 ^M 'o >^ 2 S C ■ H £ 3 K Own 53 fc 3 rx t c -a t

  • --- o o

i-S «  ~«3 O O CO H snjoijdsonj spiodiq SSAllOBJJXa snouaSoijju-uo]^ u snjoqdsoqd I | UOU3EJJ ajqegtuodESun iejoj^ | | I0J3jsa]0H3 I I spiodiq Hsy saAnoEJjxa snouaSojj'iu-uojsj l«£ do 6 b> «  uiajojj I o J3;bav •uj3 aSa am JO iqSia^ f^ f-* b* li^ i^ mvo ID tD lira a s s '^ o -SS ^2 2o.s2 C V C >; Itil t « « c « a c c c c c s c a c c Kc bi be M en 0000 - o tc Uc Vh I- ' >-i 2 2 ^ oC;'S^ o'o.Ji --3 ^:3 2 N T3 Ca N — ^ t-, o f- ^ ^ , ^S^ ^— P- 15 c 3 O O 'E C N li CSS «  H >^ - "I I* ffl>^"0 o 5J C ij!!0.>Sl fc <3 •* o pn 00 n t^ O r^ b>cb b b* op ^00 00 00 M N>0 ri inb'b ■* N OO O l/l w N O " o 00 '-' O' ro O QC 00 30 O 1^00 00 00 00 ■*vO « 00 o o "i-oo o b " 6 N O f^ I 00 N rf C poo PJ 00 M N t^OO OM^ O N 00-^riON^O> f^ iri f*^ r^ f^O m b^ NNNNNNNN COO OO 00 N O' N O 00 r^ o 00 c , N N N\b b CO r*) CM^ Tj- iri sO sO o> N o o t^ 1^ o OO O r^ O^ f^ N f^ ^ b.M «  o ■^oo fn H^ io f^ b* 11 N N N O' 'j- O 30 6 b " ^ N 't ^ in m\0 N (vj f^ inOO V CM MM on in Tj- o ui O o \0 M in ^ r^ 'noo M r^ O'O oo \0 -^ in in r^ ^oo tn tm •III 5 aOl ri^! Omen OU O EI S'^ '^ . -s Hi hcuHHE-' E-h • c o .■5 ^ '-'^^-^ JJ — c^ QQQ C o ^ ^<J8 tfS 0:5 o CO sruoijdsoiij I0J3js3i0q3 spiodiq qsy S3AI}3EJ)X3 snouaSoJi'ra-uoisj ua3oj}Tjsj I I IBJ UI3JOJJ sruoqdsoqj U0U3BJJ spiodiq I I qsy o op saAijOBJixa SnOU330Jl'TU-UO]sI upjOJJ i91EjV\ boob •uj3 333 aqj jo jqSp^ | | moo m N f^ m * •<J-vO O O t~ N o ino \0 o o p p o o o o o b b b b b b b T)-o o^o o o o o ■*p;tr^r»9>J~ b ^ " boovo b t^ t^ £^ t^ r^^ m^o vt 1 I I I I I I I ■~ 1-1 ii « 2 "s ? 3 ^ £3 r r w O .'t! ^ -^ 0, -s^ Q Q m fl^^UKPQ .2S o^'i-ii-a i2 w 0^= •a • E c.E-S 0.1 3 M O C Q 0) u cs 3 0*^ taJ ffiXUJOw IX CI, E-u « .s M _ - ?> C 00 ^ 1 -^ 1 1 1 > 1? °~ ^ T3 C s. N It 'Si 1 rov (188 1903) Conte ( 922) rcmiet & wOO N ,_, o 1^ a "ir^^^f^^Sr^ -^'^i^a Tichoi Farkas Vaney Russo Faur^ E p5 mil s 5 fc^ b o O I 1 A A 000 ° I N 0* N 6 U1 r» 1 o o b b S 5 s s a US «  iapu omct lu 1 &0 cue ••* a-»* a ■s^j: a "c ^"^^ tij c c c '^ ^— ' -000 J3 »■ u u u 3 3 3 ^ !<! « «  i-i CDCfiC/D ^ I 5 a S ^ OS 03 s ^. a ^ •- :2 4) 55 ^ -a ~ ■= <; b J U CO S 2 O •-•2 "55 S 9 E Z 3 312 THE UNFERTILISED EGG AS A [pt. m Fremiet & Dragoiu found the dry material of the egg of the frog to consist of 60 per cent, protein and only 14 per cent. fat. And this has been the experience of all those who have analysed amphibian eggs. Next to the hen's egg, the eggs of fishes have probably received the most attention. The recorded analytical figures for fish eggs are often deceptive, for many analyses have been made of salted fish roes and egg-preparations such as caviar, but the greatest care has been taken not to include in Table 30 results which might have been vitiated in that way. The question is complicated by the fact that analyses of the purified constituent egg-substances prepared from preserved material may well be admitted into consideration, for, except in certain cases, they would probably not undergo much change during the process of preservation. As in the case of reptiles and amphibia, the fish egg is characterised by its predominance of proteins as the food of the growing embryo. It should be remembered that, in all these comparisons, the yolk of the hen's egg is a more proper standard of reference than the whole egg-contents, in which case the differences become even more remarkable. This generalisation appears at all points; thus, the brook-trout, a fresh-water fish, has 30 per cent, of protein and only 9 per cent, of fat, and the herring has 26 per cent, as against 3 per cent. In Table 31 the protein/fat ratios are collected together, and the difference emerges there with great clearness. Though it is obvious that fishes vary considerably among themselves as to the fat-content of their eggs, yet all of them have more protein than fat. The only fishes in Table 31 which come near to being exceptions to this rule are the sturgeon and the dogfish, both of which have an unusually high amount of fat in their eggs (a fact which accounts for the superiority of Russian caviar over other varieties, for the former is made chiefly from the eggs of sturgeon) . Besides this general difference between the egg of the bird and that of the fish, there are many others, but they concern the chemistry of the individual components rather than the rough constitution of the egg as a whole, and will, therefore, be dealt with later on. If Table 31 be further studied, it will be seen that, as far as can be known at present from the few analyses of crustacean and cephalopod eggs, the superior proportion of protein over fat holds good there also. Curiously enough, the only analysis we have for a gastropod egg gives a picture more resembling the egg of the hen, comparatively equal amounts of fat and protein being present. SECT. l] PHYSICO-CHEMICAL SYSTEM 313 The position of affairs may perhaps be summarised by saying that it is only the birds which have been successful in producing an egg really well stocked with fat, though the reptiles clearly show an approximation to this achievement. Does this mean that the storage of fat in the egg is particularly associated with terrestrial embryos? The facts and arguments to be brought forward in later chapters (see Sections 7-7, 9-15 and 1 1-8) make this hypothesis a very likely one, but, as Table 31 shows, the silkworm (the only representative of terrestrial Table 31. Protein/ fat ratio in various eggs. VERTEBRATA Amniota Aves Hen (yolk) „ (whole egg) 0-450 1-035 Reptilia Anamnia Lizard Grass-snake Tortoise 1-450 0-298 2-271 Amphibia Pisces Frog Cod Sturgeon Herring Carp Trout Dogfish Salmon INVERTEBRATA 2-6lO 12-550 1-937 8-333 9-101 3-518 1-050 2-400 ECHINODERMATA Echinoidea Sea-urchin 2-370 MOLLUSCA Cephalopoda Gastropoda Octopus Limpet 5-750 1-245 Arthropoda Crustacea Arachnida Insecta Crab King-crab Silkworm 2-709 3-166 2-139 Annelida Polychaeta Sabellaria 3-162 arthropods) does not seem to have succeeded in storing fat in its egg to any great extent. It might, of course, be argued that this was one of the factors which prevented the insects attaining any considerable size and rivaling reptiles and mammals for the possession of the land. The mammals gave up the heavy fat storage in the egg when they invented viviparity and the fully developed placenta. In 314 THE UNFERTILISED EGG AS A [pt. iii this connection the monotreme egg would be a chemical study of great interest, and it is characteristic of the exasperating fragmentation of this field of work that all we know about the monotreme egg is that its membrane seems to have the properties of a keratin. The suggestion that the metabolism of the fowl, operating on a continuously high level of energy turnover, would naturally tend to fill up the eggs with fat, and is associated with the well-known higher temperature of the avian body (Wetmore) may not be without value, but any special emphasis on fat metabolism in adult birds is precluded by the statements in Schulz's review. It is very significant that as animals became more complicated and more adaptable to varied surroundings, higher, in fact, in the taxonomic scale, they loaded their eggs to a greater extent with yolk. Since the extra material was usually fatty acids, this process appears strikingly in Table 31 . The effects of the yolk have long been familiar to embryologists, and have been best described, perhaps, in a passage by Milnes-Marshall. "The immediate effect of a large amount of yolk", he said, "is to retard mechanically the processes of development, but the ultimate result is to shorten them. This paradox is readily explained. A small egg, such as that of Amphioxus, starts its development rapidly, and in about eighteen hours gives rise to a free-swimming larva, capable of independent existence, with a digestive cavity and a nervous system already formed ; while a large egg such as that of the hen, hampered by the great mass of yolk by which it is distended, has, in the same time, made very little progress. From this time onwards, however, other considerations begin to tell. Amphioxus has been able to make this rapid start owing to its relative freedom from yolk, but now this freedom becomes a retarding influence, for the larva, containing within itself but a very scanty supply of nourishment, must devote much of its energies to hunting for and to digesting, its food, and hence its further development will proceed more slowly. The chick embryo on the other hand has an abundant supply of food in the egg itself and has no occasion, therefore, to spend its time searching for it, but can devote its whole energies to the further stages of its development. Hence, except in the earliest stages, the chick develops more rapidly than Amphioxus and attains its adult form in a much shorter time. The tendency of abundant yolk to lead to shortening or omission of the ancestral history, is well known. The embryo of forms well provided with yolk SECT, i] PHYSICO-CHEMICAL SYSTEM 315 takes short cuts in its development, and jumps from branch to branch of its genealogical tree instead of climbing steadily upwards. Thus the little West Indian frog, Hylodes, produces eggs which contain a larger amount of yolk than those of the ordinary English frog. The young Hylodes is consequently enabled to pass through the tadpole stage before hatching, and to attain the form of the frog before leaving the c:gg\ the tadpole stage is, in fact, only imperfectly recapitulated, the formation of gills, for instance, being entirely omitted." The more yolk, then, the longer the embryo can remain an embryo before having to face the external world, and the more preparations it can make for that event. It is probable that this question is intimately bound up with the penetration of fresh-water surroundings by the originally marine forms. "It has long been noticed", said Milnes-Marshall, following the classical exposition of Sollas, " that marine animals lay small eggs whereas their fresh-water allies lay eggs of much larger size. The eggs of the salmon or trout are much larger than those of the cod or the herring, and the crayfish, though only a quarter the length of the lobster, lays eggs of actually larger size. The larger size of the eggs of the fresh-water forms appears to be dependent on the nature of the environment to which they are exposed. Considering the geological instability of the land as compared with the ocean, there can be no doubt that the fresh-water fauna is, speaking generally, derived from the marine fauna, and the great problem with regard to fresh-water life is to explain why it is that so many groups of animals which flourish abundantly in the sea should have failed to establish themselves in fresh water. Sponges and Coelenterates abound in the sea, but their fresh- water representatives are extremely few in number; Echinoderms are exclusively marine ; there are no fresh-water Cephalopods, no Ascidians, and of the smaller groups of Worms, Molluscs, and Crustacea, there are many that do not occur in fresh water. Direct experiment has shown that in many cases this distribution is not due to the inability of the adult animals to live in fresh water, and the real explanation appears to be that the early larval stages are unable to establish themselves under such conditions. To establish itself in fresh water permanently an animal must either be fixed, or else be strong enough to withstand and make headway against the currents of the streams or rivers it inhabits, for otherwise it will in the long run be swept out to sea, and this condition applies to larval 3i6 THE UNFERTILISED EGG AS A [pt. iii forms equally with adults. The majority of marine invertebrates leave the egg as minute ciliated larvae, which are quite incapable of holding their own in currents of any strength. Hence it is only forms which have got rid of the free-swimming ciliated larval stage, and which leave the egg as organisms of considerable size and strength, that can establish themselves as fresh-water animals. This is effected most readily by the acquisition of yolk — hence the large size of the eggs of fresh-water animals — and is often supplemented by special devices." Here is an explanation for the well-known paucity of eggs in freshwater plankton. In certain cases it is possible to induce an embryo to skip the larval stage which it should normally pass through. Thus Child could abolish the free-swimming larval stage in the ascidian Corella willmeriana, simply by removing the eggs from the parental atrial chamber {p¥L j'^.) to normal sea-water (/>H 8-4). Giard had also noticed the discrepancy in egg-size between closely related marine and fresh-water forms, and had classed it among those cases where like adults have unlike larvae ("Poecilogony"). The classical instance is perhaps that of the shrimp Palaemonetes varians, one variety of which {microgenitor) lives in the sea near Wimereux and has eggs 0-5 mm. diam. (32 1 per female) and another of which {macro genitor) lives in fresh water at Naples and has eggs 1-5 mm. diam. (25 per female). Giard has reviewed this subject in a very interesting paper. "Dans un groupe determine", he said {(Euvres diverses, p. 18), "la condensation embryogenique va en croissant des types marins aux types d'eau douce ou terrestres." The correlated proposition, namely, that the fresh-water forms generally lay fewer eggs than the marine ones, is illustrated by the following instances collected by Carpenter: No. of eggs laid per female per annum A Lamellibranchs ... Gastropods Fishes Crustacea Marine form Ostrea edulis i ,800,000 Buccinum undatum 12,000 Haddock 9,000,000 Lobster 5,000 Fresh-water form Uniopictorum 220,000 Anodonta cygnea 18,000 Average of many snails 100 Average of many limpets 6 Ovoviviparous pond-snails 15 Brook-trout 750 Crayfish 200 Another reason for the poverty of fresh-water fauna was suggested by von Martens who pointed out that the fresh-water climate, with its periods of desiccation and freezing, was much more severe than SECT, i] PHYSICO-CHEMICAL SYSTEM 317 that of the sea. But even these two causes together cannot fully account for the phenomenon, for there are many cases of individual species which they will not cover; thus the Cephalopods, which hatch out as minute but very active copies of their parents, i.e. which pass their larval stage within the egg, and which should therefore be immune from the disadvantage described by Sollas, never penetrated into fresh water. A third reason must be added to those of Sollas and of von Martens. As will be shown in Sections 12 and 13 the marine invertebrate embryo depends largely on the salts of the sea water for its supply of ash, and therefore could not be expected to develop in a medium very poor in inorganic matter. Colonisation of the fresh water could not occur, then, until animals had begun to provide in each egg sufficient ash to make one finished embryo. There seem to be few data concerning the capacity of marine invertebrate eggs to develop in fresh water, although the adult animals have been found often enough to accustom themselves to a fresh-water environment (see the instances given in Semper) . Many studies of the effect of hypotonic solutions on marine embryos can, however, be called to mind, and in all the cases the results are teratogenic. The fate of the Cephalopods, it is interesting to note, is explained by this third factor, for Ranzi has demonstrated the intake of the salts in the sea water by the octopus egg. As for the general statement that animals can afford their young a better chance of survival by providing them with larger amounts of yolk and therefore a longer incubation-period, there is a striking parallel here with the seeds of leguminous plants which are packed with nourishment. In the Origin of Species (6th ed. p. 56), Darwin wrote, "From the strong growth of young plants produced from such seeds as peas and beans when sown in the midst of long grass, it may be suspected that the chief use of the nutriment in the seed is to favour the growth of the seedlings, whilst struggling with other plants growing vigorously all round". It is interesting that the birds show an adaptation exactly similar to the poecilogony of the invertebrates and fishes. Tree-nesting birds are usually nidicolous, but the defenceless state of the newly-hatched squab has brought it about that ground-nesting birds are usually nidifugous. As Table 30 shows, the composition of the eggs of all animals other than those of the frog, the silkworm, and certain fishes, is still, to 3i8 THE UNFERTILISED EGG AS A [pt. m use a phrase of William Harvey's, "hid in obscurity and deep night". It is as yet much too early to try to draw any conclusions from the very fragmentary figures which are all that we have at our disposal, and we may well admit that one of the most urgent needs of chemical embryology is a much wider extension of our knowledge of the static chemistry of the egg. This is a quite indispensable preliminary to the investigation of the metabolism of the embryo in the lesser known forms. The attempt has already once been made to link up in some way the chemistry of the egg with what is known of the type of embryonic development which takes place in it. Wetzel in 1907 analysed the eggs of a sea-urchin, a crab, a cephalopod, and an elasmobranch fish. He pointed out that the eggs he studied were examples of varying richness in yolk, of total and partial, equal and unequal, superficial and discoidal cleavage, as well as chemical systems. Taking the egg of Strongylocentrotus lividus as his first case, he regarded it as typical of a class of alecithic eggs, of a total and equal cleavage type, and he drew attention to the fact that it was rich in water and in salts, but poor in fatty substances, in nitrogen, and in phosphorus. Similarly, in the case of the mollusca, where there is no very definite type of development, the egg of Sepia could not stand as representative of any wider class than the cephalopods, but, as far as it went, it showed that the cephalopod egg was rich in nitrogen, poor in fat and inorganic substances, with a moderate phosphorus and water-content. The decapod Crustacea, to which Maia squinado belongs, have a purely superficial type of cleavage, with no cell-multiplication in that part of the egg which holds the yolk. Accordingly, the egg possessed a moderate fat and water-content, a moderate ash, and much protein and phosphorus. The mammalian ovum is still as unknown chemically as it was when Wetzel was writing, and it may be found to have a constitution not unlike the alecithic echinoderm eggs. For the eggs of birds (and of reptiles, which only differ from them in having very little egg-white) Wetzel found a low protein and water-content, a high proportion of fat and ash, and a large amount of calcium and phosphorus. Here cleavage would only take place at one isolated point on the surface of the mass of food-material. In the amphibia, the richness of yolk, while much more significant than in lower classes, does not reach the level of birds and reptiles. SECT. I] PHYSICO-CHEMICAL SYSTEM 319 and this is duly reflected in the chemical composition by the moderate water-content, the high proportion of protein which is yet only double that of the fat. The case of the dogfish is again different, for there the egg is rich in yolk and the cleavage is meroblastic; thus the water is rather low, the fat rather high, the nitrogen very high, and the ash and phosphorus moderate. But these conclusions of Wetzel's, interesting though they are, cannot really be assessed until a great deal more comparative work has been done. They must rather be taken to represent the kind of correlation we may hope for in the future. However, one of Wetzel's generalisations may be accepted, if with some reserve. He pointed out that the fat-content of eggs showed great variations, rising from 12 per cent, of the dry weight of the Sepia ^gg to 66 per cent, of the dry weight of the (yolk of the) hen's tgg. Again, the nitrogen gave very variable results, rising from 5-3 per cent, of the dry weight in the (yolk of the) hen's ^gg to 6-9 per cent, in the egg of the grass-snake, 1 2 per cent, in the egg of the dogfish, and even in the case of the cod 14 per cent. On the other hand, the phosphoruscontent varied only between the (outside) limits of 2 • i per cent, for the sea-urchin tgg and 3-6 per cent, for that of the grass-snake. Wetzel, therefore, suggested that a distinction might be made, at any rate, roughly, between those constituents of the egg which may serve as sources of energy for the growing embryo, and those which in no circumstances do so. Protein, fat, and carbohydrate would come in the former class; phosphorus (for nucleoprotein) and cholesterol, for example, would come in the latter class. The former would show great variations among eggs of different species, the latter would not. He thus supposed that one might be able to deduce, as it were, the constitution of any given egg, if one knew what substances, and in what proportions, were used by the embryo as combustible material during its development, as well as the constitution of the newly born or hatched organism. From this standpoint Wetzel distinguished four types of substances in the unincubated egg : ( i ) material for the embryo to burn during the course of its development, (2) constituents of the finished protoplasm of the embryo, (3) constituents of the finished embryo, but not for incorporation into the protoplasm itself, but into the paraplasm (in Le Breton's terminology), (4) the protoplasm of the original egg-cell. No aspect of chemical embryology needs attention more 320 THE UNFERTILISED EGG AS A [pt. iii urgently than this, and the correlation of chemical constitution with developmental type should offer a most attractive field for research. But it is not only correlations of this type that lie hidden under the enigmatic character of analytical figures. The water-content of the eggs may have a powerful effect on the sex-ratio, for King found in 191 2 that reducing the water-content of fertilised frog's eggs considerably lowered the proportion of males, while increasing it by means of treatment with dilute acid considerably raised the proportion. A discussion of these facts in relation to genetics as a whole will be found in the review of Huxley. It is probable that the effect which delayed fertilisation has upon the sex-ratio is to be explained by difference in water-content of the eggs. Hertwig was the first to observe this delayed fertilisation phenomenon in some work which he published in 1905, and since then it has many times been observed not only for amphibia but also for trout (Kuschakevitsch; Huxley; Mrsic) . Riddle has suggested that the mammalian egg may be subject to such influences as it passes from ovary to uterus. He quotes van der Stricht's histological work on the bat's egg during this process, and points out that the swelling of the yolk-granules would indicate an absorption of water. The exact degree of hydration of the mammalian egg might thus conceivably have an effect on the mammalian sex-ratio. Table 30 has several more important points which have not, so far, been touched upon. It is interesting to follow in the figures of Milroy the difference between the fish eggs which float at the surface of the water during their development (pelagic ova), and those which sink, or rather float, at lower and denser levels (demersal ova) — the former have a water-content of about 90 per cent., the latter of about 70 per cent. A knowledge of the chemical composition of fish eggs throws a great deal of light upon their distribution in the sea, and so indirectly upon ecological problems. Their fat-content, for example, has been treated from this point of view by Polimanti, whose work will be discussed in the section on the general metabolism of the embryo; and the investigations of the specific gravity of fish eggs, which are discussed in Section 5, have also an important bearing upon these problems. Another point worth notice is the approximately constant percentage of cholesterol in different eggs, nearly always about 500 mgm. per cent, of the wet weight, a proportion which, roughly speaking, holds for the egg of the hen as well. SECT, i] PHYSICO-CHEMICAL SYSTEM 321 It would be as well to emphasise the fact that no principle of selection has been used in the preparation of Table 30, on the ground that results such as those of Roffo & Correa on a Brazilian gastropod, and McCrudden on fresh-water fishes, which seem obviously wrong, may not be so at all. The estimation methods and analytical processes which are by general consent judged most satisfactory at the present time cannot be considered in any way final, and to have excluded certain results on account of the technique employed in obtaining them would not have been justifiable. Table 30 does not, therefore, absolve investigators fi'om the duty of looking up the original papers in such cases as touch them most closely, and forming an independent judgment, according to the best opinion of the time, on the stress which can be laid upon them. It is needless to say that I leave out of account all doubtful figures in the generalisations made here. I -12. Egg-shells and Egg-membranes Very little is known about the relative proportions of yolk, white, and shell, in the eggs of the lower animals, or rather, in most cases, egg-contents and shell or surrounding membrane. Table 32 gives a few figures. The discrepancy between the results of Ford & Thorpe, on the one hand, and Wetzel, on the other, is very strange, especially as they both used Scyllium canicula eggs, but it is probably due to insufficiency of the statistical element. Ford & Thorpe's proportions are more likely to be accurate. Much work, however, has been done on the membranes and hard coverings which invest the unincubated eggs of diflferent kinds of animals. For instance, the gelatinous substance which surrounds the undeveloped amphibian egg was examined chemically by Brande in 1 810, who noticed that it absorbed water and was not precipitated by tannin or by strong acids. Later work has shown that it consists almost entirely of mucoprotein and water. Wetzel's figures for its weight are shown in Table 32. Giacosa isolated mucin in a pure state from it in 1882, and the figures which he obtained for its percentage composition are shown in Table 33. He was able to show the presence of a reducing sugar on hydrolysis, but he could isolate nothing else from the jelly, and therefore concluded that it was pure mucin. The presence of glucosamine in the mucoprotein was afterwards confirmed by Hammarsten, by Schulz & Ditthorn and by Wolfenden, who confirmed Giacosa's finding that 322 THE UNFERTILISED EGG AS A [PT. Ill Table 32. In % of total egg-weight Species Egg-membranes Whit( Herring Carp ... Cod ... Pike ... 2-4 3-7 4-4 4-1 — Dogfish Silkworm 5-4 26-9 8-87 (wet) 19-3 36-5 Trout ... Octopus 25-97 (dry) I3-57-20-29 86-0 — Yolk Investigator and date — Konig & Grossfeld (191 3) 3? 33 75-3 Ford & Thorpe (1920) 36-5 Wetzel (1907) — Tichomirov (1882) 14-0 3J Kronfeld & Scheminzki (1926) Ranzi (1930) Tomita's figures. Marine turtle ( Thalassochelys cortica) Weight Shell White Yolk Total m gm. 2-0 13-5 18-9 34-4 % 5-8 39-2 55-0 Wetzel's figures. I Frog {Rana temporaria) Ovarial egg (no jelly)... Egg with unswoUen jelly Jelly alone Swollen jelly ... Water content of ovarial egg jelly „ „ egg and jelly Empty dry jelly Dry egg Dry egg + dry jelly ... Thus of dry weight egg jelly Weight in mg. 1-897 4-674 2-777 8-97 0-62 0-90 1-52 Melvin's figures. Shell-weights of insects Squash-bug {Anasa tristis) Luna moth [Tropoeoa luna) Cecropia moth {Sarnia cecropia) ... Smartweed-borer {Pyrausta ainsleii) 52-5 78-65 67-48 59-28 40-72 % of total weight of eggs 29-2 23-3 22-0 31-0 it was remarkably resistant to putrefaction, and studied the effect of enzymes such as pepsin upon it. The resistance of frog ovomucin to putrefaction was for long a puzzle to biochemists, but it seems to be explained by the unwillingness of most SECT, i] PHYSICO-CHEMICAL SYSTEM 323 bacteria to grow on pure proteins, and as the jelly contains no enzymes of an autolytic character no protein breakdown products are formed, and consequently no bacterial growth takes place. This might be considered a protection of the developing embryo from bacterial attack. It is very probable, moreover, that the mucoprotein acts as a source of nourishment for the young tadpoles immediately after hatching, for they invariably attach themselves to it after they emerge from the egg-membrane, and hang on to it by their oral suckers (for histological details consult Nussbaum and Lebrun). On the other hand, development will readily proceed in the absence of the jelly, for as Hluchovski has shown it is disintegrated by exposure to ultra-violet light and may thus be removed without harming the eggs. The swelling which takes place in the gelatinous covering when the eggs are shed into the water was studied as long ago as 1824 by Prevost & Dumas, who measured the size of the eggs at intervals after they were laid. Their table is as follows : Hours after laying Diameter of egg (mm.) 2-5 1-5 5-0 2-5 6-3 3-5 7-1 4-5 7-2 5-5 7-1 6-5 7-3 They observed that dyes would pass through the jelly as soon as it had swollen, but not before. Similar work by Wintrebert on Discoglossus pinctus gave the following figures : after laying Diameter of egg (mm, o-oo 2-5 X 2-3 0-03 3-0 X 2-7 016 3-3 X 3-0 0-66 5-8 X 3-2 800 5-OX4-6 As regards the mineralogical and morphological structure of the egg-shells of the lower animals, a good deal is known, and for full detail the reviews of Prenant and of Biedermann should be referred to. The majority of reptile egg-shells have their calcium carbonate in the form of calcite, as Kelly; Schmidtt, and Meigen have shown, but the two first-named investigators discovered that the tggsheUs of chelonia were of aragonite, and later Lacroix observed a similar phenomenon in the case of certain saurians. The tgg -a a »-l ^-v a plH lO u o d V5 ,— ,0^ .§5 V cco_i B > '0 cS "C a S fe S E -= '*;i: 3 cS r3 k! u >ffio;z: |i| CO p^ i^ ^ ?, - o SOOS o bp '2 co^ cr ■" ^ fi i- -> S w JJ p U . S V! C3 O O^ O Cj :0 Ui ^ Si S2 c .E > ffi T3 ^, ^ < oac. ca V V - > > u u V hJi-I u ■^ (S CO f~. i^ r^ CD ^ 6 Th' lO m 1 1 1 ^ 1 1 1 ^ LO O) ■^ 1 1 1 1 Tj- 1 1 1 ^ 01 CO S (N <N (M tr> CO "^ -co locf J^ •^ CO CO 6 ' ' '-' 066 i-l Ci 6 V 1 CO (M y 01 CO cog CO coo <>' CO r^CT^T** in CO «  6 -H 6 « 6 6 6 T^COCOw CO ix> ■* 0^^ «3 CO to CO inOi ^ "^ lO lO io 10 tJ" N4 »-t 0^ 1 1 r^ 1 m loco 1 1 CO ^90 CO 1 1 f,.r- 1 ' ' f^f- f^ r^!0 1 1 ^"^ 1 1 1 " ^^ CO to 1 1 04 CO 1 \ 1 COM CO lO lO 10 Tf" CO II I ^ COtO COCO " lOTfOO," r^ lO lO r- f^ O CO '^'-o <-0 O O " o o O " o o 6 "^ lO lo rhco r~- 01 coto COCO " coco O) (O 10 in loto CO I oy^irS^l 2L^~ a)iocr)"-*o< I o^inpr^l ^'TP o«o»-'"" i,M«i-<i-i 000 inoj 00050 -^o o^5< ;*co eo eo inioinininin'OiO'^eo o o •* Oi r-~ o CO CO (O coco <o r^to m I T^jH in CO I 01 0) 01 irj eo CO S- - ^ lij i^ e.9 •2t3 3^ _c j^xi a, 3 o 3 -Cl -fi -TJ cj 3 S -s y .^-S 3 t

-yS .SJ^

fc. Q. lU .. peovin akdown


Paranucl Ichthulic .5 ■5 S "-S 3 <u.y .S Ui u 1— 1 G«  u P ^-5 3 (I 3 lU 3 T3 3 p c« 

§ a o.g .0 cs

c'3 g3'3^-" it^.-s^

dono us b <i "^ r^ « 


i: c 1 7. :: m « iS < r. u ^ ►- 2 TILl se sna iile CO s Rep Tortoi Grass Croco( Q. — ' < 2 fa -'-'i b ^ s . . S ,« • •"w S^ M 3 0, fcj HOUUffi • 3 «  =«  Co • W)--^ bo c « S - ^ •"CTs-i CT3 t^TJ >>u J; S « t, O^ O t« c« u e) u 11 O tJ< CO CO 01 " c^ c< C be rt ^<2 c3 « O 3 3 g 3 Ih Ci< ^ ^co .^^ 01 CI ^ OICO o w c 2 M O o " u o >S9 ►" t< " ea o w O bo JS CO '-' +3 ^ ^ (-, bD 3 o C t; O lO 10 060 O) o 0» Tt" o»<r> o o CO lO x^ ■*co o O c< ij* CI ^ U 0< CO" t~-T}<T}>C5-* C^OOO-OOCOtJ" ^66666666 0'-io»oo»cooco O ■<t' C< coco '-' t^'O C5 lO O coco 1^ I I I I CT> •* LO COK3 CD ii CTlCO >pcp O ~ CO | | | | ^ O o r^ CT) inco CO CO ■* -" lO •* ci tOl^CfOO'-iOCOOlCOI^t^'-'O in o< T^ Lo^i in lo ■^ -^ LO'-is -rt" ot tJ" 0<tJ<O0<OOOCD coto r^Loo<^cp o< io io io io LOCO " t} (V f^ C50 " CO p o( o I "4f ^^ o< I ot I r^co I ^ 'p CO'^ o UD i^ r^ •- lO LO ^ CO - CD - CD CO CDco cr> r^ co'O ■* 10 ■-■ LO CO LO t^ f^ f~(i) to o o o CJ5 o» to CT> '*'-^ LO incp to CO lO 0) CO" CO" OI^lOn «  lO'OLOtOiOiO'^'^iOiO <o - - ei C^i-D CO CO t-D f^to to rt :; ;i 3'3'S'3 J2JS "3-C U U - tJ K>l C C fl c c be— ■?!— d


"YolkIchthi ThuicI Ichthu Kerati hthu bum hthu M ^J2 ^<^ <^ nS

  • " — — — G*"*-* C —

3 3 3 3;S'3"3-S 3 „^^^-c;5x-c (S-3 cjooo'jaos/t-> bO bo « =s S g 'I O 1 : :s2 ? ? - : • * C I 1^ :2.S ,0 -2 ^ "g a c3 ^ g^'C ^r*^ ^ -_- :cr)a,05i>5 ^' « '. iS ^ C •^--' bo c bo "3 u bo ' he ' .S c «  C &,'t^ 3 ■ 2 t«  C "0 a; i- I-c IJ (- > w cs ii H(^OQ X ho< ffiUEH u bo Ki O OQ anjs — c 'op a o z ^ ^3 2, 3 c , a' ^ ^ 6^ 326 THE UNFERTILISED EGG AS A [pt. m membranes of snake's eggs which show all variations as to lime-content (see Table 9) are, as Kelly has shown, composed of amorphous and unstable calcium carbonate. The eggs of gastropods, such as Helix, Ampullaria, Bulimus, Amphidromus, etc., are, as Turpin {Helix aspersa) and Rose {Helix pomatia) , besides the workers mentioned above, have demonstrated, like those of birds in having their lime in the form of calcite. For a general theory explaining these differences see the paper of Prenant. The shells of eggs may also contain calcium phosphate. In the hen and in birds generally there is very little, but the globules seen in their egg-shells are believed to be calcium phosphate, though no analysis has given a figure of more than i per cent, of this salt. In other eggs, however, there may be more; thus Gmelin found 7-3 per cent, in the egg-shells of a tortoise, and Kelly noted its presence also in those of Bulimus and Lophohelia, though she gives no analytical figures. It is interesting to note that the mineralogical form of lime in the egg-shell may vary during the development of the embryo; thus Kelly says that the shell of many full-grown mollusca is conchite, while that of their respective embryos and eggs is calcite. Kelly found that the organic substance was a remarkably constant proportion of the shells of mollusca, reptilia and birds (see Table 9). Some eggcoverings contain almost no water at all (birds), others have more than the egg-contents, as has been shown for the trout's egg by Kronfeld & Scheminzki (membrane 75 per cent., egg 66 per cent.). By far the commonest substance of which egg-membranes are composed is keratin, though this protein seems to take many forms, and not to have exactly the same properties in different situations. The earlier workers were content to assert the presence of it on the basis merely of solubility tests. Thus in 1874 Schenk studied the egg-shell of Raia quadrimaculata, and decided that it was 95 per cent, keratin after the application to it of the protein colour reactions and an examination of its behaviour towards various solvents. The same conclusion was arrived at by the same methods by Hussakov & Welker for the egg-cases of Raia erinacea, and the Port Jackson shark, Heterodontus philippi. The keratin of these egg-cases was insoluble in all solvents except acid and alkali. They found that sulphur was present, but no phosphorus, and they were unable to find any reducing sugar after total hydrolysis. Irvine, using an optical test for chitin, found SECT, i] PHYSICO-CHEMICAL SYSTEM 327 none in elasmobranch egg-cases. Krukenberg in 1885 decided that the egg-case of Scyllium stellare was of a keratinoid nature, because of its percentage composition, in which he found a marked amount of sulphur. He observed the interesting fact that the egg-cases of this fish, while still in the uterus of the parent animal, would dissolve in pepsin and trypsin, while after they were laid they would not dissolve in solutions of either enzyme. He also isolated tyrosine and leucine firom the keratin of the egg-cases of Scyllium stellare. He made very similar researches on the egg-cases of Scyllium canicula and Myliobatis aquila, finding that they possessed rather different properties and seemed to be of different constitution; thus on hydrolysis he recovered a great deal of leucine and hardly any tyrosine from the keratin of Scyllium canicula, while from the keratin of Myliobatis the yields were precisely reversed. The latter substance was also considerably more resistant to digestion than the former, and Krukenberg considered that the former was not a keratin at all. He had already decided (wrongly, as it turned out) that the shell-membrane of the hen's egg was mucin, not keratin, and now he concluded that this also applied to the egg-case oi Scyllium stellare, as well as to that ofLoligo vulgaris, of which he made a separate examination. He thought it possible also that the jelly which surrounds the egg in the ovo viviparous selachians might be a mucin too, especially as, according to Schenk, it was not precipitated by chromic acid, and he himself found that it was extremely resistant to digestion by enzymes. This material has received no further chemical investigation since the time of Krukenberg. Other workers who identified the proteins of egg-membranes by the aid of colour tests and solubility reactions were Leuckart, who showed, as far as anything could be shown with such preliminary methods, that the membranes of planarian eggs were of chitin, and Yoshida & Takano and Jammes & Martin, who drew a similar conclusion about the coats of the eggs of Ascaris lumbricoides, which they found were readily soluble in gastric juice or in any acid.^ The case of the parasitic nematodes is of special interest, for the chitinous membrane does not arise until after the fertilisation of the egg, being, therefore, in a sense, analogous to the fertilisation membranes of echinoderms. Whether the chitin is formed as it is required during these early stages, or whether it is already present in the unfertilised egg-cell in some soluble form, is uncertain. Faure-Fremiet in an ^ See also Campbell on the chitin of insect egg-membranes. 328 THE UNFERTILISED EGG AS A [pt. iii attempt to throw light on this question, prepared pure samples of chitin from the newly fertilised eggs ofAscaris megalocephala by boiling them with strong potash, and identified the chitin chemically, isolating glucosamine hydrochloride from it. Remembering that Weinland showed that chitin is probably formed from glycogen during insect metamorphosis, Faure-Fremiet estimated the glycogen in the Ascaris eggs before and after fertilisation. Before fertilisation there was an average amount of 20 gm. per cent, dry weight, but afterwards only 4-67, the extreme values being 5-91 and 3-23, so that no less than 17 per cent, of glycogen had disappeared. Estimations of chitin in the egg-envelopes after fertilisation gave results of between 8-3 and 10-7 per cent, dry weight of glucosamine (calculated as glycogen) with an average of 9-23. The total glucose, then, in the fertilised eggs was 12-83 to 15-08, as against 20-0 in the unfertilised ones, a loss of 7 to 9 per cent. All the glucose lost, therefore, could not have transformed itself into chitin, but must have had some other destination, perhaps butyric and valerianic acid if Weinland's view is correct. The eggs o^ Ascaris have also an " ovospermatic membrane", but for the discussion of the significance of this reference should be made to the memoir of Faure-Fremiet, and nothing is known about it chemically. Their third membrane, the internal one, would seem to be composed to a large extent of ascaristerol (see p. 352), for the histological evidence demonstrates a collection of the ascaristerol globules at the periphery of the cytoplasm. After fertilisation, FaureFremiet found the saponification number of ascaristerol lowered from 199 to 145, from which he concluded that its constitution had been slightly altered. Zavadovski has also described the egg-shells of many nematodes. • Neumeister, who found more than 5 per cent, of sulphur in the shells of the reptiles, Calotes jubatus, Ptychozoon homalocephalus, and Crocodilus biporcatus, concluded that they consisted of a true keratin, and the reactions given by the egg-membrane protein of a monotreme, Echidna aculeata, led him to the same conclusion in that case also. Table 9 gives the figures which he obtained for the calcium and other constituents of some of these egg-shells, as well as the very similar investigations of Wicke & Brummerstadt on Alligator sclerops. From these fragmentary results, it would seem that the eggmembrane protein is here keratin, and a quantity of calcium is secreted into the membrane by the animal, varying in amount from SECT, i] PHYSICO-CHEMICAL SYSTEM 329 90 per cent, to 10 per cent,, according to the species. Again, the egg-membrane of the Brazihan gastropod studied by RofFo & Correa is said, on the basis of qualitative tests only, to be a true keratin, containing no reducing sugar and associated with no other substances, save 2*45 per cent, of ash. It contained calcium the amount of which did not vary during development. The transparent horny egg-membrane of the selachian Mustelus ^ laevis, which disappears half-way through the development of the ■ embryo, has also been investigated by Krukenberg, who compared it with the egg-membrane of the grass-snake, Tropidonotus natrix. The former resembled the shell-membrane of the hen's egg rather than the true keratin of the Myliobatis egg-case. The latter seemed to have some of the properties of elastin and some of those of keratin ; from it he was able to isolate a reducing carbohydrate as well as glycine, tyrosine and leucine. Krukenberg was also one of the earliest workers to make quantitative investigations on this subject. His figures for the protein of the egg-shells of Murex trunculatus and the whelk Buccinum undatum, which are given in Table 33, led him to make a new class of such substances, the conchiolins. As no data exist for the sulphur content of most of these proteins, it is impossible to say whether they are keratins or not, and the whole subject needs re-investigation. About five years later, Engel also investigated the egg-membrane protein of Murex, and, obtaining 0-5 per cent, of sulphur from it, concluded, its other properties taken into account, that it was a keratin. Engel also agreed with Hilger, whose figures for the egg-membrane of the snake, Coluber natrix (see Table 30), suggested an elastin as its principal component. He had not been able to find any sulphur in it. About the same time, Wetzel examined the conchiolin in_the_egg-shells of, Mytilus edulis, and obtained from it, after hydrolysis, leucine, tyrosine, glycine, Various hexone bases and ammonia, but no phenylalanine. The first efforts at quantitative discrimination between egg-membrane proteins were contented with ascertaining the elementary composition ; thus von Fiirth analysed the protein of Loligo vulgaris eggs in this way (39 per cent, glucosamine), and Verson, and later Tichomirov, decided that the egg-shell of the silkworm, Bombyx mori, was a keratin-like body (3-7 per cent, of sulphur), though, owing to its unusual properties, they called it chorionin. Of these two lastnamed analyses, it is probable that Tichomirov's is the more accurate. 330 THE UNFERTILISED EGG AS A [pt. m for he was more careful to remove all the adhering silk than was Verson, and Farkas' independent work agrees rather with his. It is, at any rate, clear that the shell-substance of the silkworm's egg is not chitin. According to Lavini the inorganic constituents of the silkworm egg-shell are potassium silicate, sulphate, and carbonate, to the exclusion of all other salts. The work of Pregl and of Buchtala in 1908 is perhaps the most thorough investigation of the amino-acid distribution of an eggmembrane protein. The figures they obtained are given in Table 34. The keratin of the egg-case' of Scyllium stellar e was the only one of which they made a complete amino-acid analysis ; for that of Pristiurus melanostoma and Scyllium canicula they only determined the cystine content and large groups such as the monoamino-acid nitrogen. Scyllium ovokeratin seemed to follow very closely in its constitution the ovokeratin of the hen, according to the figures of Abderhalden & Ebstein, which have already been discussed, but separated itself off very sharply from it on account of its high tyrosine content. The ovokeratin of the tortoise Testudo graeca, which had been investigated two years previously by Abderhalden & Strauss, was again different, having no tyrosine, but a very high percentage of proline. As far as this work goes, it would seem right to conclude that, though the eggs of different species may use similar proteins in their external membranes, the constitution of these proteins may vary very considerably. The work of Steudel & Osato, and of Osato, however, brought a new factor into the problem. Their analyses of the egg-membrane protein of the herring's egg, which are shown in Tables 34, 38 and 39, gave results which differed from the usual keratin figures, but which very closely approached the analyses which they were making at the same time of the ichthulin of the herring's egg. Thus the amide nitrogen (2-05 per cent.) was lower than any of the keratins, but approximated instead to the i-8i per cent, of herring ichthulin. What appeared to be the case on a general survey turned out to be certainly so when the amino-acid distribution was examined, for the two sets of figures almost exactly corresponded. The properties of the eggmembrane protein and the minute amount of sulphur in it precluded its classification as a keratin, and the fact that no reducing sugar could be discovered among its breakdown products was convincing evidence against its being a mucin. Osato suggested that it was 330* Table 34. Distribution of amino-acids in egg-proteins. Amide N ... Glycine Alanine Valine Leucine Proline Phenylalanine Aspartic acid Glutamic add Serine Tyrosine Cystme ... Hiiiudinc ... Arginine Lysine ... Tryptophane Humin Unidentified Total di-amino Total mono-am Non-amino N c ii Scl U tl Present 19'40 — 53-73 2-76 4.4 _ ^ ^Jolie O'bo 103 10-6 021 0'44 0-30 i-H'i 114 '•7 0-32 '3'5l' l-o6 3-2 '■45 io-i6 0-29 3-7 — — 9-B6 — None Present Present Present Present Present 50-7 — 2-56 9-04 Present 13-72 3-8o 0-37 4-i6 — None 6 3-8 3-66 Ii-ig o-t 0-93 0-62 0-34 None 82 8-83 Present 'None 92 n-og o-ig O'SS 07 8-33 0-39 0-22 l-8l 2-05 7-56 — ■■77 — — — 6 1 -55 3.90 None SECT, i] PHYSICO-CHEMICAL SYSTEM 331 simply an insoluble modification of ichthulin. As he pointed out, industrial use has long been made of insoluble forms of proteins, such as casein, and there was no reason why the egg-membranes of certain eggs, at any rate, should not be insoluble modifications of the proteins of their yolks. Steudel & Osato also suggested that the ovomucoid of the egg-white of the hen might be a phylogenetic reminiscence of the mucoprotein with which the amphibian egg is surrounded. For a review of this work see Steudel. The eggs of salps and tunicates are surrounded by a coat of very much smaller cells which act as some sort of protection for the developing embryo inside. Zavattari has demonstrated histochemically the presence of an abundance of glycogen in these test cells, and believes that they have a nutritive function. If so, this would be a third case where such an active participation of the shell or case in embryonic metabolism would have been noted, the two others being the abstraction of calcium from the shell of the hen's egg, and the contribution of amino-acids by the egg-case of the silkworm. A good deal is known about the osmotic and other properties of the membranes of amphibian and fish eggs, but these are so intimately associated with the physico-chemical processes taking place during development that consideration of them will be postponed to Section 5. It will suffice to mention here the experiments of Peyrega, who found that the egg-cases of Scyllium canicula were permeable to salt. He fitted up osmometers with small pieces of the case as the membranes, and observed that it took about 20 days to establish osmotic equilibrium with respect to solutions of sodium chloride about as strong as sea water, when distilled water was put on the other side. These egg-cases have also been shown by Needham & Needham to be permeable to urea and ammonia. 1-13. Proteins and other Nitrogenous Compounds The principal protein substance which is found to occur in the eggs of all known animals closely resembles the vitellin of the hen's egg. It has even been found, according to Chatton, Parat & Lvov, in the food-reserves of infusoria. The early analyses of the eggs of the pike by Vauquelin in 181 7, of the barbel {Cyprinus barbus) by Dulong d'Astafort in 1827, and of the trout [Salmo fario and Cyprinus carpio) by Morin in 1823, ^^^ to no more than the view that an albuminous substance w£is present in them. But with the work of 332 THE UNFERTILISED EGG AS A [pt. iii Gobley on the hen's egg, which has already been described, a more solid basis for comparison was achieved, and Valenciennes & Fremy, in a memoir which received a prize from the Academy of Sciences and which was translated into English, proceeded to examine the eggs of as many species as were available to them. Gobley's only excursion into comparative chemical embryology had been a detailed analysis of the carp's egg, published in 1850, but he had not been slow to point out the differences between this analysis and that of the hen's egg. His figures are shown in Tables 2, 30 and 33, where it will be seen that he got a value of 15-76 per cent, protein (wet weight) for the hen, and 14-23 per cent, for the carp, but 31-43 per cent, fat for the hen and only 2-57 per cent, fat for the carp. The carp's egg had, he found, about 10 per cent, more water than the yolk of the hen's egg, but only a third of the lipoid substances. Fremy & Valenciennes specially directed their attention to the protein fraction, and attempted to discover whether the vitellin was the same in all eggs. For the most part they relied on histological appearances (the "dotterplattchen" were greatly discussed at this time), but they also examined the solubility relationships of the proteins from each egg, and in some cases subjected the purified substances to elementary analysis. The figures they obtained for the different compounds are all given in Table 33, and the eggs they investigated in Table 35. They were able to isolate a number of vitellin-like proteins, soluble in salt solution and precipitated by the addition of water. They compared vitellin with fibrin, and concluded that the two substances were almost identical, in spite of slight differences in the analytical figures — "for bodies of this nature", they said, "which are not crystallisable and insoluble in water and which are therefore very difficult to purify, where is the chemist who could answer for i per cent, of nitrogen in an elementary organic analysis?" Ichthin, which they isolated from fish eggs, differed from vitellin by not becoming an opaque mass when placed for a long time in boiling water, and by giving a violet instead of a blue colour when treated with boiling hydrochloric acid. Ichthidin, another product offish eggs, differed from ichthin in being soluble in water. Ichthulin, the third member of the group, differed from the others in not being soluble in all dilutions of saline, but in being precipitated from the aqueous extract by further addition of water. As for emydin, it closely resembled ichthin, and it is SECT. l] PHYSICO-CHEMICAL SYSTEM 333 not easy to see why Valenciennes & Fremy did not identify it with that substance. The remaining egg-proteins, which they did not further investigate, they referred to under the generic name of albumen. Table 35. Investigations AvES Callus domesticus Pisces of Valenciennes & Fremy. Vitellin Elasmobranchs Raia clavata Ichthin Torpedo martnorata >j Scyllium canicula 99 Galeus canis jj Alustelus laevis 99 Squatina angelus 99 Raia fullonica >j Raia rubus j> Teleosteans Cyprinus carpio Ichthidin and ichthulin Labrax lupus Ichthulin and ichthidin Alugil chelo Scomber scombrus Pleuronectes maximus Pleuromctes solea Solea armorica Unidentified species of salmon >) eel Albumen Reptilia Testudo mauritanica Emydin Cistudo europaea jj Unidentified species of lizard Vitellin jj grass-snake )> >> viper „ (?) Amphibia }> frog Ichthin >> newt 5> Crustacea jj lobster Albumen Ar-achnida and Insecta — — Albumen MOLLUSCA — Not albumen The differences between the compositions which Valenciennes & Fremy found for these substances are not great, and it is very doubtful whether they are more than modifications of the same substance, especially as these workers admittedly had great difficulty in obtaining pure preparations. But the problem of the identity of the vitellins is not yet settled. The later investigations are all grouped together in Table 33, and the differences between the preparations can easily be seen to be small. The work of Plimmer & Scott proved that ichthulin is a phosphoprotein closely allied to vitellin. Among the more interesting observations must be mentioned those of Levene & Mandel; Levene, and Walther, on ichthulic acid obtained from the ichthulins of various fish eggs by digestion 334 THE UNFERTILISED EGG AS A [pt. iii with pepsin and other methods. These with their very high phosphorus content approach closely the " paranucleins " or vitellic acids obtained from the vitellin of the hen's yolk by Levene & Alsberg and others. Evidently there are several possible stages of breakdown, for Walther's ichthulic acid only contains 2-8 per cent, of phosphorus, while that of Levene has as much as 10-4. Here, also, however, there are great variations; thus, while nearly all the ichthulins studied have from o-6 to i -9 per cent, of phosphorus, the preparation of Steudel & Takahashi from the herring's egg has only 0-014 P^r cent. In the yolk of a dogfish egg, Zdarek found no less than three proteins, the third of which may possibly correspond with Konig & Grossfeld's albumen class. In 1908 Alsberg & Clark claimed that phosphorus was quite absent from the principal protein of the egg of an ovoviviparous selachian, Squalus acanthias, but some twenty years later I re-examined the question and obtained without difficulty o-6 per cent, from selachian ichthulin (derived from the same species). This yolk also contains a second protein, thuichthin, corresponding closely in properties and constitution with the ovolivetin of the hen studied by Kay & Marshall (see Tables 10 a and 33). Gray has studied the properties of the ovoglobulin or ichthulin of Salmo fario. If the yolks are poured into water, a dense white clot is formed and the water becomes cloudy. The precipitate is soluble, however, in acids, alkalies and neutral salts. When the egg-cell dies, the egg becomes opaque, and this must certainly be due to the precipitation of the globulin, for by placing dead white eggs in normal sodium chloride solution they rapidly become clear and resemble normal eggs, but regain their opacity when removed to distilled water. The clearing process takes 15 minutes but the precipitation takes i| hours. Evidently the dead protoplasmic membrane can no longer retain in the egg the electrolytes necessary for solution of the ichthulin. Further work on the properties of teleostean ichthulin was done by Runnstrom. 1-5 parts of egg "Pressaft" having been added to 1-28 parts of water and the ichthulin precipitated, the effect of various ions on its solubility was tried. The anions placed themselves in the order: SON > I > NO, > SO. > CI > acetate. SECT. I] PHYSICO-CHEMICAL SYSTEM 335 Thus for 2 c.c. of potassium chloride solution, 0-3 c.c. of distilled water had to be added to get coagulation, but to 2 c.c. of KSCN solution, as much as 6-4 c.c. The cations went as follows: Ca > Mg > Sr > K and Na. The egg-white of the dogfish egg was thought by Brande in 1810 to be identical with the jelly surrounding the egg of the frog, but whether the former really consists of mucin and not albumen cannot be definitely stated, for no work has since been done on it. However, my wife and I, in our work on the eggs oi Scyllium canicula, frequently observed a coagulation of the egg-white with acetic acid, which would point to the latter possibility. The proteins of the echinoderm egg have never been properly investigated. Vies, Achard & Prikelmaier have estimated from cataphoresis experiments that the average isoelectric point of the Paracentrotus lividus egg-proteins lies between 5-0 and 5-8 pH., but their grounds for this figure are not free from criticism. Vies & Gex, in some interesting experiments, have studied the normal unfertilised sea-urchin's tgg spectrophotometrically. The absorption spectrum of the normal egg has peaks or bands at wave-lengths of 490, 395, 370, 315, and 230 Angstrom units, and a marked trough between 260 and 240 A. This curve is very peculiar, for on the one hand it shows much transparency in the ultra-violet although most organic substances do not, while on the other hand there is nothing at all corresponding to the bands of absorption about A 275 which all proteins give. This absorption is brought about by the cyclic amino-acids in the protein molecule, and it is quite impossible that these should be altogether absent from the egg-proteins of the sea-urchin. Vies & Gex considered various technical possibilities which might explain these effects, but did not think that any of them would account for what was perhaps the most remarkable part of the investigation, namely, the finding that on cytolysis ("white") a perfectly definite and clear absorption spectrum for protein revealed itself In the intact egg, then, this must be masked by something else. Speculation on the nature of this mechanism would be easy, for all kinds of eflfects might be responsible, e.g., formation of complexes, reduction equilibria, and satisfaction in vivo but not in vitro of residual valencies in the protein molecule. If this very interesting work should lead in the future 336 THE UNFERTILISED EGG AS A [pt. iii to a revivification in a subtler form of the old biogen molecule theory (though it is to be hoped that it will not), not only as regards the egg-cell but as regards protoplasm in general, we shall at any rate possess in the spectrophotometer a powerful means of studying the untouched normal cell-interior. Doubt exists with respect to the presence of reducing carbohydrate in the ichthulin molecule. Levene & Mandel obtained minimal quantities of laevulinic acid from their cod ichthulin, but this finding was associated with the presence of purine bases. Six years earlier Levene had been unable to find a trace of glucosamine in cod ichthulin. Similar negative results were obtained by Steudel & Takahashi on herring, and by Hammarsten on perch, ichthulin. But the presence of glucosamine in notable amounts has been reported for Torpedo ichthulin by Rothera, and for carp ichthulin by Walther. While it is possible, and even probable, that ichthulins from different fish eggs may vary much, it would be very desirable to know to what extent this is the case, and a comparative study of ichthulins is much needed. As we have seen Levene & Mori have isolated a trisaccharide from avian vitellin. Closely allied to the question of the presence of carbohydrate groupings in the ichthulin molecule is the equally disputed problem of the presence of purine bases in the undeveloped tgg. We have already seen that Miescher's identification of nucleoprotein with vitellin was quite erroneous, and have described how he was set right by Kossel. For the hen's &gg, it is now fairly clear that nucleins are present only in exceedingly small amounts at the beginning of development, not exceeding, for instance, i or 2 per cent, of the total nitrogen or phosphorus. But there has been more difficulty in deciding what is the real state of affairs in the eggs of fishes and aquatic invertebrates. Walther (carp), Hugounenq (herring), Linnert (sturgeon), and Hammarsten (perch), all examined the ichthulin of these eggs for nucleic acid, and all failed to find the least trace of it. Henze, on the other hand, working with the whole tgg of the cephalopod. Sepia officinalis, isolated considerable amounts of purines together with no less than 1-15 gm. per cent, of a pentose. Tschernorutzki a little later found that 10 per cent, of the total phosphorus of the herring's egg could be accounted for as nucleoprotein phosphorus, and the nucleoprotein itself amounted to i-ig gm. per cent, dry weight. Masing; Tichomirov, and Needham & Needham reported SECT, i] PHYSICO-CHEMICAL SYSTEM 337 quite similar results with the sea-urchin's egg, the egg of the silkworm and the eggs of various Crustacea, echinoderms and an annelid. In the sea-urchin egg purine bases were found accounting for 6 per cent, of the total nitrogen as nucleoprotein nitrogen, while in the case of Bombyx there were 20 mgm. per cent, dry weight. Again, Levene & Mandel isolated from their ichthulic acid in 1907 0-344 P^^^ cent, of guanine, 0-307 per cent, of adenine, 0-360 per cent, of uracil and 0-309 per cent, of thymine. Mandel & Levene were also able to isolate nucleic acid from cod's eggs. It would certainly appear from this evidence as if ichthulin and vitellin may be associated with small quantities of nucleic acid. In this connection it is of interest that Calvery has evidence that the chick embryo can synthesise "yeast-" as well as animal nucleic acid. Steudel & Osato have also obtained guanine and adenine from herring's eggs, but this was in the nonprotein nitrogen fraction, and there was therefore no evidence from their work that any preformed nucleic acid was a constituent of the egg. The most exhaustive investigation of the problem was that of Konig & Grossfeld, who in 1913 set out definitely to clear up the discrepancy. As perhaps might have been expected, they found that they could isolate purine bases after hydrolysis from all the fish eggs they studied, but only in small quantity; their results are shown in Table 36. The question of nuclein synthesis by the developing embryo will be discussed in relation to these findings in Section I0'3. Table 36. Investigations of Konig & Grossfeld. Total purine bases isolated Mgm.% dry weight Herring Carp Cod ... Pike ... Sturgeon 0-408 I -060 2-440 0-014 0-230 But the exact relationship between the nuclein and the vitelHn remains exceedingly obscure. It is possible that in one and the same egg there may be more than one modification of vitelHn, apart altogether from the insoluble form suggested by Steudel & Osato. All the knowledge that we possess at the present time 338 THE UNFERTILISED EGG AS A [PT. Ill on this point is of an unsatisfactory histological nature, and any discussion of it must inevitably include an unprofitable proportion of guesswork. Thus, Jorgensen differentiated histologically between two substances which seemed to be present in the unripe egg of Patella vulgata, ergastoplasm No. i and ergastoplasm No. 2, one at least of which was responsible for the formation of the vitelline globules. Faure-Fremiet & Garrault identified ergastoplasm No. i with the mitochondria, and ergastoplasm No. 2 with the fatty constituents of the yolk. But if two forms of vitellin existed, one in loose combination with a nuclein and the other free, the staining reactions of histological elements mainly constituted by one or other of these Table 37. Phos Millon Trypto Glucos Investigator Protein Iron phorus test phane Sulphur amine and date Ichthulin None Much Positive Negative Present None McCrudden (1921) Albumen Traces Present ,, ,j Much ,, Ichthulin None A little Negative )> None >j Albumen ,, Much ,, Present Ichthulin — Present — — None Levene (1901) ,, Present J, — — Present Walther (1891) ^^ — — — — None Hammarsten (1905) " — Present — — — Valenciennes & Fremy (1854) ,, — jj — — >j — Gobley (1850) substances would very likely differ, and it is possible that an explanation on these lines may in the future correlate the chemistry with the histology of the yolk. The vitellin question has been in a measure reviewed by McCrudden, whose table (given in Table 37) illustrates the difficulty of summing up the findings of investigators at all succinctly. The amino-acid analyses (Table 34) are rather more interesting. We have data for the vitellins of the herring, the trout, the cod, and the sturgeon among fishes, the frog among amphibia, the grasssnake among reptiles, and Hemifusus tuba, a gastropod. To this may be added amino-acid analyses of the mixed egg-proteins of the seaurchin egg and the eggs of the brook-trout and the giant salamander, as well as the albumens of cod and sturgeon and the mucoprotein of Hemifusus. If the fish ichthulin analyses of Iguchi or Hugounenq be compared with those of Table 1 1 for the vitellin of the hen, no very marked differences can be observed, although the predominancy SECT, i] PHYSICO-CHEMICAL SYSTEM 339 of arginine and lysine over histidine, which is a constant feature of the ichthulins, reaches greater values in the latter than in the case of bird vitellin (see Table 38). Again, bird vitellin always shows a notable proportion of proline and leucine, and this is also the case with the vitellins of the lower animals (e.g. 10 per cent, of leucine in gastropod vitellin, 19 per cent, in snake vitellin and 9 per cent, in herring ichthulin), though the amount of proline is usually not so great. The only instance of a real divergence between bird and other vitellins would appear to be the glutamic acid content, which is always high in the former, although this amino-acid is absent from the latter. Table 38. Hexone bases of yolk-proteins. In gm. % original In % total nitrogen protein Investigator Species Protein Hist. Arg. Lysine Hist. Arg. Lysine and date Herring Ichthulin 2-45 I4"50 10-07 I'^S 6-33 7-40 Steudel & Takahashi (1923) Egg-menibrane 3-99 14-41 7-51 2-09 6-35 5-55 Steudel & Osato (1923) protein Hen Vitellin — — — i-go 7-46 4-81 Osborne & Jones ( 1 909) Herring Ichthulin 0-40 2-70 2-00 — — — Hugounenq (1904) (clupeovin) Sturgeon Ichthulin 0-47 0-97 o-oi — — — Konig & Grossfeld (1913) Cod Ichthulin 0-55 0-54 0-02 — — — ,, ,, Trout Ichthulin 0-54 0-41 o-oi — — — ,, ,, Gastropod Ichthulin None 3-73 0-86 — — — Komori (1926) Frog Vitellin 1-14 1-06 0-29 — — — Galimard (1904) (ranovin) Snake Vitellin 0-30 0-32 1-45 — — — ,, If now Table 39 is considered, it will be seen that variations are present in the general analysis of these proteins, but that they tend to cancel each other out among the groups. Thus the mono-aminoacid/di-amino-acid ratio is very constant indeed in different ichthulins, although Rothera himself considered that he was dealing with two entirely different proteins, the vitellin of the Torpedo egg and that of the sturgeon. It is unfortunate that Komori's examination of gastropod vitellin was confined to the estimation of the amino-acids by isolation, and did not include a van Slyke determination of the relative amounts of mono-amino and di-amino acids. In contradistinction to the ichthulins, the mixed egg-proteins studied by Russo and Gortner show more variation, though the former's values for two sea-urchin ^gg proteins agree well with the usual vitellin figure. Masing, however, was not able to find any phosphoprotein phos 340 THE UNFERTILISED EGG AS A [PT. Ill phorus in sea-urchin eggs, and Needham & Needham found only very little. It is interesting to note that the ratio is subject to large fluctuations among the keratins of the egg-cases. As for the albumens which Konig & Grossfeld isolated from the eggs of the sturgeon and the cod, they seem to approach in their composition, in so far as data for the hexone bases permit one to form a conclusion, the ovoalbumen in the hen's egg. The 8 per cent, of tyrosine obtained from the sturgeon ovoalbumen is, however, remarkable. The mucoprotein which Komori found around the eggs of the gastropod Hemifusus tuba, and which he partially analysed, is not sufficiently well characterised to be compared except roughly with the mucoprotein of the amphibian egg-jelly. Table 39. In % total nitrogen Species Torpedo {Torpedo marmorata) Sturgeon Dogfish (Scyllium stellare) „ (Pristiurus melanostoma) ,, (Scy Ilium caniculd) Hen Herring ... Sea-urchin Brook-trout Giant salamander Hen Protein Ichthulin Ovokeratin Ichthulin Egg-membrane protein Mixed eggproteins (total) Mixed egg-proteins (coag. only) Mixed eggproteins Vitellin (for i comparison) 15-67 1609 i5-o8 14-33 14-23 1 6-43 14-09 ipz 8-49 849 1-26 6o-20 9-51 5-09 5-13 449 660 i-8i 0-99 0-56 0-14 0-24 0-21 63-60 79-66 66-45 64-19 73-70 61-77 25-10 27-65 15-78 28-78 30-75 2050 27-02 2-40 2-30 5-04 2-31 209 3-55 2-29 Investigator and date Rothera (1904) Buchtala (1908) Steudel & Takahashi (1923) Steudel & Osato (1923) Russo (1926) 7-33 2-05 — 62-11 25-91 2-40 — 284 45-70 17-30 2-64 — — — 62-20 29-80 209 ,, 1-82 — — 61-55 28-25 2-18 Gortner (1913) 2-25 — 1-63 S-55 53-73 29-35 1-83 67-10 25-10 2-67 Plimmer (1908) The general distribution of nitrogenous substances in the eggs of the lower animals is shown in Tables 40 and 41. Pigorini's investigation of the silkworm egg is suggestive, but his data about the different protein fractions are insufficient to enable us to form any judgment on their relation to those so well known in the bird's egg. The very large amount of mucoprotein in the silkworm ovum is certainly remarkable. In Table 41 are placed the few data which we have on the relative amounts of protein and non-protein nitrogen in different eggs, and the way the protein is divided between keratin, albumen, and ichthulin or vitellin. Clearly enough there is great variation, and a rough dichotomy into two groups, one in which the SECT. I] PHYSICO-CHEMICAL SYSTEM 341 non-protein nitrogen accounts for from 14 to 35 per cent, of the total nitrogen, and one in which it only accounts for less than 10 per cent, of the total nitrogen. It is evident from the work of Konig & Grossfeld that all the fishes examined belong to the first of these categories, although within the group there are wide divergences, such as the minute amount of albumen apparently present in the trout's egg and the low non-protein nitrogen of the herring's egg. Good agreement is to be noted between the results of Levene and Konig & Grossfeld, who all worked on the cod; and, although nothing concerning the non-protein nitrogen can be gathered from the figures of Kensington and Hugounenq, their results do show general agreement as regards the partition of nitrogen among the proteins. The only reptile on whose eggs work has been done which could be incorporated in the table is the grass-snake, and there, although no non-protein nitrogen figures are available, it is interesting to note the very high proportion of keratin. Table 40. Silkworm (Bomfryx: mon). (Pigorini, 1Q23.) In % of total protein A Protein sol. in water but not Protein sol. in Protein sol. in Protein sol. in coagulable by heat, and distilled water 10 °„ salt sol. dilute alkalies yielding glucosamine on (albumen) (vitellin) (nucleoprotein) hydrolysis (ovomucoid) 29-20 8-57 11-45 5090 The second principal group, consisting of those eggs which have a relatively much lower percentage of non-protein nitrogen, contains two members, the hen and the silkworm. The former may be said with a high degree of probability to be characteristic of all nidifugous birds, and perhaps of nidicolous ones also, but whether the latter is at all representative of the centrolecithal insect eggs may be considered doubtful. The sole insect egg which has been investigated chemically, so far, is that of the silkworm, and until more evidence is available the hen and the silkworm will have to be placed together in this second group without comment. It is significant that, in the hen's case, the percentage of albumen is greater than in any other, a fact obviously referable to the large amount of egg-white present in that egg. Finally, it is of interest that the sea-urchin's egg seems to have a protein/non-protein nitrogen ratio very like that of the fishes, but situated on the low protein edge of their limits of variation. 342 THE UNFERTILISED EGG AS A [pt. iii Table 41. Distribution of Gm. % wet weight Species Water Protein (Nx 6-25) Protein (by diff.) Protein of eggmembrane Albumen Ichthulin Nitrogen (direct) Free bases and amino acids Fat Ash Carp ... 66-15 2770 2997 363 16 43 4-432 9-91 2-48 1-40 Pike 6353 28-13 33-01 375 2-38 17-29 4-500 9-59 1-40 2-06 Trout 6385 27-81 30-81 1-76 0-15 24-33 4-450 457 3-71 1-63 Herring 69-22 26-32 2521 3-20 4-83 13-68 4-212 3-50 4-19 1-38 Cod 72-10 23-02 24-44 2-57 2-70 11-47 3-683 7-70 1-33 2-13 Salmon — — — — — — — — — Herring 65-00 28-ss — 0-79 28-70 — 3-62 Sea-urchin (Strongylo centrotus lividus) Silkworm 80-50 6-47 35-00 10-20 3-00 3-90 2-25 66-24 22-00 — — — 3-67 0-875 — Grass-snake (Tropido notus natrix) Cod 58-94 94-67 19-24 — 12-71 0-72 5-81 — Fresh- water gar 5390 26-20 — — — — 0-138 Hen (average results) whole egg Turtle (Thalassoclielys corticata) yolk — — 11-81 0-45 6-13 5-23 2-91 0-500 0-033 —

  • With so % mucoprotein and

Within the non-protein nitrogen fraction itself there are some fragmentary data for the distribution, as may be seen from Table 42 . Unidentified compounds usually account for from 20 to 35 per cent, of the total non-protein nitrogen, and free amino-acids for approximately half of it. Among those identified by Steudel & Osato were histidine, arginine, lysine and cystine. The ammonia may vary from 4 to 25 per cent., and the purine bases from 15 to 40 per cent. As far as can be seen at present, the hen's egg seems to possess the greater part of its non-protein nitrogen in the basic fraction. The most interesting point brought out by the table is probably the significant quantity of urea shown to be present by the analyses of Steudel & Osato, amounting to no less than half of the total nonprotein nitrogen, and it is possible that a good deal of the unidentified nitrogen of Konig & Grossfeld might be accounted for in this way. The presence of nitrogenous excretory products in the undeveloped egg, though at first sight paradoxical, is nevertheless undoubtedly a fact in the case of some aquatic organisms. The hen's egg contains hardly a trace of urea at the beginning of development but that of a selachian fish contains a good deal (see Section 9- 1 1 ) . SECT, i] PHYSICO-CHEMICAL SYSTEM 343 the nitrogen in eggs. Gm. % dry weight (ash free) ^ A . ^ % of the total nitrogen Egg , — ^ ■ — \ Pro- mem- Free Free tain Pro- brane Nitro- bases and Protein Ker- Albu- Ichthu- amino- (N x tein (by pro- Albu- Ichthu- gen amino- Investigate total atin men Hn acids 625) diflF.) tein men Hn (direct) acids Fat Ash and date 66-9 121 548 331 85-37 92-37 1119 50-64 13-65 30-54 7-64 — Konig & Gr (1913) 70-96 11-35 721 52-4 2904 82-75 95-93 10-90 6-92 50-25 13-08 27-87 407 — „ 85-20 15-7 0-49 790 1485 80-56 89-25 5-10 0-44 70-48 12-89 13-24 IO-7S — 86-15 12-70 9-15 543 13-9 89-52 85-75 10-88 16-43 46-53 14-33 "-Qi 14-25 — 68-41 10-50 i-oi 469 315 8933 94-84 9-97 10-48 44-51 14-29 29-88 5-55 — „ (loo-o?) — 11-70 880 — 87-80 — — 10-30 77-5 — — 4-50 7-50 Kensington ( (loo-o?) 2-7 973 — __ — — — — — — — Hugounenq 62-0 — — — 379 — _ — — — — — — — Russo (1926) 94-0 — 27-4 8-05 6-04* — — — — — — — — — Monzini(i9: Pigorini (ic 96-0 — — — 3-98 65-25 — — — — — — — — Russo (1922) (loo-o?) 66-0 3-74 30-2 — — — — — — — — — — Galimard (19 66-0 — — — 33-0 68-09 — — — — — — — — Levene (i89( _ ___ — _________ Nelson & Gi (1921) 96-4 406 49-8 42-5 366 — — — — — — — — — — Q8-5 — — — 1-5 — — — — — — — — — Tomita (1921 chorionin in addition. As is well known, these fishes have a special relation to this substance. In 1858 Stadeler & Frerichs isolated "kolossale Quantitaten von Harnstoff" from the organs of plagiostomes, obtaining a solid mass of urea nitrate when they added nitric acid to their final concentrates. One liver of an adult Scy Ilium canicula gave them 2 oz. of urea, and similar high figures were reported for Acanthias vulgaris. Teleostean fishes, however, and the cyclostome, Petromyzon planeri, yielded practically no urea, at any rate not more than would be present in mammalian tissues. Stadeler confirmed the selachian results on Raia batis and clavata and on Torpedo marmorata and ocellata. In 1 86 1 Schulze repeated and confirmed Stadeler's work on Torpedo, and in 1888 Krukenberg published an extensive work on the subject, in which he related his unsuccessful attempts to demonstrate urea in the bodies of teleosts {Lophius piscatorius. Conger vulgaris, Acipenser sturio), a cyclostome [Petromyzon fluviatilis and Ammocoetes) and a cephalochordate (Amphioxus lanceolatus) , although he found large amounts of it in the bodies of elasmobranch fishes [Scyllium stellare, Mustelus vulgaris and laevis, Acanthias vulgaris, Squatina angelus, Torpedo marmorata, Myliobatis aquila) and in the holocephalic Chimaera 344 THE UNFERTILISED EGG AS A [pt. iii monstrosa. Particularly interesting were his experiments with eggs — he isolated considerable amounts of urea from a 5 cm. embryo of Mustelus laevis, and from the yolk of Scyllium stellare and Myliobatis aquila eggs, but he could find none in the surrounding jelly or "white ". An Ggg ofPristis antiquorum yielded 3920 mgm. per cent, (wet weight) and a Torpedo ocellata egg 1 740 mgm. per cent. An Acanthias vulgaris embryo 1 7 cm. long had 3360 mgm. per cent, in its muscles, 1800 mgm. per cent, in its liver, and 2640 mgm. per cent, in its unused yolk. Other work on urea in selachians was done by Grehant and by Rabuteau & Papillon. Table 42. Distribution of non-protein nitrogen in eggs. % of total non-protein N (including purine N) g-| z § iz g 2 -g I I |Z w.« „ „ ^ 2^ c S "G 2 o op Investigator Species HSo? cq <fe^ P U D U hUa and date Herring — 198 — 44-3 359 — — — — — Konig & Grossfeld (1913) Carp ... ... — 39-8 — 36-1 24-1 — — — — — >> >, Sturgeon ... — 25-2 13-6 55-4 189 — — — — — ,, ,, Herring ... ... 2060 244 67 21-6 — 519 None 18-3 — — Steudel & Osato (1923); Steudel & Takahashi (1923) Herring 1443 16-91 23-42 41-65 1802 — — — — — Yoshimura (1913) Silkworm ... 440 — 4-44 54-30 34-60 — - — — 610 6-7 Russo (1922) Hen (aver, figures) — 88-80 4-22 7-04 — None None Trace — — — Fresh- water gar 299 92-00 — 4-02 — — — 4-0 — — Nelson & Greene (1921) (not ripe) More light, however, was thrown on the reasons for this richness in urea when in 1897 Bottazzi working on the osmotic pressure offish blood, found that the elasmobranchs differed fundamentally from teleosts in being isotonic with sea water. Serum A Selachians Torpedo marmorata —2-26° Trygon violacea —2-44° Teleosteans Charax pimtazzo —1-04° Serranus gigas — i -03° Bottazzi observed that the selachian osmotic pressure would correspond to some 3-9 per cent, sodium chloride but laid no emphasis on the fact that selachian blood did not contain anything like so much ash. It was left for Rodier to show that the difference was made up almost wholly by urea. Duval has since found that the salts alone would only give an osmotic pressure of A — i-o6°. "High bloodurea", as Smith says, "is a phyletic character of the orders Selachii SECT, i] PHYSICO-CHEMICAL SYSTEM 345 and Batoidei", and its osmotic function was well shown by the reciprocal relation between salts and urea which Smith found to hold in selachian tissues and fluids. Blood-urea mgm. % Smith (1929) Selachians Dogfish {Mustelus canis) 880 Denis (1913) Selachians Dogfish {Mustelus canis) 800 Sandshark {Carcharias littoralis)... 1000 Skate {Raia erinacea) 868 99 Teleosteans Mackerel {Scomber scombrus) 86 Goosefish {Lophius piscatorius) ... 40 Flounder {Paralichthys dentalus) 46 In view of all these facts it is not surprising that Needham & Needham in 1928 found about 5 mgm. of urea nitrogen present in the Scyllium canicula egg at the beginning of development ; and 888 mgm. per cent, of urea in the undeveloped Acanthias vulgaris egg. Gori, again, found 7 10 mgm. in undeveloped Torpedo eggs. But since urea accumulation is closely confined to elasmobranchs it is unlikely that the results of Steudel & Takahashi and of Konig & Grossfeld can be interpreted as being due to urea. The presence of urea has also been reported in the undeveloped eggs of "ants and flies" (in small quantities) by Fosse. Further details would be desirable here. There is reason to believe that nitrogenous substances other than those already mentioned are present in certain eggs. Thus Yoshimura and Poller & Linneweh isolated trimethylamine, tetramethylenediamine and choline from fresh herring eggs, and there is a certain probability that fish eggs also contain betaine. As the characteristic smell of fish is due to these amines and related substances, this is not very surprising. Brieger is said to have found neuridine in fish eggs, and Schii eking isolated spermine from echinoderm eggs in 1903. Taurine and glycine were found in echinoderm eggs by Kossel & Edlbacher. Of the manner of formation of ichthulin in the maturation of the ovum we know absolutely nothing. Paton & Newbigin concluded from a very few analyses that the phosphorus was brought to the ovaries from the muscle of the salmon as inorganic phosphorus, but, in view of what is now known about the organic phosphorus compounds of blood, this appears rather unlikely. 346 THE UNFERTILISED EGG AS A [pt. in I '14. Fats, Lipoids and Sterols Studies on the fatty substances of the undeveloped eggs of different animals have resulted in much interesting information. There has been, of course, a great body of histological work, and the yolks of all kinds of eggs have been repeatedly subjected to microscopic examination (for example, Kaneko's study on the silkworm); but, in spite of many attempts, I have not succeeded in finding more than a few hints in this literature which are of value to the chemical worker. This subject has been dealt with in a general way by Ransom and by Dubuisson, to whose papers those interested in the histological aspects of yolk must be referred. Of the way in which the fat and the protein are intermingled in the yolk we know practically nothing, and it would be most desirable to investigate the yolk with the methods which modern colloidal chemistry has developed. But that the association between fat and protein indicated by the histological evidence is not very close is shown by the interesting centrifugation experiments of McClendon on the amphibian egg. If the egg of the frog is centrifuged for five minutes under the right conditions, it separates into three perfectly distinct layers, the upper one being oily and yellow, the middle one translucent, colourless and protoplasmic, and the lowest one black, containing practically all the yolk. By using a considerable number of eggs, McClendon was enabled to obtain suflticient material for the chemical analysis of each layer. The figures he obtained are shown in Table 43. It is evident from a slight inspection of his results that the upper layer is composed mainly of neutral fats and a little lecithin, and the middle layer of water, salts and protein, with no fats or lipoids. The lowest and much the largest layer is made up of the vitellin (ranovin or batrachiolin) together with the major part of the lecithin. It is interesting that the association between the phosphoprotein and the lipoid was the only one that centrifuging could not break, for, as we have already seen, the observation of a loose lecitho-vitellin combination in the hen's egg is very old. McClendon found that mitotic figures were all present in the middle layer, and that this centrifuging produced a variety of monstrous embryos. He was led to regard the protoplasm of the egg as constant in composition throughout, but "anisotropic as regards its axes, in other words crystalline in structure". SECT. l] PHYSICO-CHEMICAL SYSTEM 347 McClendon extended his observations to the egg of the sea-urchin, Arbacia punctulata. Separated by centrifugal force, this egg divided itself into four layers, as Lyon had already described, {a) a layer of yolk bodies and red pigment granules extending from the centrifugal end about half-way to the equator, {b) a layer of similar yolk bodies but without the pigment granules, {c) a translucent fluid layer extending almost to the centripetal pole and containing the nucleus, and finally {d) a very opaque layer or cap of minute volume, sitting on the centripetal pole. When the crushed eggs were centrifuged, the material separated into two layers, {a) and {b) being indistinguishable, centrifugal and containing the egg-membranes, and [c) centripetal, {d) not being perceptible. McClendon analysed the layers in the same manner as those of the frog's egg — the figures are given in Table 43. Table 43. McClendon' s figures (1909). In the /o in the layers % , of I" 1 "a lay i ers dry weight phosphorus A ll la Is / t! 2 c t< —1 1) u

  • -> tJ

a & 2 1 Layers of centri o-o6 6 Upper centripetal 50 50 8o-o 4-0 8 8 Trace 0-34 1-4 0-41 — fuged egg of (fatty or oily) Rana pipiens 016 16 Middle (protoplasmic) 82 18 7-5 II-5 60 21 Trace 0-05 i-o 0-37 ~ ~ 0-78 78 Lower centrifugal (yolky) 48 52 24-0 60 10 60 Trace 270 1-2 1-33 ~ ~ Layers of centri — 32-5 Centripetal (proto 88 12 49-0 — 20 3I-0 2 36 16-66 3-24 13-45 1-24 fuged egg of ^r plasmic) bacia punctulata — 67-5 Centrifugal (yolky) 79 21 38-2 — 10 51-8 2 74 12-84 1-6 10-6 2-02 A short consideration of them shows that centrifugal force is not nearly so successful in separating the egg of the sea-urchin into chemically unlike layers as it is in the case of the frog. This fits in perhaps with the long-established fact that centrifugal force interferes far less with normal development in the sea-urchin's egg than it does in the frog's egg (Morgan and Lyon). It was very noticeable that, whereas the frog's egg separated out into layers of markedly different water-content, this did not take place in the sea-urchin's egg. In the case of the centrifuged frog's egg, again, there were big differences between the phosphorus contents of the different layers, but in that of the sea-urchin's egg this only applied to the residues which were mainly protein. McClendon surmised that the inclusion 348 THE UNFERTILISED EGG AS A [pt. iii of the membrane proteins in the centrifugal layer caused this effect. It is of course a fact of the first importance that normal development can follow centrifugation and this will receive attention later (see Section 3 and the Epilegomena) . I shall only mention here as one of the best instances of this phenomenon, the work of Schaxel on the axolotl egg. Here centrifugation caused atypical discoidal cleavage which nevertheless resulted in a normally proportioned embryo. Thus normal conclusions can follow abnormal distribution of the so-called "organ-forming substances". For further details of these experiments, see Morgan and Bertalanffy. The early work of Gobley on the fat of the hen's and the carp's egg has already been described. He isolated glycerophosphoric acid from the latter, and pursued further his investigation of lecithin, concerning which it is of interest to note that Sacc contested his claim to have found organic alcohol-soluble phosphorus. Sacc believed that the fats contained dissolved in them a quantity of inorganic phosphorus. Gobley, however, was easily able to disprove this view and to show the identity of carp's egg lecithin with brain lecithin. Data which have accumulated since Gobley's time on the fatty substances of the eggs of the lower animals are collected in Table 44, and may be compared with those in Table 22. One of the most striking differences between the hen's egg and other eggs is the relatively low iodine value of the fatty acids of the former, both free and combined in lipoids. The neutral fat of the hen's egg has an iodine value varying roughly between 60 and 90, but for fish eggs the figures vary from 90 to 150, and the same rule holds generally of the lipoid fatty acids, for they average 60 in hen and 100 in fish eggs. The saponification numbers, on the other hand, are much the same throughout the two tables (from 170 to 200). The conclusion might therefore be drawn that egg fats differ rather more as to the number of unsaturated linkages in their acids, than as to the length of their chains. Nevertheless, there are remarkable exceptions to these generalities, the fatty acids of the echinoderm eggs, for example, having enormous saponification and high Dyer numbers, and therefore presumably only very short chains of carbon atoms. Arbacia is more remarkable in this than Asterias. Yet, though they are exceptional in that respect, they have iodine numbers very like those of fish-egg fats. Another point of interest is that the cholesterol/fatty SECT. I] PHYSICO-CHEMICAL SYSTEM 349 acid ratio, as shown by expressing the cholesterol in percentage of the fat present, is rather constant, never going below 4 and never rising above 12. This may have some connection with the physical Table 44. Data for fat fraction of eggs. 1 "o-tio a <S 1 3 ■3 a ^1 l^^l J3 0? £ C a ?f H ^ 5 ^ <4-r ."2 ."S "S - c 3 ■$t Qj cj.S .2 ■3 c 0? '0 '0 •s| &3 .-a (u u — 2 J5 4^ 11 'S 2 3 u p Investigator Species ^ ^ m c Q^Z c u^ Ji fc fe cfiJ: D^ and date Amphibia Frog {Rana tempor 123-0 — — 5-97 — — — — — Faure-Fremiet & Dragoiu aria) (1923) Fishes Sturgeon ... 107-6 191-4 — 4-35 12-92 — — — — Konig & Grossfeld (191 3) Trout 128-3 181-8 — 6-52 41-10 o-ig 0-19 — — „ ,, Cod 148-4 1 76- 1 — 12-05 35-19 0-21 0-43 — — ,, jj Herring ... 123-1 230-6 — 6-94 43-61 — — — ,, J, Carp 78-9 186-9 — 10-98 59-19 — — — — 55 55 Pike — — — — — 0-27 0-22 — — 55 3J Trout {Salmo fario) 108-6 219-8 — 6-23 37-50 — — — 1-7 Faure-Fremiet & Gar (132-9 189-8 Fatty acids of the ph osphatide fraction' rault (1922) Carp {Cyprinus carpio) — 140-0 ■ — 5-99 60-00 — — — 3-36 55 55 (64-4 — Fatty acids of the phosphatide fraction' Dogfish (55-88 — Fatty acids of the phosphatide fraction^ Ponce (1924) Shark (Lepidorhinus — — — — — — — 17-3 — Tsujimoto (1920) kinbei) ECHINODERMS Sea-urchin {Arbacia 147-0 606-0 4-001 — — — — — — Page (1927) punctulata) Sea-urchin [Echinus 145-0 195-0 — — 29-40 — — — — Moore, Whitley & Adams esculentus) (78-8 225-0 Fatty acids of the phosphatide fraction] (1913) Sea-urchin {Arbacia — — — — 50-00 — — — — Matthews (191 3) punctulata) Sea-urchin [Paracen 150-0 — — — — — — — — Ephrussi & Rapkine trotus lividus) (1928) Starfish (Asteriasgla 112-5 318-8 3-778 — — — — — — Page (1927) cialis) POLYCHAETE Polychaete worm — — — 8-85 — — — — IO-6 Faure-Fremiet (1921) {Sabellaria alveolata) Nematode Roundworm {Ascaris — — — 5-0 — — — — 80-0 Faure-Fremiet (191 3) megalocephala) State of the egg-cell, and will be referred to again (see Section 12-5). The lipoids, expressed as lecithin in per cent, of the fat present, show greater variations, but it is not possible to say at present what the significance of these may be. 350 THE UNFERTILISED EGG AS A [pt. iii The mention of squalene in Table 44 indicates the existence of an egg-constituent, our knowledge of which is of very recent origin. In 1 906 Tsujimoto isolated from the liver oils of elasmobranch fishes a saturated hydrocarbon of approximate formula C30H20, and in 191 6 published a further study of it. Its properties and constants are given in Table 22. In 1920 he reported that he had been able to isolate it from the egg-yolks of two elasmobranchs, Chlamydoselachus anguineus and Lepidorhinus kinbei, where it made up no less than 13 per cent, of the egg (wet weight) and, in another case, 1 7 per cent, at least of the total fat fraction. There the matter rested until 1926, when Heilbron, Kamm & Owens, taking up the question of its presence in eggs once more, isolated it from the undeveloped yolks of Etmopterus spinax, Lepidorhinus squamosus and Scymnorhinus lichia. In the fully developed eggs of the first-named of these three, practically none was present, indicating that it must either have been combusted or absorbed during development. Further researches on the embryological significance of this compound are greatly required. It is possible that some hydrocarbon of this sort may explain certain obscure points in the chemistry of the egg, for instance, the oil extracted by Dubois from the locust's egg {Acridium peregrinum). It contained 1-92 per cent, phosphorus, and was present to the extent of 4*5 per cent, of the wet weight of the egg, no small proportion. Kedzie studied a similar oil which he obtained from the egg of the American locust. A question which is perhaps related to the general problem of the egg-oils is that of the oil-globules of the yolks of some of the teleostean fishes. In 1885 Agassiz & Whitman divided all pelagic eggs into those which had the oil-globule and those which had not. But it was soon found that this method of classification was valueless, for the appearance of the globule is rather erratic; thus, although Lota vulgaris (van 'S>2ivs\he\ie) , Brosmius (anon.) and Motella mustela (Brook) were all found to have it, the common pike's egg does not have it (Truman). Ryder first suggested that the oil-globule might have a relation to buoyancy, but Prince, reviewing the whole subject a little later, pointed out that this could hardly be so, for the salmonoid fishes all have them, and yet their eggs never float. Moreover, out of 22 teleost eggs with no globule, 17 are pelagic, while out of 24 teleost eggs which have globules, only 15 are pelagic. Ryder replied to this by partially withdrawing his theory, and Mcintosh SECT, i] PHYSICO-CHEMICAL SYSTEM 351 simultaneously showed that the eggs of the catfish, which are undoubtedly bottom ova, have large oil-globules. Another theory was put forward by van Bambeke, who believed that the oil-globule was a special form of yolk, and of a purely nutritional significance. Prince criticised this view on the ground that the oil persists in the yolk after the liberation of the embryo from the egg-membrane, and travels beneath it as it swims about. This would not, however, negative the possibility that the oil was used for larval rather than embryonic nourishment. Van Bambeke' s claim that a protoplasmic thread passes from the oil-globule to the germinal disc was almost completely disproved by van Beneden. His and Miescher, examining the oil histochemically, found that it only stained very slowly with osmic acid, and therefore differed profoundly from the yolk, and, although it was soluble in ether, it contained no more than a trace of phosphorus. It is remarkable that the oil has never been subjected to a proper chemical examination, especially in view of the extensive zoological literature on it. What we know of its properties faintly hints, perhaps, that it may be a hydrocarbon like squalene, and the whole question, indeed, holds out great possibilities for physiological as well as chemical work. The oil must readily dissolve lipochromes, for the pink pigment of the salmonoids is found in it. Prince's own theory was that the globule was a constituent of ancestral significance, a vestige from the time when, as Balfour showed, the teleostean yolk was very much larger than it is now. The nutrition view is probably the best. The lipoids and sterols of the eggs of the lower animals are very little known, and their further study is much to be desired. Page in 1923 described a sterol — asteriasterol — which he isolated from the eggs of Asterias forbesii and which turned out to be closely related to, though not identical with, ordinary- cholesterol; the eggs of Arenicola cristata, on the contrary, yielded a sterol absolutely identical with the well-known substance as it occurs in mammals. Ten years previously, in a less accurate study, Matthews had failed to find any cholesterol at all in the eggs of Asterias forbesii, though he had been able to isolate some from those of Arbacia punctata. From the former he got a jecorin-like substance, containing 10 per cent, of glucosamine, which was probably a mixture of kephalin, cerebrosides, "protagon" and various carbohydrates. Page's later study of the fats and lipoids of the echinoderm egg led to the conclusion that (qualitatively) there was more kephalin in the eggs of Arbacia than 352 THE UNFERTILISED EGG AS A [pt. iii in those of Asterias, and more lecithin in the eggs of Asterias than in those of Arbacia. Asterias contains large amounts of soaps, and its oil is present in much greater abundance than the oil of Arbacia; moreover, it contains more sulphur compounds (sulphatides?) decomposable with potash than does the Arbacia egg. Page, Chambers & Clowes made a study of the effects of various cytolytic agents on the eggs of Asterias separated by microdissection into their cortical and endoplasmic components. They used for this purpose hypotonic sea water, digitonin and saponin, and found that digitonin caused slow cytolysis of the cortical and rapid cytolysis of the interior protoplasm when the two were isolated, whereas hypotonic sea water caused slow cytolysis of the interior and rapid cytolysis of the cortical protoplasm. If these results do not actually demonstrate that the greater part of the asteriasterol is localised in the outer and fertilisable parts of the egg, they at any rate suggest a new method of investigation which may help to solve many similar questions in the future. Runnstrom has studied the lipoids of the echinoderm Qgg in relation to its coloured interference fringes and its membrane properties. Among the sterols existing in eggs must be mentioned a substance which has long been known to occur in the ova of Ascaris, and which has been called "ascarylic acid". Faure-Fremiet identifies it with the droplets or crystals described in the egg oiAscaris by van Beneden. It was isolated simultaneously by Faure-Fremiet from the eggs and by Flury from the whole body of the nematode ; the former worker found that it accounted for 22 per cent, of the dry material. Ascarylic alcohol, ascarylic acid, or, as it would probably be best to call it, ascaristerol, seems to exist in the egg-protoplasm in combination with palmitic, oleic, and perhaps stearic acid in ester form. FaureFremiet & Leroux studied its properties, and proposed the provisional formula of C32Hg404 . Its saponification number was 199, and its m.p. 82°, it did not give the cholesterol colour-reactions, and its molecular weight was close to 511. Flury considered it to be related to oenocarpol. Acaristerol seems to be strictly confined to the eggs, for even the parietal cells of the ovary and uterus do not contain it, as Faure-Fremiet showed by means of histochemical tests. Nor is it present in the testes and spermatozoa. It may at present be classed with the sterols, like asteriasterol. Ascaris eggs also contain o-i6 per cent, dry weight of ordinary cholesterol. SECT. l] PHYSICO-CHEMICAL SYSTEM 353 Table 45. Distribution of phosphorus. In % of the total P f _3 u* Sh u i "o gfln 15 p-^ ^Oh Oh \-i ^ U ^ JJ . w A cx 1^ M)-? £P3 3 3 fl 2 G 3 Investigator Species HDh Hi OI -Si 7 -S fl and date Bird Hen 61-4 9-5 9-5 None 1-6 27-5 29-1 Plimmer & Scott (1909) Amphibian Frog (ovarian) 26-2 4-3 4-3 None 7-6 61-9 69-5 Plimmer & Kaya (1909) Fishes Sturgeon 28-8 i6'9 — 9-9 None 54-3 54*3 Plimmer & Scott (1908) Herring — — — — 63-0 63-0 ■>■) J3 Grey mullet — — — — — 48-0 48-0 5» J> Trout 26-0 — — — — 34-6 — Faure-Fremiet & Garrault (1922) Herring 33-2 — — — — 66-8 — Yoshimura (191 3) Haddock — — 21-22 — — — Milroy (1898) Herring — — — — lO-O 90-0 — Tschernorutzki (191 2) Salmon 37-8 i8-9 — Traces — 43-3 — Paton (1898) Herring — — + + — — Steudel & Takahashi (1923) ECHINODERMS Sea-urchin {Strongylo 43-0 33-1 — — — — 23-8 Robertson & Wasteneys centrotus purpuratus) (1913) Sea-urchin {Arbacia — — — — Much None — Masing (1910) punctulata) Sea-urchin [Arbacia 29-2 47-0 — i-i — — 23-5 McClendon (1909) punctulata) Sand-dollar {Dendras 6-09 46-95 17-55 29-4 32-0 I2-I 44-1 Needham & Needham ( 1 930) ter excentricus) Starfish {Patiria mini 32-3 51-8 32-0 20-0 15-40 Trace 15-4 J> 59 ata) Crustacea Sand-crab (Emerita 28-2 6 1 -40 42-10 19-3 10-82 Trace 10-82 >J J> analoga) Brine-shrimp [Artemia salina) Gephyrea 5-9 56-4 38-3 18-1 37-9 None 37-9 )> >5 Gephyrean worm ( Urechis caupo) Nematode 27-4 56-90 40-60 16-3 15-80 Trace 15-8 5> J) Roundworm {Ascaris) 25-6 — — 20-0 54-4 — 54-4 Faure-Fremiet (191 3) Miescher (1872) reported that in the salmon nearly all the phosphorus is in organic form.

  • This fraction will include pyrophosphate P.

t This fraction will include guanidine phosphoric acids (arginine or creatine phosphate P). + "Present in some quantity." The lipoids of the mammalian egg-cell have recently been the subject of some work which is interesting, though, Uke all histochemical studies, very difficult to appraise. Following on Russo's claim to have found two different sorts of eggs in rabbits, varying N E I 23 354 THE UNFERTILISED EGG AS A [pt. m in their reactions to staining methods, Fels in 1926 confirmed this difference for the human egg-cell, some specimens of which showed a strong lipoid-reaction (Ciaccio and Smith-Dietrich methods) in the nucleolus while others did not. Fels' illustration is certainly striking. Leupold had already put forward the view that eggs whose nucleoli were rich in lipoids produced females, and the remainder males, but all the evidence, however, is against sex-dimorphism in the mammalian egg (see Parkes' review). These observations, together with those of Pollak on the presence of Reinke's crystals in the egg of Macacus rhesus, and similar work by Limon and" von Ebner (on Cerrus capreolus), are all that we have on the chemical constitution of the mammalian egg-cell. Closely connected with the lipoids of the egg is the distribution of phosphorus compounds in it and Table 45 gives what is known upon this subject. It is interesting to see how the phosphoprotein phosphorus varies, in some eggs being very large in proportion to the total phosphorus, in others being almost insignificant. Masing was wrong in saying that the echinoderm egg has none at all, for Needham & Needham in 1929 observed quite a high percentage in the &gg of the sand-dollar. It is significant in view of what has already been said about the pre-eminence of birds in storing fat in their eggs, that the hen's egg has 20 per cent, more phosphorus in lipoidal form than any other egg investigated. The fishes rank in this respect with the echinoderms and annelids, little diflference being noticeable between alecithic and lecithic eggs. Perhaps this famous distinction involves neutral fat rather than lipoids. It is to be noted from Table 45 that the inorganic phosphorus content of eggs is very variable; in many cases almost none is present, but the haddock's ^gg seems to have no less than 20 per cent, of the total phosphorus in this form. About the same proportion is present in the nematode egg, ii Ascaris can be taken as representative. FaureFremiet was able to identify the calcium phosphate in the egg-cytoplasm with the "hyaline balls" described by van Beneden, using various histochemical reactions (McCallum, Prenant, etc.). Pure calcium phosphate, according to Faure-Fremiet, accounts for 0-4 to 0-6 per cent, of the dry weight of Ascaris eggs, an inconsiderable amount in view of the share it takes in the appearance of the cytoplasm as a whole. SECT, i] PHYSICO-CHEMICAL SYSTEM 355 1-15. Carbohydrates The carbohydrates of the eggs of the lower animals have been less investigated than anything else — a summary of our quantitative knowledge concerning them is shown in Table 46. The presence of glycogen in insect and mollusc eggs was noted by Bernard and in those of arachnids by Balbiani. For the reasons mentioned above, it is difficult to know how trustworthy the figures for carbohydrates are, so bad have the methods been in the past. Faure-Fremiet & du Streel's figure for the glycogen of the frog's tgg must surely be too high, for most of the other workers are agreed on a value of about 2 gm. per cent, wet weight. In the case of animals other than amphibia, the figures are too scattered to permit of any generaUsation: thus, though glycogen was not found in herring's eggs by Steudel & Osato, Gori did note its presence in Torpedo eggs, and the eggs of the reptiles Vipera aspis and Elaphis quadrilineatus, in addition to free carbohydrate. Steudel & Osato pointed out that many histologists such as Goldmann had published results concerning the fish egg which might lead one to suppose that very large amounts of glycogen were there. That this was not found by chemical methods ought to induce, they felt, a more cautious attitude towards histochemical work than was customary; indeed, much of what is called glycogen histochemically can certainly not be glycogen. Greene's carbohydrate figures for the eggs of the king-salmon Oncorhynchus tschawytscha, from Cahfornian rixers, were of special interest, for, throughout the maturation period, the carbohydrate content of the egg remained the same. The duration of the fast did not affect it at all. Quahtative investigations of carbohydrate in eggs have been made by Anderlini on the silkworm egg and by Konopacki, who observed the presence of glycogen microchemically in the perivitelline fluid of the frog's tgg. As has already been mentioned, a carbohydrate group is undoubtedly contained in the mucoprotein of the amphibian egg-jelly, and von Furth's analysis of the egg-cases of the squid Loligo vulgaris showed that their protein also contained a carbohydrate, but whether these play any part in the sugar supply for the developing embryo remains an obscure point. Haensel found an amount of glucose in the frog's tgg which is shown in Table 46, but he also tried the effect of keeping the eggs in solutions of various mono- and di-saccharides, 23-2 356 THE UNFERTILISED EGG AS A [PT. Ill Table 46. Mgm. % wet weight Mgm, % dry weight . ' , , ' ^ 6 ^ 4 ^ % y ^ h y c S-o iJ ^ l-S.^ >■ Investigator Species H-c'fa O h-o^ta O ^nd date Reptiles Turtle ( Thalassochelys corticata) White ... ... ... — Trace — — — — Tomita (1929) Yolk ... ... ... — 100 — — — — ,, Tortoise {Testudo graeca) ... — 140 — — — — Diamare (1910) Amphibia Frog {Rana esc. and fuse.) ... — — 2,520 — — — Kato (1909) ,, ,, ... — — 1,100 — — — Athanasiu (1899) Frog {Rana temp.) ... ... — — — — — 7810 Faure-Fremiet & Dragoiu (1923) Frog {Rana esc. and fuse.) ... — — 2,500 — — — Bleibtreu (1910) Frog {Rana temp.) ... ... — — — — 193 — Gori (1920) ,, ,, ... ... — — 10,140 — — — P'aure-Fremiet & Vivier du Streel (1921) „ ,, ... ... 604 — — 1906 — — Needham (1927) ,, ,, ... ... — — 2,528 — — — Haensel (1908) ,, ,, ... ... — — 1,650 — — — Goldfederova (1925) Fishes Herring — 500 — — — — Steudel & Osato (1923) King-salmon {Oncorhynchus — 96 — — — — Greene (1921) tschawytscha) Trout {Salmo fario) ... ... — — 340 — — — Faure-Fremiet & Garrault (1922) ECHINODERMS Starfish {Asterias glacialis) ... — — 20 — — — Dalcq (1923) Sea-urchin {Echinus esculentus) — — i ,360 — — 8980 Moore, Whitley & Adams (1913) Sea-urchin {Strongyloeentrotus 1360 — — 543° — — Ephrussi & Rapkine lividus) (1928) Insects Silkworm {Bombyx mori) ... — — 1,110 — — — Pigorini (1922) ,, ,, ... — — 1,980 — — — Tichomirov (1882) ), „ ... — — — — — 3080 Vaney & Conte (1911) Bee {Apis mellijica) ... ... — — 2,500 — — — Straus (191 o) Cephalopod Octopus {Sepia officinalis) ... — — None 3620 1000 None Henze (1908) ,, As glucose ... ... 2700 — — — — — ,, ,, As pentose ... ... 2380 — — — — — ,, Polychaete Polychaete worm (^aie/Zana — — 1,270 — — — Faure-Fremiet (192 1) alveolata) Nematode Roundworm {Ascaris megalo- — — — — — 2105 Faure-Fremiet (1913) cephala) 356* Table 47. Ash content of eggs. WllOLB EOU H m I'ARTB Pike (£j« lueiw ) Sturgeon [Acitie Sea-urdiilt [Arb ■jntlurw) Starfuli LUUtia. Sea water (Wo ih Hole) Sea waier {Clia tnga cxpcdilion) Sea- urchin (Sir neylQcentrolus lividtu) Dogfish {^p'""' Spider-crab (A/ , cmicula) ... no vtTTUCOsa) ... Octopiu tStpia Ifianalu) Hen (6'<i//uj Join '»""")■■• MotluK ( Volula Dogfuh {Squaiin otonllUas) Frog {Rana Umporaria) l-rout (&ilmo fontimlii) Salmon {Saimo ndar) ... Wrauc (Labrax lupus) ... Torpedo [Torptdo dctllala) Dogfiah [Snilium canicula) Spidcr-aab {Maia vnrucosi Octopu) (Stpta ojficinalij) a pmtuloia) Gephyrcon worm {Sipuncutus nitdui) . Lugworm [.irtnitola claparti"" Herring {Glupea hariagus) Waicr-sol. Salmon .. Plaice Dogfuh \Squatut acanthiai) (cgg-jclly) Sawfish {Pritlii aniiquorum) Mtlhi:quivalcnLa prcx SO, PO, a — 55 Na Mg Ca V»7 01Q7 413 < .V»4 ■1 325 '17 9-8o 300 10 — I -05 4 107 — 0-99 46-6 - 5 i 1-04 z 1-04 16-29 ^fj^ ■6'65 ■iS-29 3-9B g-6i 11-51 1707 ^bs — 10-79 ^3^ 4<'85

&.Grossreld(i9i3) 2-08

'■«<> Page (1927) a-27 — ■ O4-0 7S-3i 409 ia-37 „ 0-004 5'-7 45-^2 Si-s^ 54*42 57-'i — 4-64 — 34-a 54-03 47 b'^'b'^ 5^'^ Diiunar, ualUnger, vol. 1 ■64 13-44 9"3i I3'5i Goblcy {1850) - — — — Wetzel (1907) — — 33-7 t' z 'U *;? ■54 — afj-g — 3'-9 — '"3 — 559 5 — — 7a — — 7-18 ,8-, 11) b-,!, 4-4 lo/i 10-31 13-93 3-3 11-33 'i-7«  6 — — Silkworm iBombyx n Turtle [Timloiiochew coTticata): Whole egg While Yolk >9'03 30-S8 Btalasccwicz (1926} 6i-oi 63ja RoBb & Correa (1927) McCailum (1926) Greig {1898) Milroy [1898) Perugitt (1879) KrukeabcTS (1888} Uf^UID OP TIIB ' Hen {Callus domaliats) Frog {Rana lembotana) Trout {Salmo/ontinatis) Torpedo {Torptdo ocillata) Spiaer-crab ^oifl wrrwfwfl) ... — Sea-urchin (raracentntus lividus) Octoput [Stpia qfficimlii) Schroder (1909} Karaahima (igag) i?-j Bialasccwiti (i< SECT, i] PHYSICO-CHEMICAL SYSTEM 357 to see if they would grow richer in glycogen. They all did; in fact, he was able to double their glycogen content by this simple means (glucose acted better than sucrose, sucrose than lactose, and lactose than glycerol, though even the latter substance gave an effect) . These curious observations have never been confirmed, and can hardly be said to carry conviction as they stand. Diamare obtained discordant results in his researches on the sugar of various eggs ; thus, he got a rather low value for the free glucose of the egg of Testudo graeca, but none at all, either free or combined, from the eggs of Scyllium catulus or Torpedo marmorata. No explanation can be given for this fact. In connection with carbohydrates, it should be remembered that viper venom, which is in all probability a glucoside, has been shown by Phisalix to be present in active form in the yolks of viper eggs. I -16. Ash We come now to the inorganic substances of eggs. Iron has been shown to be present by microchemical tests in many eggs, such as those of Limnaea, Tubifex, Rana esculenta (where it is massed at the light ventral pole) and Pisidium, by the work of Schneider. Dhere found traces of iron and copper in the eggs of Sepia. Warburg found 0-02 to 0-03 mgm. iron per 100 mg. nitrogen in the unfertilised sea-urchin €:gg', part of it seemed to be in ionic form and part not. According to Wilke-Dorfurt, there are 4-8 mgm. per kilo iodine in oyster egg-shells. Ash analyses of eggs have been made by several workers, whose results, it may be remarked, would have been more easily comparable if they had expressed them in the same way, instead of in nine or ten different ways, omitting in some cases the figures which would enable them to be calculated into a form comparable with each other. Table 47 summarises what is known about the distribution of inorganic substances in eggs. It has entailed a good deal of calculation, for only one of the previous investigators expressed his results in terms of millimols and milliequivalents, and unless this is done it is impossible to gain any idea as to the relative preponderance of cation and anion. The first thing which should be noted is the fact that, when the salts are expressed in per cent, of the total ash, potassium is always there in greater amount than sodium, and nearly always to a greater extent than any other metal. This seems to be quite characteristic of the ovum, though in other systems of SECT. I] PHYSICO-CHEMICAL SYSTEM 357 to see if they would grow richer in glycogen. They all did; in fact, he was able to double their glycogen content by this simple means (glucose acted better than sucrose, sucrose than lactose, and lactose than glycerol, though even the latter substance gave an effect) . These curious observations have never been confirmed, and can hardly be said to carry conviction as they stand. Diamare obtained discordant results in his researches on the sugar of various eggs ; thus, he got a rather low value for the free glucose of the egg of Testudo graeca, but none at all, either free or combined, from the eggs of Scyllium catulus or Torpedo marmorata. No explanation can be given for this fact. In connection with carbohydrates, it should be remembered that viper venom, which is in all probability a glucoside, has been shown by Phisalix to be present in active form in the yolks of viper eggs. i-i6. Ash We come now to the inorganic substances of eggs. Iron has been shown to be present by microchemical tests in many eggs, such as those of Limnaea, Tubifex, Rana esculenta (where it is massed at the Ught ventral pole) and Pisidium, by the work of Schneider. Dhere found traces of iron and copper in the eggs of Sepia. Warburg found 0-02 to 0-03 mgm. iron per 100 mg. nitrogen in the unfertiUsed sea-urchin egg ; part of it seemed to be in ionic form and part not. According to Wilke-Dorfurt, there are 4-8 mgm. per kilo iodine in oyster egg-shells. Ash analyses of eggs have been made by several workers, whose results, it may be remarked, would have been more easily comparable if they had expressed them in the same way, instead of in nine or ten different ways, omitting in some cases the figures which would enable them to be calculated into a form comparable with each other. Table 47 summarises what is known about the distribution of inorganic substances in eggs. It has entailed a good deal of calculation, for only one of the previous investigators expressed his results in terms of millimols and milliequivalents, and unless this is done it is impossible to gain any idea as to the relative preponderance of cation and anion. The first thing which should be noted is the fact that, when the salts are expressed in per cent, of the total ash, potassium is always there in greater amount than sodium, and nearly always to a greater extent than any other metal. This seems to be quite characteristic of the ovum, though in other systems of 358 THE UNFERTILISED EGG AS A [pt. iii the organism other relations are found; thus corpuscles and plasma of some mammalian bloods have converse potassium/sodium ratios, and, as a general rule, potassium preponderates in cells while sodium preponderates in media. Of the anions PO4 usually takes up much the greatest part, but SO4 may in certain cases equal it. In the columns on the right of the table the total anion and total cation are shown, in each case calculated as millimols and as milliequivalents, the former giving an idea of the total number of molecules present, the latter of the total number of valencies. Study of the anion/cation ratio expressed as milliequivalents per cent, wet weight provides an important key to the constitution of the egg, for it shows roughly to what extent anion or cation is held in combination with protein or lipoid, or other organic substances. We have already seen that in the case of the hen's egg, taking both yolk and white into account, the anion/cation ratio is more than unity (Bialascewicz's figures give 2-17), showing that a quantity of sulphur and phosphorus is in organic combination — a conclusion which fits in admirably with all that we know of the hen's egg from other sources. The same relationship is seen in the figures of Konig & Grossfeld for the three fish eggs they investigated, the pike, the cod and the sturgeon. On the other hand, the figures of Page for two echinoderm eggs give ratios much less than unity, demonstrating the organic combination of a good deal of the cation. It may be noticed that the analyses of Dittmar and Page for sea water give ratios in the very close neighbourhood of unity, as would be expected, and indicate at the same time that the ratio cannot be regarded as significant to less than o-og. From what has been said, therefore, it might be concluded that the yolk-laden eggs of the fishes, like that of the hen, have a ratio above unity, while the alecithic echinoderm eggs have ratios much below it. But there are exceptions to this generalisation. The ratio of unity for the carp egg which is given by Gobley's results may perhaps be neglected, owing to the date of the work (1850), and the similar value obtained by Roffo & Correa on a gastropod egg may also be regarded as suspicious because of the enormous amount of sodium chloride that appears in their analysis. But the careful work of Bialascewicz in 1926 does not altogether support the generalisation. His figures for the fish egg are in good agreement with those of Konig and Grossfeld, but his anion/cation ratios for the echinoderms SECT, i] PHYSICO-CHEMICAL SYSTEM 359 do not go below unity, though they approach it much more nearly than do the fishes. Further work is needed to clear up this contradiction. In one case, however, Bialascewicz got a ratio below unity, that of Arenicola claparedii, so that in a general sense his investigations are not opposed to those of Page and Konig & Grossfeld. McCallum's low ratio for the egg of the herring is difficult to explain, but Perugia's analysis of the egg-jelly of ovo viviparous selachians fits in well enough with the majority of the other evidence. Attention might also be drawn to Bialascewicz's high ratio (15) for the eggs of the octopus. Sepia, which would appear to be extraordinarily poor in metallic ions (cf. p. 317, Section 13 and the Epilegomena). Some further light is perhaps thrown on the inorganic composition of eggs by Wetzel's figures for insoluble and soluble ash. He submitted the eggs of various animals to examination, with the following results : % dry weight Species Total ash Insol. ash Sol. ash Sea-urchin {Strongylocentrotus lividus) ... 9-7 2-4 7-2 S^iideT-cvah {Maia squinado) ... ... 4-12 0-27 3-8 Octopus {Sepia officinalis) ... ... ... 2-2 0'59 i'6 Tio^sh. [Scyllium canicula) ... ... ... 5-5 1-15 4-3 In all cases he found more soluble than insoluble salts, i.e., more chlorides than sulphates and phosphates. Table 48. Bialascewicz' s figures. Concen Vol. of tration Vol. of liquid CI in I c.c. Total CI inter c.c. of after Deeree ultra in ultra micellar yolk dilution of filtrate filtrate fluid per Species taken (c.c.) dilution (mg.) (mg-) I c.c. yolk Hen (yolk) 4-8 10 2-o8 I -08 1-080 — >3 4-8 20 4-17 0-472 0-944 0-541 >> 4-8 30 6-25 0-306 0-918 0-569 4-8 40 8-33 0-222 0-888 o-537^ »J 4-8 50 10-40 0-195 0-975 (0-754) It is very interesting, as Bialascewicz points out, that the mineral composition of terrestrial and aquatic animals should be so alike. The preponderance of potassium which is seen in the hen's tg^ does not change as one passes to organisms laying their eggs in an environment containing far more sodium than potassium. Thus, although 36o THE UNFERTILISED EGG AS A [pt. iii the normal sea water has twenty times as much sodium as potassium, fish eggs often have quite twenty times as much potassium as sodium. There would not appear to be in this connection any difference between homoio-osmotic and poikilo-osmotic aquatic animals. It is also obvious from Table 47 that aquatic eggs often have very much less salt in them than the ambient medium, and this would be a special case of the phenomenon found in all marine animals, and termed by Fredericq "Mineral hypotonicity". Bialascewicz arranged the animals he studied in a list of ascending concentration of metalHc ions as follows : Metal gramions Species per litre Octopus {Sepia officinalis) ... o-oi6 Gephyrean worm {Sipunculus nudus) 0-064 Spider-crab {Maia verrucosa) 0-079 Wrasse {Labrax lupus) 0-091 Herring {Clupea harengus) (McCallum) o-ioo Dogfish {Scyllium canicula) ... 0-107 Sea-urchin {Arbacia pustulosa) 0-159 Sea-urchin {Paracentrotus lividus) ... o-i8o which would also be an ascending table of taxonomic groups, were it not for the high metal content of the echinoderm eggs, which exceed even the fishes. There are other points concerning the relative amounts of salts in the eggs which require mention. McCallum, who had for a long time previously been studying the proportion of salts in the ash of animals and parts of animals with reference to the composition of sea water both now and in earlier geological epochs, made an analysis of herring's eggs in 1926. He had previously differentiated between palaeo-chemical salt ratios in bloods, namely, ratios resembling that which pre-Cambrian sea water can be calculated to have possessed, and neo-chemical salt ratios, namely, ratios resembling the sea water of the present day. Thus Limulus polyphemus and Aurelia flavidula, the king-crab and the medusa, which have always been marine animals, now approach the modern sea in the composition of their vascular fluids, but the lobster Homarus americanus, the selachian Acanthias vulgaris, the frog, dog, and man, for instance, all have ratios resembling the composition of the sea water at the appearance of the protovertebrate form. He had also identified the kidney as the organ responsible for maintaining the palaeo-ratios in the salts of the blood. In order to explore the possibility of identifying a palaeo-ratio SECT, i] PHYSICO-CHEMICAL SYSTEM 361 in the contents of the cell itself, he had recourse to eggs, and for those of the herring obtained the following distribution : Ratios on the basis of Na 100 Na K Ca Mg CI 100 216-7 ii'4 18-7 356-8 This stood in marked contrast not only with the vertebrate bloodplasma but also with the Archaean sea water calculated for the time at which life first began to appear in it, thus : Vertebrate blood-plasma (dog) 6-6 2.8 0-7 139-5 Archaean sea water 100 100-250 10 0-05 But after extraction of the dried eggs with water in a Soxhlet apparatus, the determination of the ratio of salts in the soluble part gave results more like the ratio for the Archaean sea water: 100 219-9 5'6 1-6 359-2 McCallum therefore concluded that the soluble part of the ash of the herring's egg exhibits a palaeo-chemical ratio. The bond shown here between the metals and the organic substances is useful in reminding us that even in fish eggs, where the anion/cation ratio is well above unity, some of the metal as well as the acid radicles may be united in organic combination. The relation between the salts in the intermicellar fluid of yolk and those in the dispersed phase itself has been studied by Bialascewicz and by Vladimirov. Bialascewicz worked firstly with the yolks of Torpedo eggs, but also with those of the hen and the trout. He prepared series of mixtures of the yolk with diluents in different concentrations, such as isotonic solutions of lithium sulphate and lithium nitrate, or in some cases distilled water, and then, submitting the mixtures to ultra-filtration, he estimated the ash and its composition in the filtrate and the residue. He first found that the percentage of chlorine bound to the dispersed phase in the ooplasm was practically independent of the degree of dilution, and from this fact he was able to calculate the volume of the intermicellar fluid of the yolk (see Table 48). For the hen's egg this was 0-549, per 362 THE UNFERTILISED EGG AS A [pt. m wet substance, and for the egg of Torpedo ocellata a similar calculation, based on cryoscopic experiments, gave a value of 0-482. On the basis of these figures, he proceeded to study the partition coefficient of each individual ion as between dispersed phase and intermicellar liquid. In Table 49 these partition coefficients are given; they represent the ratio amount of ion in the continuous phase or intermicellar liquid j amount of ion in the dispersed phase. It will be noted from Fig. 17 that as dilution of the original yolk goes on the ratios in some cases change, but in others remain constant. Thus the chlorine of the trout and the hen egg yolk remains constant at 0-5 in the latter and 1-02 in the former case, showing that Table 49. Bialascewicz's figures. a l-H 5 31 i §1 si 34 Is ►3t ■f-s SI 5 .2 ^ ^1 11 •2 ~ K 0-722 I -000 0-890 0-768 I -000 0-870 0-967 I -000 0-945 o-8oo Na 0-942 0-567 0-509 0-331 0051 — 0-080 0728 I- 000 I -000 Ca 0-093 0-391 0-274 0-169 0-321 0-760 0-474 0-696 0-505 I- 000 Mg 0-295 0-460 0-321 0-380 0-157 0-410 0-707 0-631 0-272 0-491 P 0-025 0-244 o-ioo 0-275 — — 0-040 0-318 o-i86 — CI 0-555 0-905 I -000 0-567 0-943 I -000 0-970 I- 000 I -000 0-766 it is very stably combined in the dispersed phase, though in different proportions according to the animal. Thus there is considerably more chlorine in the dispersed than in the continuous phase of the yolk of the avian egg, while in the fish egg there is a very slight excess of chlorine in the continuous phase. In all other instances, however, both as regards the hen and the trout, the excess of ion is in favour of the dispersed phase, the colloidal aggregates of which may therefore be looked upon as reservoirs of ash. Nevertheless, there is a good deal of the sodium combined in the continuous phase, and not a little of the potassium, though here the trout differs from the hen, for the potassium ratio is about 0-9 in the former case and only 0-7 in the latter. All the other ions have lower ratios than these; magnesium, calcium and phosphorus, for instance, are all present to a much greater extent in the dispersed than in the continuous phase. These experiments show also exactly how firmly the ions in the dispersed phase are bound there, and with what ease they may be washed out into the ultra-filtrate. It is apparent from SECT. l] PHYSICO-CHEMICAL SYSTEM 363 S 2 °'^^ £ oT 0-8 it'-r CO) 0-63 « E 0-2 0-lb-. 3sl Bialascewic ^ -sSf. Fig. 17 that the phosphorus, chlorine, and probably sodium in the dispersed phase, are intimately united there, for, however great the dilution of it, they do not increase in the ultra-filtrate. Magnesium and calcium, on the other hand, show a comparative readiness to pass out of the dispersed phase as the dilution is increased. The behaviour of the potassium is the most pecuHar, for, as dilution goes on, the calculated concentration of this ion actually decreases, but as the decrease is slight it is probably due to experimental error, and it was treated as such by Bialascewicz himself. Thus, of the ions bound to the dispersed i | o-s phase, the cations sodium and potassium, and the anions of §? 0-3^, chlorine and (presumably) phosphate, are firmly attached, while the cations calcium and magnesium are not, and can easily be washed out. The high proportion of phosphorus in organic combination should be remembered here. Bialascewicz also pointed out that the partition coefficient or ratio followed with dilution a practically rectilinear course, so that some idea of the ratio in the natural undiluted yolk might be obtained by extrapolation. These figures so obtained are shown in Fig. 17, from which it may be deduced that the ions follow the order phosphorus, calcium, magnesium, chlorine, potassium, sodium, beginning with the one most of which is in the dispersed phase and ending with the one least of which is so distributed. Fig. 1 8 shows another aspect of the passage of ash from dispersed to continuous phase. In succeeding papers Bialascewicz extended these researches to the eggs of amphibia, some other fishes, Crustacea, molluscs, echinoderms and annelids. He reported that the intermicellar liquid varied much in its relative amount, accounting for from 20 to 63 per cent, of the whole ooplasm. From the data in Table 50, however, there does not seem to be a very close relation between the relative volume extrapolated 2 values for undiluted ooplasm 4 6 degree of dilution Fig. 17. 364 THE UNFERTILISED EGG AS A [PT. Ill 100 90 €1 -0% 70 M, Ca, 6 • — ■ «..;«  of continuous phase and the percentage dry weight of the system. Bia lascewicz's tables give the concentration in percentages of the principal ions in the intermiceliar liquid of different eggs, and these are conveniently summarised in Fig. 19, taken from his paper. From this it is obvious that all the eggs studied have about the same proportion of potassium, but that the other ions are rather variable. There is much more calcium, relatively, in the continuous phase of the yolk of the hen's egg than in that of any of the others except the crustacean Maia verrucosa. Similarly, there is more magnesium, relatively, in that of the frog than in any other egg. A very interesting comparison may be made between the distribution of ions in the continuous phase of the eggs and that in the serum of Table 50. Bialascewicz's figures. 0-1 0-2 0-3 Concentration of the bhree elements in the continuous phase (mgrnt per cc) Fig. 18. 0-4 Continuous phase cc. per cc. In% of vol. Egg ooplasm ooplasm % dry weight Scyllium canicula 0-83 17-0 — Salmo fontinalis 0-79 20-8 — Salmofario ... — 41-5 (Faur^-Fremiet & Garrault) Torpedo ocellata 0-41 59-0 — Acanthias vulgaris — — 47-3 (Zdarek) Arbacia pustulosa 0-82 17-8 — Paracentrotus lividus .. 0-79 20-7 22-6 (Wetzel) Rana temporaria o-6o 39-9 42-6 (Kolb, Terroine, etc.) Callus domes ticus 0-55 45-1 50-3 (Kojo) Sepia officinalis 0-50 50-0 47-3 (Wetzel) Maia verrucosa 0-37 63-2 43-6 (Wetzel) SECT, l] PHYSICO-CHEMICAL SYSTEM 365 the corresponding adult animals. Fig. 20, taken from Bialascewicz, shows that the potassium preponderates in the former and the sodium in the latter, while the other inorganic substances are more or less equally distributed. As there is no difference in electrolyte con 70 -S 604 to 50 40 30i 20 10


o CO I O o o ^ to I ^= Potassium ^ = Calciu = Sod'rum Fig. 19 ^ = Magnesium centration between the continuous ooplasm phases of fresh-water and marine animals, one must conclude that salts do not account for the properties possessed by the latter, and that crystalloidal organic compounds, such as taurine, urea and glycine, play an important part in keeping up a high osmotic pressure. Thus Bialascewicz found a concentration of 8-43 gm. per litre of urea in 366 THE UNFERTILISED EGG AS A [PT. Ill undeveloped Torpedo ocellata eggs, but none in those of Arbacia, Sepia, or Maia. Vladimirov occupied himself with the egg-white in the egg of the bird. In connection with other investigations which dealt with the Salmo fonblnalis PcGassium Sodium Magnesium Calcium Torpedo ocellaba Maia verrucosa Continuous phase of egg-yolk Serum of adult animal Fig. 20. water metabolism of the egg (see Section 6-4) he measured the electrical conductivity of the egg-white in the unfertilised ovum, using the Kohlrausch and Holborn apparatus, and obtaining a value of 7*6 X io~^. By the aid of a dialysis method he calculated the electrical conductivity of the intermicellar fluid of the egg-white, allowing for the disturbing effect of so large a concentration of pro SECT. I] PHYSICO-CHEMICAL SYSTEM 367 tein. The result came out to 10-4 x io~^. If this work were repeated for the yolk, interesting commentary on Bialascewicz's researches would be possible. It agrees with the earlier measurements of Bellini, who found the electrical resistance of the unincubated white to be Q. 1 8-8 ohms. Much further work on such properties of the yolks and egg-whites of a wide range of eggs is urgently needed, for they must obviously be of the greatest importance to the developing embryo. Such questions as the electrical conductivity of egg-cells and developing embryos are very relevant here, but must be left for consideration in Section 5. As a conclusion to this discussion of the chemical constitution of the egg, it may be admitted that great progress has been made in our knowledge with respect to it during the last fifty years. But to a discerning judgment, it remains none the less a matter for great surprise that in view of our comparative ignorance of the chemical architecture of the egg, we know as much as we do about the cominginto-being of the chemical architecture of the finished embryo. One further matter may be alluded to in this section. The composition of fossil eggs cannot be said to have much embryological interest, but it is hard to exclude a mention of them. The only analyses we have are those of ZoUer who worked on the fossil eggs of Chincha Island, off the Peruvian coast, where seagulls have been living and depositing guano from a very remote date. Zoller found that "time, which antiquates antiquities, and hath an art to make a dust of all things" had had that effect on these eggs and had reduced their water content to 14-4 per cent. There was no urea or uric acid present, although the protein had nearly all disappeared and had given rise to ammonium salts. There was no trace of fat or of carbohydrate, and the sulphur of the proteins had all turned into sulphate. Water % 14-4 Cholesterol 0-287 Phosphoric acid ... 0-045 Total nitrogen 945 Ammonia N 8-12 K 14-9 SOs 16-08 These figures make it only too clear that if palaeontology and biochemistry enter into closer relations than exist at present, it will not be by way of the chemical analysis of fossil eggs. More hopeful approaches will be found in Section 9-15. SECTION 2 ON INCREASE IN SIZE AND WEIGHT 2-1. Introduction We have so far been considering the unfertilised egg-cell and its reserves of nutrient material as a physico-chemical system, and we must now proceed to summarise critically what is known about the alteration the egg undergoes in passing into the state of the finished embryo. Subsequent sections will take up the chemical changes during this process in all their complexity, but first the apparently simple phenomena of change of weight must receive consideration. To this undertaking special difficulties are attached; for example, the act of birth or hatching itself, important though it is for the chemical embryologist as the term of his investigations, is yet purely arbitrarily and conventionally chosen as such, and, as far as the organism itself is concerned, may be relatively unimportant. The age at which birth takes place varies in different animals considerably, and may occur earlier or later in development, cutting across cycles of growth at almost any point. However, the study of growth in weight and alteration in shape, is an essential preliminary to the study of the chemistry of the embryo. I do not propose to spend any time in the discussion of definitions of growth. The actual data which we have concerning pre-natal growth will be found in Appendix i, where they have been placed in the hope that a collection of them will be of assistance to chemical embryologists. No previous assemblage of them has been made, and they are to be found scattered all through the literature. Biochemists have in the past been insufficiently careful to check their results on embryos against normal tables of weight, length, age, etc. The predecessors of this section are the chapters on growth in d'Arcy Thompson's Growth and Form and Faure-Fremiet's La Cinetique du Developpement. These authors gave a full criticism of the whole subject, but without special reference to the development of the embryo. Moreover, much has been done since they wrote, and their treatment differs in various ways from what follows here. PT. m, SECT. 2] ON INCREASE IN SIZE AND WEIGHT 369 It is obvious that the growth of an embryonic organism can be measured in many ways besides that of increasing weight. Its enlarging dimensions in various directions of space can be measured, or its volume, or the quantity of various constituent substances. More will later have to be said about the way in which these different quantities may be thought of as fitting in together and changing with age. But the simplest manner of representing growth will probably always remain the measurement of the increase in weight of the total mass, and it is this which is now to be considered. The relation of this factor to the age and the length of the foetus is a point of capital importance to the chemical embryologist in the knowledge of his material. It is true that the data are fragmentary enough, restricted as they are almost entirely to various mammals and the chick. 2'2. The Existing Data Cephalopods. Octopus. A remarkably complete set of data for the embryonic growth oi Sepia is given by Ranzi, and this is almost all we have as regards invertebrate development. Insects. Silkworm. Luciani & LoMonaco have studied the curve of growth through the successive moults in the larval condition, but, in spite of their work and of many other researches on the silkworm larva and ^gg, I cannot find any in which the increasing weight of the embryo itself has been measured prior to hatching. Fishes. Trout. Weighings of fish embryos have been exceedingly few in number, owing to the smallness of their size and the difficulty of separating them from the yolk. Kronfeld & Scheminzki, however, have made some estimations of the increase in weight of trout embryos, and their figures, together with those of Gray, are shown in Table i of Appendix i. 24 370 ON INCREASE IN SIZE [PT, m Amphibia. Frog. In the case of amphibia, where the cleavage in the egg is more or less inclusive of the yolk-laden portion, it is not possible to obtain data for the weight of the embryo itself, for, before hatching, although the protoplasm is constantly increasing at the expense of the yolk, the two elements cannot be separated, and therefore cannot be weighed in isolation. This appears in the figures of FaureFremiet & Dragoiu ; Schaper ; Davenport ; and Bialascewicz, and must always be taken into account when differences between species in water-content and other constants are under consideration, for much confusion may be caused by not distinguishing carefully between yolk plus embryo and embryo alone. Reptilia. Snake. Bohr's very few figures on Coluber natrix are all that are available. (Appendix i. Table 2.) Birds. Chick. It is on this animal, as might be expected, that the greater part of the work on embryonic growth has been done. Hasselbalch, in the course of his work on the respiration of the chick embryo, obtained a regular series for a race not given. These corresponded well enough with the earlier data of Falck (also from an unknown breed), which were the first to be published, appearing in 1857. Hasselbalch's curve is shown in Fig. 21, in which for the first time we see the usual ' ' embryo-placenta relation ' ' in the form of a weight of extra-embryonic structures larger than the embryo in the earliest stages, but soon falling below it.^ relation between the two as follows: 4-0 C3.0 2'0 1«0 O Membranes • Embryo Hasselbalch calculated the See also Fig. 521. SECT. 2] AND WEIGHT 371 Wt. of embryo + wt. of membranes Day Wt. of embryo 8 9 ID 15 16 1-917 1-652 1-613 I -108 1-128 17 I-IIO 18 1-090 Other sets of weight data have been reported by Lamson & Edmond, by Murray and by Needham, for White Leghorn chicks, and by LeBreton & Schaeffer for chicks of an unspecified race. These are all placed in Table 3 of Appendix i, where it will be seen that the general agreement between them is good. The values obtained by Murray; Byerly^ and Schmalhausen are probably the most accurate, for the conditions were very carefully controlled. Hanan's values are lower than all the others. It is unfortunate that Schmalhausen does not state what breed of hen was used in his experiments, though he does mention that it was not genetically pure. Some early measurements by Welcker & Brandt are not included in the table, for they do not appear to be trustworthy. Other measurements which are useful are those of Edwards, who has published a peaked curve showing the length of the primitive streak during the first 50 hours of incubation. ^ Schmalhausen's work on the growth of parts of the chick embryo will be dealt with later : he was preceded by Falck, who measured and weighed various organs but did not use enough material to make his figures valuable to modern workers. Mammals. [a) Mouse. In 1923 LeBreton & Schaeffer published figures for the embryonic growth of the mouse, but these were not very numerous. The only other work on this subject is that of McDowell, Allen & McDowell, probably the most accurate and satisfactory study of prenatal growth in any form that at present exists. Their figures are given in Table 4 of Appendix i, and the curve obtained from them in Fig. 22. This is drawn on arithlog paper, the ordinates in logarithmic ruling giving the actual weights in gm., the abscissae in arithmetical ruling giving the age and the number of the individuals. On each day the range of the individual unclassified weights is shown by a vertical line which is itself used as a base-line for the frequency distribution of the classified individual weights. The number of cases in the ^ See also Fischel and Leva. 24-2 372 ON INCREASE IN SIZE [PT. Ill distribution is shown by the distance to the right of the vertical baseHnes, and can be judged by the frequency-scale at the bottom of the WEIGHT GRAMS 1 -000 •1000 •0100 •00100 •00010 •00001 1 e^. J .u^^ \-y^ ^^^ > \^' '^ ^ > , zz Y - k/^ T V^Z ' />H V^ t / / / / / J 1/ .c;r.Al E np FRFniifNniF'=^ 1 I 1 1 1 1 1 1 1 1 1 1 1 1 20 40 60 1 9 10 n 12 13 14 15 CONCEPTION AGE Fig. 22. 16 17 18 19 chart. The means, weighted by the number of individuals in each litter, are shown as dots on the vertical base-lines, and it is through these, of course, that the "normal curve" would be drawn. The continuous curve in the graph is one drawn to a formula which SECT. 2] AND WEIGHT 373 Foetus of albino rab will be discussed later, in Section 2-4 (p. 393). The lay-out so made reveals several interesting features; it appears, for example, that there are always individuals on a given day which are equal in weight to the mode of the day before. McDowell, Allen & McDowell consider that this is evidence of a possible delay of as much as 24 hours between copulation and fertilisation, but, whether this is so or not, it certainly equates with exactly similar variations found in the chick both in the early stages (primitive streak) and in the later ones of organ-growth. Further, the modes and means are generally close together, though less so at the beginning of development than at the end, and the latter do not approximate to a straight line. A glance at the graph also shows that the highest individual weights on each day tend to form a curve parallel with that of the means throughout development. (b) Rat. Donaldson's comprehensive monograph of 191 5 includes a discussion of the growth of the rat embryo, but much less work has been done on this animal than upon man, for in the latter 5 case the ad hoc labours of obstetricians have often provided much valuable material for the biologist. However, Stotsenberg's work gave a good account of the matter, and his figures are reproduced in Table 5 of Appendix i, and in Fig. 23. They begin from the 13th day after insemination, before which weighing is difficult, and they continue until birth, which takes place at the 22nd day. This prenatal period would appear to be one complete growth-cycle, if we may judge from the work of Donaldson, Dunn & Watson on the post-natal growth of the rat. Huber has studied the growth of the rat embryo in its earliest stages prior to fixation to the uterine wall. He states that the eggcell of the rat approaches the uterine end of the oviduct while in the two-cell stage, segmentation being slow and proceeding as the transit takes place. Fig. 24, reproduced from his monograph, is a 16 17 18 Fig. 23. 374 ON INCREASE IN SIZE [PT. Ill photograph of a model of the oviduct with its contained eggs. By reconstruction methods at a magnification of looo diameters of the ova Huber was able to determine the volume changes during segmentation as follows: Age bays ^ , Hours Stage Average vol. (cubic mm.) I Pronuclear 0-000156 2 o 2-cell 0-000162 3 I 4-cell 0-000173 3 17 8-cell 0-000184 1 1 -cell 0-000210 There would, therefore, appear to be a certain increase in volume during these very early stages, but as the specific gravity changes are not known it is difficult to understand what it may imply. There is at present a great gap in our knowledge of the embryonic growth Fig. 24. of the rat between the early point at which Ruber's studies end and the later one at which those of Stotsenberg begin. Huber himself suggested that the slow development of the ovum of the rat during its passage down the oviduct was best accounted for by the lack of any food-supply for an alecithic egg until fixation to the uterine wall had taken place. As the whole embryonic period of the rat is only 22 days, it is of great interest that the first four days should involve hardly any increase in size. This fact renders of no significance the calculated weights of rat foetuses given by Donaldson, Dunn & SECT. 2] AND WEIGHT 375 Watson in their earlier paper, for, in assuming that embryonic growth in the rat followed a quite similar course to that taking place in man and the rabbit, they did not allow for the long time taken for the rat egg to pass through the oviduct after fertilisation. Thus they arrived at the result that the rat embryo of 15 days should probably weigh 2-6 10 gm., whereas by direct measurement Stotsenberg found that it only weighs o- 1 68 gm. Their calculated figures are consequently not included in Appendix i. (c) Guinea-pig. The most usually quoted work on the embryonic growth of this animal used to be that of Read, who used a very indirect method of measuring it. He weighed the pregnant female every day between insemination and birth and then each foetus with its membranes and fluids, from which data, assuming that growth had taken place regularly, the weight of one embryo could be calculated. He concluded that the guinea-pig passes through two growthcycles during its intra-uterine life. But no satisfactory conclusion can really be drawn from such figures, subject as they are to all kinds of complicating factors, and, like the earlier ones of Minot on the guineapig, obtained in the same indirect way, they are better discarded. It is needless to point out that differences in the weight of mother + embryo due to defaecation, filling of caecum, etc., may amount to grams, while the weight of the embryo is still only milligrams. In the absence of any other figures, they had their importance, but in 1920 Draper made a complete study of the embryonic growth of the guinea-pig. Together with the few fragmentary (but direct) figures of Hensen, and the careful work of • 90 — 80 • / • / / • 70 " / • 60 - E .7 ( 50 - C tj 40 .«; •/ A * • 30 ~ • / • / 20 J 10 ) / 1 • 11 days, 1 1 10 20 30 40 50 60 70 Fig. 25. 376 ON INCREASE IN SIZE [PT. Ill Ibsen, and Ibsen & Ibsen, Draper's figures form the standard series, and are shown in Tables 6 and 7 of Appendix i. As is generally known, the guinea-pig differs from most other mammals in being born much later in its life-span than is usual, so that its lactation period is exceedingly short and it is able to eat green food a very few days after its birth. This is reflected in its gestation time which is relatively long. 20 '0 CO / / 15«0 —~ -0/ CO ^ E ® / a '^ / %. / a> / 10*0 — c ^ a:> K®/ ^ • uv- of 'membranes ? • 5«0 yA IX, Age ab which wbs. of embryos &, _>*x'^ membranes are equal O'O -1--^ i^;;i^ Aqe in days 15 20 25 30 35 40 Fig. 26. 45 50 55 60 65 During the 64 days of its development in utero, the guinea-pig increases its weight to about 85 gm. and its length to 10 cm. This process is shown in Fig. 25 taken from the figures of Draper. In Fig. 26, which gives an enlarged view of the lower part of the growthcurve, the increase in weight of the placenta, the membranes, and adnexa, together with the amniotic and allantoic fluid, is also shown. The extra-embryonic structures reach a more or less constant weight about two-thirds of the way through development, but, as can be seen from Table 5 of Appendix i, the values from which this curve was drawn are very divergent. In comparing the growth of the SECT. 2] AND WEIGHT 377 embryo with the growth of the membranes, it is interesting to see that for the first month the latter weigh much more than the former, after which, for a certain period, they grow together at the same rate. But soon the curves diverge, and the membranes hardly grow any more, while the embryo continues to increase greatly in size. Evidently when the membranes and placenta have reached a sufficient size to meet the utmost further demands of the embryo they grow no more. There can be little doubt that the size of the placenta exercises an influence on the growth of the embryo, and is of the highest importance from the point of view of embryonic nutrition. The amniotic liquid bears the same relationship as regards weight to the embryo as do the placenta and the membranes. 100 Fig. 27. Fig. 27 shows Draper's curve for the length of the embryonic guinea-pig. Ibsen's work led to much the same conclusions as regards the relations between embryo and adnexa as that of Draper. Ibsen found that the number of foetuses in the uterus exerted an effect on the growth-rate of each one, thus the larger the litter the slower the rate of growth of the individual foetus. The early growth of the placenta is more rapid than that of the foetus, but they reach the same weight on the 25th day, after which the foetus outstrips the placenta very soon. Placental weight and the weight of the membranes towards the end of pregnancy are closely correlated with uterine crowding, but this is not the case with the decidua basalis, which corresponds to the maternal part of the placenta. Minot considered that the amniotic fluid of the cow and of man decreased in 378 ON INCREASE IN SIZE [PT. Ill amount after the middle of pregnancy, but this was not found to be the case by Ibsen for the guinea-pig. Ibsen constructed from his experimental data the interesting diagram shown in Fig. 28, which shows the percentage of the whole system occupied by embryo, placenta, decidua basalis, and amniotic fluid from the 20th day onwards. The embryo does not rise in per cent, after the 55th day, the placenta remains very much the same all through, the decidua basalis is much smaller relatively at the end than at the beginning and so is the amniotic fluid. Up to the 50th day Ibsen found no correlation between foetus-weight and placenta-weight, but afterwards there is undoubtedly such a correlation, evidently due to crowding. [d) Rabbit. Much less work has been done on this form than might have been expected from its easy availability, but the figures of Chaine (the standard ones) are given in Table 8 of Appendix i, together with some early fragmentary ones of Fehling. Friedenthal also gives a few data which are shown in Table 9. {e) Dog. Liesenfeld, Dahmen & Junkersdorf made a thorough study of (unfortunately only 5!) dog foetuses. SECT. 2] AND WEIGHT 379 (/) Sheep. As early as 1847 Gurlt made a study of the increase in length of the foetus of the sheep, but Colin is the only investigator who has ever determined the growth in weight (see Fig. 29). Gurlt's figures, which are quite regular, are given in Tables 10 and II of Appendix i. FaureFremiet & Dragoiu, in the course of their extended work on the growth and chemical development of the embryonic lung in the sheep, made measurements of the growth of that organ, but did not give any data on the weights of their foetuses as a whole, a very unfortunate omission in view of the incompleteness of the literature on this subject. [g) Pig. The only extensive figures in existence for the embryonic growth of the pig are due to the careful work of Lowrey and of Warwick. These are given in Tables 12 and 14 of Appendix i. Lowrey's results will again be referred to in connection with the growth of individual organs and parts in the embryo. LeBreton & Schaeffer also measured and weighed a certain number of foetuses in the course of their classical work on the behaviour of the chemical nucleo-plasmatic ratio during 60 100 Days, Sheep (Colin) Fig. 29. embryonic development. Their figures are shown in Table 1 3 of i Appendix i. \ (Ji) Cow. The embryonic \ growth of the cow has been \ investigated by several workers whose results are shown in ' Table 15 of Appendix i. Fig. 30, constructed from Franck and Hammond, should be compared with Fig. 28 for the guinea-pig. {i) Man. The embryonic growth of man has been much studied, and many thousands of embryos have been weighed. His's studies have been the principal means of fixing the relation age/length, Months, Cow (Hammond). Fig. 30. 38o ON INCREASE IN SIZE [PT. Ill and Balthazard also gives figures for this, which will be found in Appendix i (Table i6). The earlier workers, Ahlfeld; Fehling; Hennig; Legou; Faucon; and Michaelis all obtained valuable data, but it was not until 1909 that a critical examination of them was made by Jackson who analysed the figures of his predecessors, and added a large number of new ones. His results gave a continuous curve from the earliest stages till birth, which agreed with the majority of the other investigators, but not perhaps with Hennig's curve (he gave no figures), which showed a very distinct slackening of growth about the sixth month, after which the same rate was resumed. If this phenomenon is real, it may possibly be associated, as Donaldson has suggested, with a transition from one growth-cycle to another, at the end of the sixth month, when the absolute weight begins to rise so rapidly. On the other hand, the mass of data which Quetelet and others after him have analysed regarding the growth of man throughout life, would seem to show that there are three growth-cycles only, one pre-natal one, one with its maximum at 5*5 years, and the third with its maximum at 16 years. Vignes' S-shaped curve for human embryonic growth is shown in Fig. 31. Bujard in 19 14 made a geometrical analysis of the early stages of the human embryo. Jackson measured the volume and weight of all the specimens in his own collection, and for the early stages also the volumes of the His-Ziegler models. His figures are given in Tables 17 and 18 of Appendix i, and the curve which he constructed from his own data as well as those of previous investigators is reproduced in Fig. 32. The 1 6th table of the appendix shows the volumes of the HisZiegler models, and demonstrates that the human embryo, like all others, is much exceeded in size by the yolk-sac during the earliest 3500 3398<jyP3405 31410^88 3000 im(p 2000 n E TO i. - C 1000 a3 '© mcf 1 500 100 JH no. 305 7 1 _. 1 260 280 300 1 1 i > f 1 1 1 ■ 3 4 5 Months Fig. 31 .270 290 310 Days SECT. 2] AND WEIGHT 381 stages of development. Jackson, who adopted the Minot method of measuring the growth-rate, concluded that the rate was 9999 per a) E o > E o <4 © E j3 CO 0) (0 E o^ c X) j: 3250 3000 2750 2500 2250 2000 1750 .1500 1250 1000 750 500 250 l: •


1 i <l / / !j 1. 1 AhlfelcL's dauta. ■ehling's da-bet Jaxjkson's data, L.egoas data, Michael is' data 1 i 1 * It 2. 3. i i '/• 4 5. L 1 ! 1 1 1 1 ! 1 J / i 1 / / / / / 1 t I 1 i / 1 1 / / / / / / 2// 7 / 1 ' / / -4 '•• / ^ j^' L^. ^ ^ ' • 50 75 1( 125 150 175 200 225 250 275 Age in Days Fig. 32. cent, for the first month, 74 for the second month, and 11 for the third month. This was in general agreement with the point of view taken by Miihlmann, and Jackson emphasised it further by showing that, if the weight of the embryonic membranes and fluids was taken 382 ON INCREASE IN SIZE [PT. Ill into account, the growth-rate for the first month was 574,999 per cent. What meaning can be attached to the enormous growth-rate figures which always appear when the Minot method is used for very young embryos must later be discussed. Fig. 32, which collects together the data of many observers, shows a considerable measure of unanimity between them. Ahlfeld's figures are the only ones which show serious divergence, and they were not taken into account by Jackson in his preparation of the "normal curve". Fig. 32 shows also by points the volume of the embryo at the different stages, but it does not differ much in value from the weight in grams. The specific gravity of the foetus does not, according to Jackson, remain precisely the same throughout develop SECT. 2] AND WEIGHT 383 merit, but changes from very slightly above i-o in the early stages to 1-05 in the later ones. Probably this is associated with the progressive loss of water as the embryo develops. One of the first to investigate quantitatively the growth of the human embryo was Boyd in 1861, who studied the weights of all the principal organs in embryos from 8*5 to 85 oz. He did not give figures for individual embryos from which a curve could be constructed, but simply divided them into large groups such as "prematurely-born", etc. Legou's data, already referred to, were worked over again in 1903 by Loisel. Zangemeister, more recently, has published figures for human embryonic growth — these are shown in Table 17 of Appendix i. Other data for embryonic growth in man will be found in the papers of Fesser ; Toldt ; Meyer ; Heuser ; Bedu ; Sombret ; Arnoljevic ; Stratz; Borri; Corrado; Balthazard & Dervieux; Ecker; Hamy; Kolliker; Cruickshank & Miller; Browne; and Friedenthal. Scammon & Calkins, who have made a great many measurements in recent years, have constructed a three-dimensional isometric projection, from which the height, weight and age of a human embryo may be read off if any one of them is known (see Fig. 33). The best recent paper on the whole subject is that of Streeter. Sandiford has shown that the weights and surface areas of foetuses fall on straight lines when plotted on double log paper. For further information on surface growth see Scammon & Klein. (j) Whale. Some information on the embryonic growth of the whale is contained in the papers of Harmer; Risting; Hinton; and Mcintosh & Wheeler, but it mostly concerns increase in length. 2-3. The General Nature of Embryonic Growth We may now turn to the theoretical aspect of the matter in the attempt to find out what interpretation can be placed upon them. We may in the first place take as a simple example of an embryonic growth-curve the work on the growth of the chick (White Leghorn) of H. A. Murray. Table 51 shows, firstly, the actual weights of the embryos on each successive day, secondly, the amount gained in each such 24-hour period, i.e. the amount of substance actually added on to itself by the embryo during the lapse of the time in question. This is known as the daily increment. In the next column the averages of the daily increments are placed, and these figures, known as the 384 ON INCREASE IN SIZE [pt. hi mid-increments, represent for each point which begins one period and ends another how much substance the embryo is adding on to itself between the times (a) half-way through the last period, and (b) half-way through the period to come. In other words, the midincrements convert the daily increments into terms of the points instead of the spaces between the points. If now the mid-increments Table 5 1 . Growth of the chick embryo ( White Leghorn) . H. A. Murray's figures. Percentage Daily Mid growth-rate .Age Wet weight Dry weight increment increment of dry in days (mg-) (mg.) (dry weight) (dry weight) substance 5 221 "•75 "•85 6 423 23-6 19-4 15-7 66-5 7 735 43-0 30-8 25-1 58-4 8 1,189 73-6 44-3 37-5 50-8 9 1,817 ii8-i 68-2 56-2 47-5 10 2,661 186-3 102-5 85-3 45-7 II 3,750 288-8 160-7 131-6 45-6 12 5.105 449-5 241-0 200-8 44.7 13 6,839 690-5 409-4 325-2 47-1 14 8,974 1099-0 575-0 492-0 44.7 15 1 1 ,460 1674-0 686-0 630-0 37-6 16 14,390 2360-0 730-0 708-0 30-0 17 17,950 3090-0 797-0 763-0 24-7 18 22,030 3887-0 832-0 814-0 20-9 19 26,670 4719-0 are expressed as percentages of their actual weights of embryo at their corresponding points in time, the last column or percentage growth-rate is obtained. This last calculation is, as will be seen, the only one so far made in the table which is open to serious criticism, and it is associated with the name of C. S. Minot, who was the first to propose it as a satisfactory measure of the growth-rate of an organism. When these figures are plotted the curves shown in Fig. 34 appear. The actual growth of the embryo expressed in terms of its SECT. 2] AND WEIGHT 385 weight at any given moment gives a curve which rises steadily till the observations cease without betraying much sign of any slackening. But the increment curves, on the other hand, show an unmistakable S-shape which is due to the fact that, for the earlier periods, the — o AbsoLate weight gms. wet • >> )> j> dry — Dajly incremenbl ^ « Mid » r^^ Fig. 34 weight gained each day is very little more on one day than the gain on the previous day, while, towards the middle of development, the daily increments and the mid-increments vary much more, each one being considerably higher than the one before. On the other hand, when the end of development is approaching, the increments each day, though far higher in absolute amount than those which were made in the early stages, do not differ so materially from one another, with NEI 25 386 ON INCREASE IN SIZE [PT. Ill the result that the curve slackens off and enters a slowly-rising phase again. It is possible, of course, to calculate the average daily increment for the whole developmental period (see Table 52), and the figures so obtained have been made the basis of a comparison of animals by Friedenthal. The fourth curve, that of the percentage growth-rate, Minot's curve, as it may be called, begins at a high level and continually descends, although in this instance there is a slight kink on it midway through development, which may for the moment be disregarded. All Minot curves begin at a high level and descend as development proceeds. Now, in many cases, it may happen that not only the increments but also the whole growthprocess itself slackens off towards the end of the period taken, in which circumstance the curve relating weight of animal at any given time to age will also have an S-shaped form. It has not escaped the perspicuity of those who have considered these phenomena that this S-shaped curve has a resemblance to the S-shaped curve of an autocatalysed monomolecular reaction, and this likeness will shortly be taken up at length. Table 52. Average daily increments. Friedenthal's figures. gm. gm. Mouse o-o8 Pig 14 Rat 0-24 Man 15 Ermine ... 03 Sheep ... 26 German marmot 1-2 Seal 30 Musk 1-5 Ass 53 Guinea-pig 2-0 Rhinoceros 90 Wolf 4 Stag ... 100 Puma 5-4 Horse ... 200 Bear 7-0 Hippopotamus . . 200 Lion 10 Camel ... 400 Roedeer... II Elephant 400 A more complicated example of the various types of growth-curves is afforded by Fig. 35 taken from Brody. It shows the growth throughout the life-span of the albino rat. The curve passing through the circles shows the course of growth ; it is, in fact, the weight of the whole animal at any given moment plotted against the age at that moment. The strongly indented curve, passing through the crosses, is the line showing increment in unit time. In the data of Murray for the embryo chick the absolute growth-curve rises steadily. SECT. 2] AND WEIGHT 387 and has no slackening off or self-inhibitory phase; the increment curve is therefore singly sigmoid. But here, when the absolute growth-curve is itself sigmoid, the increment curve is symmetrically sigmoid, rising to a maximum and then falhng away again to zero during the second phase. Finally, the corresponding Minot curve is shown by the line joining the triangles, and, as usual, it declines throughout growth from an initially very high value. n 70 30 «25 n) u CO a»20 m i. OU 60 <0 2 40 ■^30 0.20 gms. 200 cent day DaysO 10 20 30 40 50 60 70 80 90 100 HO 120 130 140 150 160 lyO^lB0 190 200 210 220 o 8 18 28 38 48 58 68 78 88 98 108 118 128138148 158 168 178 188198 % I Age o U Fig- 35 There are other ways, however, in which the subject of embryonic growth-curves can be introduced. Ostwald's classical work on growth in metazoa, which appeared in W. Roux's "Vortrage" series in 1908, laid great emphasis on the value of knowing the precise route through weight taken by an organism on its way from egg-cell to finished embryo. In Fig. 36, taken from his monograph, several different curves are shown relating time to weight. At the time A, at hatching or birth, for instance, the weight of the 25-2 388 ON INCREASE IN SIZE [PT. Ill organism is exactly the same in all four cases, but the manner in which the increase in weight has taken place is in the four cases profoundly different. It is certainly quite clear that the chemical embryologist, engaged in the attempt to understand the processes which contribute to the final result, must pay detailed attention to the path by which this final result is arrived at. The four different curves in Ostwald's figure would imply four very different sets of conditions within the developing embryo. An embryo which grew according to Curve I would change very rapidly in the beginning, and afterwards change progressively less rapidly as the curve became asymptotic. The reverse of this process would happen if the embryo grew according to Curve ii, for there the process continually increases in rapidity, and is proceeding at its fastest when the point A is reached. Curve iii, on the other hand, being S-shaped, would seem to indicate the presence of an autocatalytic process, for at first the growth proceeds faster and faster, but later on, after the point of symmetry of the curve has been reached and passed, slower and slower. Several such S-shaped curves superimposed on one another make up Curve iv. As far as is known, no growth takes place in the manner represented by Curve i, but rather in that of the other three curves, though our present knowledge does not enable us to say definitely which, except in certain cases. Ostwald's monograph should be referred to by those who wish to see how he continued the discussion of growth-curves, for it is probably the best presentation of the subject, and it was certainly written from a much less doctrinaire point of view than most of its successors. The general interpretations of embryonic growth-curves may be divided into several classes. They depend more than anything else, as will be clearly seen, upon how the facts are expressed. One way of expressing them led Minot to his "laws of cytomorphosis", another led Robertson to his "autocatalytic master-reaction", SECT. 2] AND WEIGHT 389 and, more recently, still other ways have been devised. The unprejudiced investigator cannot avoid a considerable measure of scepticism in considering the claims of one way of expressing the facts over another. 2-4. The Empirical Formulae We may first direct our attention to those presentations of the facts which do not carry with them any theoretical superstructure, but aim simply at describing the data in as short a manner as possible. The first of these "empirical formulae" was that of Roberts, who in 1906 pointed out that the growth of the human foetus could be regarded as nearly proportional to the cube of the age ; thus, if the weight in grams is W and the age in days T, the formula would be W — T^. But this was only very approximate, and the curve it gave did not fit the curve drawn through the experimental data with any accuracy. Roberts, indeed, stated that his formula gave results correct to "within an ounce at the third month". "Since the weight of an embryo of the third month," was Meyer's remark, "according to the best available evidence, is considerably less than an ounce, the accuracy of Roberts' method must be fully apparent without further comment." Tuttle next introduced an equation in which arbitrary constants were introduced, thus W = 50 {T — 2)^. Later still, Jackson, whose work on the human embryo has already been mentioned, proposed the formula : where W is the weight in grams and T the age in days. This fitted the experimental points much better than the formulae of Roberts and Tuttle, but was still rather deficient, especially in the very early period and the very late period. Henry & Bastien also proposed x^ + 2^xj> — 3q>'2 — i62y = o, where x = months andj = kilos. Duvoir has reviewed the other more or less practical rules which have from time to time been proposed, such as Casper's rule that, from the fifth month onwards, the age of an embryo in man can be found by dividing the height in centimetres by 5. Balthazard & Dervieux 390 ON INCREASE IN SIZE [pt. iii altered this formula to 5-6. Again, Mall's rule states that the number of days embryo age is equal to the square root of the foetal total length in centimetres x 100. Balthazard & Dervieux have also evolved formulae relating foetal age to the length of the limb-bones, e.g.: L = femur length x 5-6 + 8 cm. L = humerus length x 6-5 + 8 cm. L = tibia length x 6-5 + 8 cm. The use of empirical formulae in the description of human foetal growth has been carried to its greatest refinement in the work of Scammon & Calkins, whose formula, n- 2-5/, L2 holds with great exactitude from the third month onwards. Another of their formulae, T = 2-134 X o-iZ X o-ooiiL^, holds with rather less exactitude from 2-5 foetal months onwards. In both these cases, T is the menstrual age in lunar months, L the total or crown-heel length of the dead body in centimetres. They also found that W= (o-26L)3-i'>8 + 4-6, 3 108 , or Z, = 3-846 VW - 4-6, where W is the weight of the dead body. From these equations, it follows that 3 108/ 15S4 / T= 2-134 + 0-3846 VW — 4-6 + 0-01627 VW — 4-6, or T= 3-0 + ^•04.gVw — 0-012, orW= 0-561 — 0-366 T X 0-061 T^. The formula of Donaldson, Dunn «& Watson, for the post-natal growth of the white rat- up to 80 days, W = a + bT + cT^, and after 80 days, W = a log T — bT — c, was of the same type as the other equations mentioned, but it had the additional refinement of including constants, a, b and c, which were variable according to sex and other factors. Murray, in his study of the chemistry of embryonic development, SECT. 2] AND WEIGHT 391 found that his series of chick embryo-weights could be accurately described by the equation: T3-6 1-496' or W=KT^-\ where K = o-668. This was not unlike the Balthazard-Dervieux formula for the human embryo: T= 19-4 X \^W. 23 Diojs 5 10 U t2 13 14 Incub&lion 2052 Fig- 37 Murray plotted the log. wet weight against the incubation age, and obtained a curve concave to the abscissa (see Fig. 37) corresponding to the curve which McDowell, Allen & McDowell got for the mouse embryo (see Fig. 22). He also found that the relation of log. weight to log. age was a straight line as far as his series of weighings went, and showed that the weighings of Hasselbalch and of Lamson & Edmond fell on the same straight line (Fig. 38). Murray's formula gave very good agreement with his figures, but these did not extend further back than the fifth day of incubation. When, later, Needham 392 ON INCREASE IN SIZE [pT. m and Schmalhausen made weighings of embryos between the second and the sixth day of incubation, it was found that Murray's formula did not hold for these earlier stages. Fig. 39 taken from the paper of McDowell, Allen & McDowell, illustrates this point. The broken line is drawn to Murray's formula, and the dotted line is an extrapolation of his curve which I made on the assumption that embryos grew at the same rate before 5 days as between 5 and 7 days, i.e. exponentially. The circles with dotted centres are the values experimentally obtained by me, the dots are those obtained by Schmalhausen, and the cross within a circle is Murray's earliest figure. 4.6 4.2 >.? 3 J.4 Of 3 2.6 2.2 fi^ ft. %^ H » t gjp ° V^ t A ^ 8 \^

  • ^

" ^ X LSkiDSon i Ldmond (a^vep^oe of lo embrvos) • H&ss2lbailch (sln^l? ojeigninos) Single cuci§hin§s l^ken in the course of other experiments in this series ■ »^ ^ ^ -" ' 1 ^ 1 0.65 0.70 0.75 0.85 0.90 . 0.95 1.00 1J)5 L05 incubation a^e (dsajs) Fig. 38. U5 1.20 1.25 UO Murray's formula gives a line consistently above the experimental points for this early period, and the exponential extrapolation is quite at variance with them. But McDowell, Allen & McDowell evolved an equation which fitted these early points (the solid line) as well as all the later ones, as follows : log W = 3-436 log {10 (r- 0-5)} + 7-626. This new equation was based on an entirely different viewpoint from that of previous workers. McDowell and his collaborators regarded not "conception age" but "embryo age" as the right zero hour to take in growth calculations. It had always been assumed previously that conception or even insemination was the right starting-point, and Brody & Ragsdale and Brody based their method for finding age-equivalence in animals on this view, while Friedenthal SECT. 2] AND WEIGHT 393 had shown a similarity in relative growth-rates by plotting the log. weights against log. conception ages. McDowell and his collaborators, on the other hand, suggested that the time of growth ought rather to be calculated from the time at which the embryo first begins to have an axis, i.e. from the primitive-streak stage. Thus the major differences between the pre-natal growth of the guinea-pig, WEIGHT . INMGS. ■200 -180 •160 •140 -120 -100 • 80 ■ 60 ■ 40 - 20 DATA OF • SCHMALHAUSEN ONEEDHAM e MURRAY INCUBATION 1 AGE FREQUENCIES ©85 077 ©88 © ©83 77 ffi200 Fig. 39 the mouse, and the chick would be accounted for by the varying times taken to get through the preliminary work of arrangement and organisation. Processes such as gastrulation, according to this view, would involve a law of growth so different from the later axial type that no formula should be expected to cover the two. We have already seen in the case of the rat's egg that some considerable time may elapse between the time of fertilisation and that of fixation to the uterine wall, during which the supply of nutrient material may be 394 ON INCREASE IN SIZE [pt. iii very different from what afterwards obtains during embryonic development. There is, therefore, much justification for the view of McDowell and his collaborators. Brody himself had come nearly to this position without recognising it, for, in a paper published in 1923, he had pointed out that, during a short period in the early stages of growth (or regeneration) the apparent observed speed seems to be slower than would be expected. Thus the curve of the fitted equation cuts the time axis not at zero, the beginning of growth, but a little later. He advanced the explanation that Durbin had already applied to the initial slow phase in the regeneration of tadpole tails, namely, that a "cap of embryonic cells" was first formed, following in its growth quite different laws from the subsequent process as a whole. "It is suggested", said Brody, "that the apparent initial slow phase of growth of the individual from the fertilised egg is due to a similar qualitative growth." (Estimated weights of eggs are shown in Table 53.) McDowell and his collaborators proceeded to show that a similar formula would fit very well the curves of growth for the guinea-pig (Draper; Hensen; Ibsen & Ibsen), the mouse (McDowell, Allen & McDowell) and the chick (Murray; Needham; and Schmalhausen). For the mouse it was : log W = 3-649 log {10 (^ - 7-2)} + 8-6587; and for the guinea-pig it was : log W = 3-987 log {10 {t - 12)} + 5-1839; The significant thing about these empirical formulae is the deduction of a certain time in each case from the conception age, thus 7-2 days in the case of the mouse, 1 2 days in the case of the guinea-pig, and 0*5 day in the case of the chick (Allen & McDowell). The evidence on which McDowell and his collaborators rested their case for this shortening of the development time was drawn from various sources ; thus, from their own histological observations they found that the primitive streak in the mouse embryo appeared about the 7th day of development, for 6-day embryos show no mesoderm, while 7-day ones do, and usually the primitive groove as well. Sobotta's work is in agreement with this estimate. Their estimate for 12 days as the time taken for the embryo guinea-pig to reach the primitive-groove stage was based on the generally accepted work of BischoflT and Lieberkiihn, while, for the chick, Duval, whose illustrations are the SECT. 2] AND WEIGHT 395 usual "normaltafeln", shows the first appearance of the primitive groove at the loth hour of development, an assessment which is agreed to by Jenkinson and by Foster & Balfour (12th hour). Table 53. Probable dimensions of egg-cells. Weight Diameter in grams in fi (Friedenthal) (Hartmann) Gyclostomata Amphioxus ... o-oooooi — Lamprey ... ... 0-004 — Pisces Sturgeon ... 0*004 — Pike o-ooi — Amphibia Frog ... 0-004 — Reptilia Crocodile ... 40-0 — AVES Hen 45-0 — Aepyornis I20000-0 — Mammalia Monotremata Platypus — 2500 Spiny anteater ... o-i 3000 Marsupialia Dasyurus — 240 Opossum O-OOI 150 Edentata Armadillo — 80 Cetacea Whale — 140 Insectivora Mole — 125 Hedgehog — 100 Rodentia Mouse ... O-OOOOOOI 72 Rat — 72 Guinea-pig O-OOOOOI 80 Lagomorpha Rabbit 0-000003 125 Carnivora Dog 0-000003 140 Cat — 125 Ferret — 120 Ungulata Horse ... — 135 Sheep ... — 120 Goat — 140 Pig 0-000003 130 Cheiroptera Bat — 100 Lemurs Tarsius ... — 90 Primates Gibbon ... — "5 Macacus — 115 Gorilla — 135 Man 0-000004 135 "The general course of pre-natal growth in the mouse, the guineapig and the chick, can be expressed by straight lines relating the logarithms of the weight and the age only when age is counted from the beginning of the embryo proper." Such is the conclusion of McDowell and his collaborators, and, though it may seem barren 396 ON INCREASE IN SIZE [PT. Ill in theoretical results, it is nevertheless based on sounder considerations than the more ambitious ones of other workers*. It is probably legitimate to assume that the laws of growth before the formation of the embryonic axis are very different from what they are afterwards. It is also legitimate to assume that the differences between the velocity constants in the three formulae are due to the varying amount of organisation which has to go on in each case before the ^ formation of the primitive < groove. Fig. 40 shows the o straight-line relationships ? found to hold by McDowell, ^ Allen & McDowell in the 9 case of the guinea-pig, mouse ^ and chick, and Fig. 41 gives further examples, from which further variants of McDowell's formulae could easily be calculated. Clearly in embryonic growth log. weight is always proportional to log. time. With respect to Fig. 39, in which the weights of very young chick embryos are given, it should be noted that the discrepancy would naturally be expected to occur only in the early stages, for in the later ones the difference between conception age and embryo age would be a smaller percentage of the total. The

  • But see p. 427.

100 • n r t I 7 1 1 o- ono / I 9 • / 7 > '-A I J \ // / ^ // /i! J '■il P ■h 7 VI' •IOC

. L / i rT ■// // / •OK s // ' / i / / 1 f UJ ho J h h^ '§ c HICK •GO •^ // ^ MURF ?AY r/ NEEDHAM c • schmalhausem; .1 •00010 I GL INFA PIG / 1 BSEN ♦ DRAPER + V.HENSEN •00001 1 2 4 10 20 4060100 200 400 600 EMBRYO AGE IN }\J"^ OF DAY Fig. 40. SECT. 2] AND WEIGHT 397 averages for the early embryos reveal the difference by bending away from the lines drawn on the basis of incubation or conception age. Schmalhausen, continuing earlier work on the growth of bacilli and protozoa, has also put forward empirical formulae for the embryonic growth of the chick, but his equation \^W= T, while fundamentally the same as that of Murray, has no velocity constant. Fig. 42, taken from Schmalhausen's paper, shows that the cube root of the weight plotted against the age only gives an approximately straight line. Schmalhausen has included in the same figure the curves obtained by other methods; thus curve P' is the Minot (percentage growth-rate) curve for the wet weight, and Ps' for dry weight, while the curve marked log o- ip is the log. weight plotted against the age. As we have already seen, in the case of McDowell's figures for the mouse, and Murray's figures for the chick, this value always gives a curve rising concave to the abscissa. The curves P and Ps in Fig. 42 represent the absolute wet and dry weights respectively. Other empirical formulae have been proposed for growth-processes from time to time. Embryonic growth can be expressed roughly by exponential curves ; thus: W = wp\ where W is the mass of the embryo at time t, w the original mass, and p a constant. Thus the equation of an exponential curve is one in which the power is always changing. Janisch has given a discussion of the use of the exponential curve in all departments of biology, and in it he shows how important this relation is in growth phenomena. The "law of compound interest "however, put forward byBlackman in 1919 for the growth of plants, and which has been shown by Luyet to be a special form of the exponential relation W= w {i + ry, has not so far been of any assistance in describing embryonic growth. Another form of the exponential curve, the arithmetical progression method, which gives the equation log W - At, 398 ON INCREASE IN SIZE [PT. Ill where A is 3. characteristic constant, was used by Faure-Fremiet in 1922 for describing the growth of a Vorticella colony, but the con 100 CHICK Hasselbalch 1900 O Series (a) D « (h) Bohr& Hasselbalch 1900 • LamsonS^Edmond 1914 Ijin 1917 O LeBrebon8(Schaeffer1923 B Murray 1926 Schmalhausen 1926 ^ Needham 1927 RAT Stobsenberg 1915 MOUSE ® McDowell etc. 1927 <§> LeBreton&Schaeffer1923 GUINEA-PIG O Draper 1920 SHEEP ® Colin 1888 PIG O LeBrebon&.Scliaeffer1923 O Warwick 1928 TROUT X Scheminski 1922 RABBIT O Chaine 1911 « Lochhead&,Cramer1908 0-0 1 00 days Fig. 41. ditions there are too far removed from those of embryonic development to make it worth while considering this aspect of the subject in detail. The formula proposed by Faure-Fremiet for the growth SECT. 2] AND WEIGHT 399 of the epithelium of the foetal lung is, however, of greater interest i W = At^w + Bfiw + Ctw + Dw, where w is the total weight of the lung at the time in question. But here the number of arbitrary constants is so large that we reach the point where the question naturally arises whether an empirical /q^ 0,1 p 7 8 3 10 i1 12 13 1'^■ 15 16 17 18 19 20 21 Fig. 42. formula is worth looking for at all. The more complicated it is, the less valuable it is, in view of the fact that, in any case, it is not intended to give us an idea of the basic factors underlying the process. 2-5. Percentage Growth-rate and the Mitotic Index We have so far been examining the results of those investigators who have taken the curves obtained by simply plotting the weight of the embryo each day during development against the time, and who have endeavoured to find a correct mathematical expression for them without a preconceived theory. We have now to consider 400 ON INCREASE IN SIZE [pt. iii the work of those who have infused more theory into their treatment of the experimental facts. Before 1890 there was no regularity in the way in which experimentalists examined their data on growth. But about that time Minot began a long series of investigations on the growth of animals, mainly the guinea-pig and the rat, in which he introduced a new method, namely, the evaluation of the growth-rate by taking it as the increment in per cent, of the weight of the animal at the beginning of the period in question. Some workers, e.g. Preyer, had already adopted this plan. The percentage growth-rate has always been found to decline enormously as development proceeds, an observation which led Minot to say that the embryo gets oldest most quickly when it is youngest. This apparently paradoxical statement drew a good deal of attention to his work at the time, and his book. The Problem of Age, Growth, and Death, included many such graphs showing how rate of senescence was greatest in the earliest periods. One of these is reproduced here as Fig. 43. Another contribution of Minot's was the conception of the "mitotic index", or the number of mitosing cells per 1000 cells in a tissue. He did not himself find time to do more than a few of these laborious counts, but he gave the following figures, which showed that the mitotic index declined with age: Development of rabbit foetus (days) Tissue Mitotic index 7-5 Ectoderm 18 Mesoderm 17 Endoderm 18 lO-O Ectoderm 14 Mesoderm 13 Endoderm 15 Blood 10 130 Spinal cord Connective tissue II 10 Liver II Skin 10 Excretory tissue Muscle 6 6 These data lent weight to his principal conclusion, which was that the younger the embryo the more rapidly it aged. Practically nothing more was done along these lines until Olivo & Slavich in 1929 reported a large series of figures for the mitotic index of the developing heart in the chick. SECT. 2] AND WEIGHT 401 Davs Mitotic period in calculatec development index hours minutes 38 minu 2 22-5 19 42 — 3 21-2 20 55 — 5 169 26 1 1 15-7 7 150 29 24 94 9 8-3 53 II IO-2 II lO-O 44 3 4-8 13 7-0 70 15 57 15 4-4 100 33 40 17 3-7 117 39 2-5 19 2-8 155 55 4-2 21 1-2 363 7 1-8 10 (after hatching) 00 — The fall in the mitotic index ran closely parallel with the fall in the percentage growth-rate of the organ, as determined by a special series of weighings. The time taken by one mitosis was calculated from these data by Olivo & Slavich : it turned out to be constant at 38 minutes. But the intermitotic period grew longer and longer, indicating that the later growth consists less than the former of proliferation and more of increase in size of the cells already formed. o o\ ( Males ] Z 5 811 17 23 29 3S3« 45 75 80 105 120 135 150 165 180 185 210 dstye 241 Fig- 43 For a long time Minot's way of looking at embryonic growth in particular and growth in general was universally adopted, e.g. by Jenkinson, and even at the present time it is much used. But the Minot curve is undoubtedly based on a fallacy, and it was not long before a feeling that this was so began to arise. It was perhaps intensified by the appearance of such estimates as that of Muhlmann, who worked out the growth-rates in early stages of N E I . 26 402 ON INCREASE IN SIZE [pt. iii embryonic development in man as 3650 per cent, and above. It was not unnatural to enquire whether the Minot growth-rate of the original dividing egg-cell was even finite. The dissatisfaction was voiced in 191 7 by d'Arcy Thompson, who wrote as follows: "It was apparently from a feeling that the velocity of growth ought in some way to be equated with the mass of the growing structure that Minot introduced a curious and (as it seems to me) an unhappy method of representing growth in the form of what he called 'percentage-curves'. Now when we plot actual length against time we have a perfectly definite thing. When we differentiate this LjT we have dLjdT which is of course velocity, and from this by a second differentiation we obtain d^LjdT^, that is to say, the acceleration. But when you take percentages of jv, you are determining dyly and when you plot this against dx you have —-^ or dx — ^ or - . ^ , that is to say, you are multiplying the thing you wish y.dx y dx to represent by another quantity which is itself continually varying, and the result is that you are dealing with something very much less easily grasped by the mind than the original factors. Minot is of course dealing with a perfectly legitimate function of x and y and his method is practically tantamount to plotting logy against x^ that is to say, the logarithm of the increment against the time. [Cf. log. weight-age curves.] This could only be defended and justified if it led to some simple result, lOr instance if it gave us a straight line, or some other simpler curve than our usual curves of growth". This criticism was justified, for the Minot curve is certainly no simpler than the untouched growth curves ; it merely falls instead of rising. But d'Arcy Thompson did not point out the presence of a definite fallacy in Minot's way of looking at growth, a physiological rather than a mathematical one. This was grasped by Brody, who has written as follows: "Minot's method for computing growthrates gives an exaggerated decline in the percentage rates of growth with increasing age simply because an arbitrary unit of time, e.g. a week, does not have the same significance at different ages. It is, for example, entirely appropriate to express the gain in weight during a week as a percentage of the weight at the beginning of the week (Minot's method) for a 6-month old chicken, because the weights (i.e. the number of cells or other reproducing units) at the beginning SECT. 2] AND WEIGHT 403 and end of the week are nearly the same as compared to the gain in weight. But to express the gain in weight during a week as a percentage of the weight of the body at the beginning of the week for a 7-day old chick embryo would be quite fallacious. It would correspond to expressing the gain in population in the U.S.A. from 1666 to 1927 as a percentage of the size of the population in 1666. The growth of the population of the U.S.A. in 1927 is proportional to the population in 1927 and not to the population in 1666. Similarly the number of cells produced in a 7-day old chick embryo should be functionally related to the number of reproducing cells (i.e. the weight) of the chick at 7 days of age and not to the number of cells at I day of age. In brief, growth is a continuous process and the rate of growth at every instant should be functionally related to the number of reproducing units at the given instant and not to the number of reproducing units which existed in some relatively remote past". In other words, Brody would prefer to ask not how much 100 gm. of embryo add on to themselves during the immediately succeeding period, but rather how many grams of those 1 00 gm. had been added on during a short preceding period. Murray's modification of Minot's method, in which the mid-increments instead of the daily increments are used as the basis for calculation, goes some way to meet Brody's criticism, for it enquires how much 100 gm. of embryo add on to themselves during half the preceding and half the following period, thus speaking in terms of a more instantaneous measure. Brody himself has made use of a similar amelioration. However, Brody's real point is that the fault lies in choosing an arbitrary length of time interval for all stages of development, in spite of the fact that they cannot possibly be equivalent for the embryo. Brody might say that the embryo cannot be regarded as having been given an equal chance to accomplish its growth in each of the daily periods throughout its development. On the other hand, it might be argued that, though this is doubtless true as regards growth in weight, it is not true with respect to the activities of the embryo as a whole, which include many other processes, such as chemical differentiation. Taking the embryo as a whole in all its activities, the arbitrarily chosen and invariant period might be regarded as an adequate one. As we shall see later, this is essentially the same argument as that used by Murray against Robertson. 26-2 404 ON INCREASE IN SIZE [PT. Ill It may, however, be concluded that the Minot curve is only useful provided no theoretical conclusions are drawn from it, and that it is retained simply as a convenient method of comparing processes. Brody's own theories will be discussed later. As against the theory which Minot built up from his experiments with growing animals, Murray has brought forward one convincing -0.15 1 -0.14 n 4 «. I \ -0.12 '0.11 -0.10 -0.09 \ \ \ \ \ -0.08 ^ -0.07 -0X?6 • -0.05 \; > \ \ -0.04 0.03 ■0.02 K X X ^^ ^^--^ ^ '0.01 A^e 5 6 7 8 9 10 tl 12 13 14 15 16 17 IS 19 ^ Fk ?-44 argument. Minot's theory of cytomorphosis involved the following propositions: (i) that the rate of growth depends on the degree of senescence, (2) that senescence is at its maximum when development begins, (3) that the rate of senescence decreases with age, and (4) that death results from the differentiation of cells. But, as Murray says, we have no real evidence to show us that the "degree of aliveness" at any given moment is in any way connected with the velocity of growth at that particular moment, or, more correctly, that the latter SECT. 2] AND WEIGHT 405 value is a true measure of the former, "There are other and perhaps more significant phenomena", said Murray, "than the growth rate, which change with age." Murray himself proposed the use of a variant of the usual Minot curve by differentiating twice instead of once so as to get the acceleration and not the velocity. Thus, after having found the percentage growth-rate by the equation dt[_w\ t' where K ^ 3-6, he went on to find the negative acceleration for each day during embryonic growth : dw d dt dt = - Kt^. This value, so obtained, is, as it were, the negative increment of the percentage growth-rate, and shows a regularly declining curve (for the chick) as in Fig. 44. Such a curve must obviously suffer from the same disadvantages as the curve from which it is derived, and does not escape from Brody's criticism that the arbitrarily chosen time-units are incommensurable at different developmental stages. In 1922 Przibram observed that in many cases of post-embryonic growth the curve obtained by calculating according to Minot's method was extremely like a regular hyperbola, but he did not find that this was true for any example of embryonic growth. We shall see later what further use of this idea has been made, 2-6. Yolk Absorption-rate Another way of regarding embryonic growth (of a lecithic Ggg) is to concentrate attention on the whole system, instead of upon the growth of the embryo alone. It was in this way that Gray treated the development of the trout embryo in the paper already mentioned (p. 369 and Appendix i). He assumed that the rate of growth of the embryo was proportional to the dry weight of the embryo and to the dry weight of the remaining yolk. This idea had already been introduced for the trout by Kronfeld & Scheminzki (see p. 369 and Fig. 41) but Gray's figures were much more complete, and showed very clearly a falling off of growth towards the end of 4o6 ON INCREASE IN SIZE [pt. iii pre-natal life, when the yolk was becoming exhausted. Thus the wet weight of an embryo plotted against the time gave an S-shaped curve, which, however, was not symmetrical, for it had a point of inflection after about 70 per cent, instead of 50 per cent, of the development had been completed. This was of course well shown on the increment curve, which was skewly bell-shaped. Assuming that growth was proportional to the amount of yolk remaining as well as to the size of the embryo already formed, Gray developed an equation ^ (where x is the weight of the embryo at time t,yQ the total yolk in the unfertilised egg, K^, the amount of yolk combusted by one gram of embryo divided by the constant k in the equation dx J It = '"^' X and y being weight of embryo and weight of yolk respectively) which he considered accounted very well for the observed facts. He deduced from it that there should be a period at the end of development when the wet weight of the larva (the whole system, embryo plus yolk) is decreasing, although the wet weight of the embryo itself is still increasing. During the major part of development the wet weight of both would increase, owing to the absorption of water from outside. From the equations the maximum weight of the larva should be reached when o-86 gm. of yolk is still unconsumed, and in actual fact Gray found the peak at a point when i*io gm. \^as yet remaining. Another possibility used by Gray to test his hypothesis was that as it was unlikely that the temperature coefficient of the growth process would be the same as that of the catabolism going on, there ought to be a measurable difference in the size of fishes raised at various temperatures at the end of their development. Experiments designed to reveal such differences gave the following results : Mean weight of 100 embryos at Temperature (°C.) the end of incubation (gm.) 15 i3-35±o-i6 ID i507±o-i8 so that the higher temperature not only accelerated the process, but, by accelerating the combustion more than the storage, led to a SECT. 2] AND WEIGHT 407 smaller finished fish. These results are curious, for it is generallyunderstood that temperature changes alter the rate at which a growth-process goes on yet not the amount of end-product formed. The work of Barthelemy & Bonnet on the frog is an exact parallel to that of Gray on the trout, for these workers raised frog embryos at different temperatures with the following results : (P.E.G.) Temperature Dry wt. of 300 Dry wt. of 300 Dry wt. of embryos (°C.) eggs (gm.) embryos (gm.) Dry wt. of eggs 8 0-378 0-334 0-88 10 0-824 0-423 0-51 14 0-594 0-318 0-54 21 0-708 0-346 0-49 Temperature Dry wt. of 70 Dry wt. of 70 Dry wt. of embryos (°C.) eggs (gm.) embryos (gm.) Dry wt. of eggs 8 0-092 0-079 0-85 10 0-128 o-ioi 0-79 14 0-097 0-082 0-84 21 0-107 o-o88 0-83 (P.E.C.) If the second and third columns of this table are compared it will be seen that in the first series the French workers did get results like those of Gray, i.e. the higher the temperature the greater the combustions and the less the storage, but that in the second series there was no such effect to be observed. The Plastic Efficiency Coefficient (P.E.C.) is the most convenient way of expressing this relation (see Section 6- 1 o) . According to Gray it should change with temperature, for assuming his trout embryos to have the same percentage composition, no matter what the temperature, those raised at 15° would contain (each) 21*3 mgm. solid and consequently (since the eggs contain 43-4 mgm. solid) would have a P.E.G. of 0-50, while those raised at 10° would contain (each) 24-2 mgm. solid and consequently would have a P.E.C. of 0*56. I shall return to this subject in the section on general metabolism of the embryo; here it is only necessary to remark that the subject is clearly not yet settled and requires much more attention than it has so far received. At the same time, returning to the main theme, it must be remembered that Gray only made use of these temperature phenomena as one of the supports for his theory. He drew another support from the fact that if his equations were correct, the product obtained by multiplying the dry weight of yolk by the dry weight of embryo should be at a maximum on the 71st day of development. This he found to be actually the case, as is 4o8 ON INCREASE IN SIZE [PT. Ill shown in Fig. 45. And the growth-rate per gram of embryo (Minot growth-rate divided by 100?) was also proportional to the amount of yolk remaining. Gray himself indicated a number of criticisms which might be brought against his views. Thus there is little a priori reason for supposing that the growth-rate of the embryo should be determined by the amount of yolk in the yolk-sac, for apart from anything else, the 10 20 30 40 50 60 70 Fig. 45. Days of development. 80 90 100 amount and nature of the syncytium in the yolk-sac wall might be a limiting factor. One would also expect that in the beginning when the yolk is very large compared to the embryo, the yolk would be present in excess, and would not exert any influence on growth-rate. 2-7. The Autocatakinetic Formulae The workers mentioned so far might be divided into three groups, those who have elaborated empirical formulae, those who have adopted the Minot point of view, and those who have treated the SECT. 2] AND WEIGHT 409 yolk plus embryo as one growth-system. Now a fourth and very large group consists of those who have been greatly impressed with the similarity which some empirical growth-curves show with the curve for a monomolecular autocatalytic reaction. This manner of looking at the subject is associated mainly with the name of Brailsford Robertson, who has in many papers and in a book specially devoted to the matter put it forward as the most fundamental approach to growth. Robertson was not, however, the first to notice the likeness. As early as 1899 it had been referred to by Errera, and Ostwald a little later. According to Monnier, Chodat of Geneva paid some attention to it in 1904. "One may regard growth, as M. Chodat has suggested", said Monnier, "as a complicated chemical reaction in which the living cell is the catalyst and the substances present are water, salts, and CO2." Four years later (in 1908 on May 9) Ostwald's monograph on growth was published in the form of an inaugural dissertation, and only ten days later Robertson's first paper appeared in the Archivf. Entwicklungsmechanik. Ostwald had treated the question in a rather unmathematical manner, but had fully explained the nature of his hypothesis ; Robertson, on the other hand, gave the S-shaped curve a detailed mathematical treatment. " The carrying out of a progressive development in time has in animals a single characteristic type; the rapidity of the process begins at a low value, increases with the continuance of the action and falls off again at the end, in other words the type of curve is S-shaped." This was as far as Ostwald went, but he did not fail to point out that the S-shaped curve was identical with that of an autocatalytic or an autocatakinetic reaction. An important point to note is that Ostwald's curves were all curves of absolute weight — he did not in any instance plot increment curves. Robertson began by an explanation of the mathematical properties of the autocatalytic curve. The differential equation characteristic of the initial stages of an autocatalytic monomolecular reaction is as follows : , -r = k-iX [a — x), which expresses in mathematical symbols the fact that the velocity of the transformation is, at any instant, proportional to the amount of material which is undergoing change and to the amount of material which has already undergone change. If, however, the reaction has 4IO ON INCREASE IN SIZE [PT, III proceeded so far that the depressant effect of the products is measurable, then the previous equation becomes — = k^x [a — x) — k^x^. Now when the reaction has proceeded half-way to equilibrium, i.e. when X = la, the equation becomes X or log log A - X A — X = Ak{t- tj), Amount transformed and this is the well-known equation for the S-shaped curve, where X represents the body-weight (not the increment) at time t, A represents the maximum of final weight which the organism is to reach, ^1 the time at which half this maximum body-weight has been attained, and K a constant which has to be determined from a known value of a; at a given time t. By differentiating, it may be found that dx/dt is at a maximum when X = \A — in other words, that the rate of increase in weight is greatest when half the autocatalytic curve has been passed through. Robertson proceeded to apply his calculation to the growth of white rats, figures for which had been reported by Donaldson, Dunn, & Watson, with excellent results over part of ^^^' 4^* the curve, though not when the age amounted to more than lOO days. At the time it seemed as if this was a most convincing example of the value of the autocatalytic theory, but maturer consideration showed that this lOO days was but a third of the possible life-span, and the comparison of the two curves as made by Lotka, for instance, does not look so impressive. SECT. 2] AND WEIGHT 411 Fig. 46, taken from Robertson's first paper, illustrates the relation between the curves for the three chemical reactions, while in Fig. 47 is seen the theoretical and the experimental curve compared. In this same first paper, Robertson applied his autocatalytic equation to the growth of man, frog, a vine, and to certain organs. The frog figures, which were those of Davenport, were the nearest approach 50 150 200 Age in days Fig. 47 300 to embryonic growth dealt with by Robertson, and they are given in Table 54. They showed a good measure of agreement, but whether it was right to say, as Robertson did in his summary, that "in all probability cell growth or the synthesis of cytoplasm is an autocatalytic reaction" is a question that subsequent workers have not by any means answered with a bald affirmative, Robertson pointed out that his new interpretation of growth curves fitted in very well with the views that had already been advanced by Loeb. Loeb had suggested that the processes of cell-division and 412 ON INCREASE IN SIZE [PT. Ill growth were simply expressions of a more or less rapid return to the chemical equilibrium between nucleus and cytoplasm which had been temporarily shifted through the process of fertilisation. He thought that there occurred in the sea-urchin's egg during the early stages of its development a great synthesis of nuclear material, and that this synthesis progressed, roughly speaking, all the more rapidly the more nuclein was formed. The velocity of nuclear synthesis, said Loeb, increases with lapse of time in geometrical progression. Further, various observers, using the Q^k, formula, had concluded that the growth-process had a "chemical temperature coefficient", so Robertson felt fully justified in speaking of a "master-reaction" of growth, and in thinking of it as autocatalytic — a reaction, which, because it would be slower than any other, would act as the limiting factor of growth, and would impress its own particular character on the general appearance of the whole process from outside. Table 54. Larval growth of frog [tadpole). Robertson's figures Davenport's calculated ft-om Days after experimental autocatalytic hatching figures (mgm.) formula (mgm.) I 1-83 1-64 2 3 4 2-00 2-03 — — 5 3-43 3-90 7 8 5-05 6-00 9 10-40 9-OI 14 23-52 23-46 41 lOI-OO 110-90 81 1989-90 112-00*

  • "Obviously another growth-cycle supervening", said Robertson

In his later writings, Robertson added further demonstrations of the applicability of the autocatalysis equation to growth, but he also added a large amount of extremely speculative considerations, such as the "nutrient level", the "endogenous catalyser", etc., into which we cannot here enter. Some of his suggestions, for example that the autocatalyst is lecithin, may be regarded as now definitely out of court. He showed, however, that the pre-natal growth in man (using Zangemeister's figures) was susceptible of description in an autocatalytic curve with its early " autokinetic " phase (Robertson's terminology), and its late "autostatic" phase. He also used, for the SECT. 2] AND WEIGHT 413 last part of the curve, data on the weight of new-born infants, born at different times, some earHer and some later than the normal. We have, moreover, already seen that the curve constructed by Vignes for human embryonic growth shows an S-shaped conformation. In 1926 Robertson published a long paper in which he reviewed the work which he and his collaborators had done on the subject of the autocatalytic theory of growth. Here he showed that the fall in the relative value of the velocity constant of (asymmetric) autocatalysis during the embryonic growth of the mouse followed almost exactly the same curve as the fall in the chemical nucleoplasmatic ratio as determined by LeBreton & Schaeffer (see Section 10-2). It is not clear, however, what significance is to be attached to this finding, especially as Crozier and Brown have shown that at least two velocity constants must be postulated in a given cycle. Robertson himself drew no theoretical conclusion from it. Robertson regards the several growth-cycles distinguishable in the life of an animal as being independent, in that they each have a different catalyst. The first of these in the case of the mouse has an equation of the type where x is the growth attained at time /, A the maximum growth attainable in the cycle under consideration, t the time required to attain half the maximum growth and k a specific velocity constant. B is another specific constant, an index of the asymmetry of the curve. The second and third cycles have the formula log-^-^ = A(^-0, where the symbols are as before. He considers that skewness or asymmetry originates probably in a progressive diminution of the velocity constant as described above. Robertson's suggestions were not allowed to go uncriticised. Meyer, in his report of data for the growth of the human foetus, took occasion to attack both them and the traditional Minot standpoint. On the whole, Robertson's autocatalysis theory emerged with less damage than Minot's cytomorphosis theory. Meyer brought into the light what had been one of the most disturbing features of the Minot method, namely, the absurd values which the percentage growth 414 ON INCREASE IN SIZE [pt. iii rate has in the earliest stages, e.g. five hundred and forty bilHon per cent. But, as Meyer seems to have completely failed to understand that Robertson and Minot were using different methods in calculating the "growth-rate", his criticism of Robertson was not of any importance. On the other hand, he did draw attention to the fact that growth-curves can be very misleading if it is not remembered that, though in the early stages the absolute growth is minute, the relative growth is enormous. "The weight of the impregnated human ovum", said Meyer, "is approximately 0-005 ^§"^-5 ^.nd yet investigators in all seriousness indicate its weight on a short ordinate reading in grams or even hundreds of grams. Little wonder, then, perhaps, that Robertson, Ostwald, and Read have unwittingly assumed that the curve of growth in man and mammals hugs the abscissa for several months as the curve of autocatalysis does." This was a good tilt at a common fallacy, but Meyer did not point out that it could be remedied by using log. paper, and he left it quite open to Robertson to reply that, even when strictly comparable quantities were taken, the S-shaped curve or a succession of S-shaped curves still resulted. More serious criticisms than these have, however, been brought against the Robertson method of treating embryonic growth. Luyet has pointed out that it may suffer more than the other methods from illusory difTerences in material. Again, Murray has written as follows: "(i) The formula demands the introduction of three separate constants which must be separately determined for every set of figures collected. (2) The equation does not give the weight as a function of age throughout life but only during an arbitrarily selected part of the growth-cycle. For these two reasons the equation as a practical simplification is not of great value. If the equation were in such a form that knowing the species, the age, the T°, and other environmental variables, one might calculate the weight and growth-rate, it might be of use. But as it stands now, it is necessary in each case to collect complete statistics and then find a mathematical expression of the figures obtained. For instance, all three constants in the equation for the growth of South Australian males differ from the constants used in the equation for the South Australian females. As one cannot extrapolate, the formula, like the man with one talent, returns what it receives. In fact as it covers only a section of the growth-curve, it yields less information than the SECT. 2] AND WEIGHT 415 original data. Moreover, as a rational account this description of the synthetic processes of growth is misleading, since (3) by this theory the growth-rate is proportional to the increment gain in weight regardless of the weight of the organism, or in other words, disregarding the amount or concentration of the reacting substances. (4) Figures for the growth of colonies as well as of individual organisms are said to be described by autocatalytic equations and are classed together. When growth is expressed in terms of percentage increase in mass, however, the important distinction between phylogenetic and ontogenetic growth is made evident ; for the former, after a short latent period, in the presence of an experimentally modified environment proceeds at a constant rate [cf Richards], while the latter does not. The S-shaped curve is the result of a limited and unrefreshed culture medium. The individual organism, however, instead of maintaining a constant growth-rate shows from the beginning considerable negative acceleration. (5) The S-shaped curve is not specific, for there are some physico-chemical processes not considered to be autocatalytic which are described by a similar curve. Finally, and this is the main objection, (6) chemical diflferentiation is not taken into account by the autocatalytic theory, which is based on the conception that there is some one master monomolecular reaction, which, being the slowest of the chain of reactions concerned in the phenomenon of growth, determines the velocity of the entire process. As there is no direct way of measuring the product of the masterreaction, the increase in the body-weight of the whole embryo is taken to represent the product. In view of the marked changes in chemical constitution which take place in the tissues with age there is no reason to suppose, and in fact it is extremely unlikely, that the total weight can be taken as an index of the amount or concentration of any one chemical substance". The tendency that has existed in the past to take the simple weight of the embryo as the sign par excellence of its developmental stage and its "aliveness" is only another case of the reluctance to judge by "ensemble", as Broca called it, instead of by single indices, which was so long the bane of physical anthropology. In this case, it is the increasing application of chemical methods to the embryo which has shown the superficiality of regarding mere weight as the pre-eminent factor. Lotka has drawn attention to this point of view, and is inclined to compare embryonic growth to the growth of a population such as Pearl & Parker have studied 4i6 ON INCREASE IN SIZE [pt. iii in Drosophila. Janisch treats the S-shaped growth-curve as the reciprocal of a catenary exponential curve. These presentations have the advantage that they do not prejudge the issue from a physicochemical angle. Exception to Robertson's views has been taken on quite other grounds by Snell, who points out that Robertson's equation ^ = K^Ax - K^x" dt ^ ^ and Crozier's modification of it ^ = (r^ + K,x) {A-x) (where x is the concentration of the end-product at time t, A the concentration of the substrate at time t, and K^ and K^ the velocity constants of the forward and reverse reactions respectively) do not take into account the fact that the system concerned is not a closed one. All the time the embryo is growing it is also eating, i.e. absorbing nutritive material, and in addition it is giving out waste products. Accordingly these equations derived from the law of mass action as we know it in the inorganic world do not allow for the effect of increasing size on the concentration of the reagents involved in growth. The equations hold true only on the condition that the volume occupied by the resulting substances remains constant, and since a growing organism is constantly increasing in -volume, this condition is not met. "When a chemical process is carried out in the laboratory," said Snell, "the reagents are ordinarily dissolved in water or some similar solvent, and the volume of the solvent is kept constant throughout the whole process. To make the conditions of a laboratory process comparable to those involved in the synthesis of new protoplasm the volume of the solvent would have to be increased as fast as the amount of the end-products is increased. As the solids of new protoplasm are formed, they do not stay in the same little parcel of liquid occupied by the old, rather they cause the liquids to expand with them. Hence the volume occupied by the end-products of growth is proportional to their amount, and the concentration of these products, instead of increasing, remains constant. This is a very important difference, for it is on the concentration and not the amount of reagents that reaction velocity depends." Thus the equations of Robertson and SECT. 2] AND WEIGHT 417 Crozier would only be true if the chick embryo, for instance, began as a kind of watery ghost of dimensions equivalent to those it normally has at hatching, and if development consisted in the gradual accumulation of solid substances within it. This singular kind of preformation certainly does not exist in reality. Snell developed another equation, ^ (where A is the initial amount of substrate, x the amount of endproduct at time /, ex the corresponding concentration at time t, Vq the volume of the organism at the beginning of the growthcycle, and iTi and K^ the velocity constants as before), which he regards as the correct form for the representation of an autocatalysed monomolecular reaction in which the conditions are similar to those in a growing animal, or in other words, where the volume occupied by the reagents increases in proportion as the end-product increases. Snell then showed that the curves obtained by this equation did not resemble any of the empirical growthcurves in the literature, and therefore concluded that there is no sound basis for assuming that the master reaction is either monomolecular or autocatalytic. Thus in the case of embryonic growth, where X — is almost constant, the curve approximates to that of a non autocatalysed monomolecular reaction, and again to nothing that is given by any actual embryo. In any case the only instance where a sigmoid curve has been shown to fit the growth of an individual foetal organ is in the work of FaureFremiet & Dragoiu on the lung of the embryo sheep. Robertson's point of view was adopted in the earlier work of Brody and his collaborators, who, however, introduced new viewpoints into it. Instead of plotting the absolute weights of the embryo against time, they plotted the increments per time-unit, thus obtaining curves similar to those in Fig. 34 above. Obviously, in a case where the absolute growth-curve was S-shaped, these increment curves would be doubly S-shaped, rising to a maximum at half-time and thereafter falling away. Or, put mathematically, from the differential equation I =Kx{A- x) NEI 27 4i8 ON INCREASE IN SIZE AND WEIGHT [pt. iii it follows that the velocity of change in an autocatakinetic system progressively rises from zero hour to a maximum value when x = ^A and afterwards constantly falls. When the data are plotted in this way the existence of growth-cycles, whether real or not, comes out much more clearly than when the absolute curves alone are used. It is often said, however, that the increment curve emphasises small fortuitous variations more than the absolute curve, and certainly there are cases, notably the chick embryo itself, where the increment curve shows up cycles which are but poorly shown on the absolute weight curve. Brody & Ragsdale in their first memoir on this subject dealt only with the growth of the cow, and concluded that one complete growth-cycle was accomplished in the foetal condition. A later paper considered the growth of the fowl, for which, on the data of Card & Kirkpatrick for growth after hatching, two cycles appeared, with maxima at 9 and 18 weeks respectively. For the growth of the embryo, the data of Hasselbalch and of Lamson & Edmond, re-arranged by Brody, gave two maxima also, but at slightly different times, thus: Lamson & Edmond 11-5 and 16-5 days Hasselbalch 10-5 and 15-0 days. Brody's figures for these curves are shown in Fig. 48. LeBreton & Schaeffer,who subsequently published a series of chick embryo weights, found maxima at 9 and 15 days, but in Murray's series there is no peak at any time, except a doubtful one on the i6th day, and a plateau between the 1 2th and 15th days. Schmalhausen's data again, when the increments are calculated, show in the case of both series peaks at 10 and 12*5 days, with an additional one, in the case of his 1927 series, at 17 days. In the face of this consensus of evidence, it is not altogether easy to conclude with Murray that "Brody's rhythmic growth curves were due to chance variations". It is true that Lamson & Edmond; Hasselbalch; and LeBreton & Schaeffer used too few embryos in their work, but even in Murray's own work, where about 650 embryos were used, there is an unexplained drop between the 17th and 1 8th days, as well as a plateau between the 12th and i6th days, both quite outside the probable error. Murray did not, it is true, get the i8th-day drop in all his experiments. In Schmalhausen's two series about 400 embryos were used. Brody's chick (daily increment curves) 4 5 6 7 8 9 ion 1213141516171819 20 3 4 5 6 7 8 9 10111213141516171S t t t t Lamson 8c Edmonds data Hasselbalch («) {b) Le Breton &. Schaeffer's chick (daily increment curve) ^ 765432i 4 5 6 7 8 9 10 1112 13141516 17 18 1920 W.Legh. | | (c) Schmalhausen's chick (daily increment curves) 7 N i^ ^ 5 6 7 8 9 10111213141516171819 4 5 6 7 8 9 10111213141516171819 1926 Max.t t 1927 t t t (d) (e) Fig. 48. 27-2 420 ON INCREASE IN SIZE [pt. m It may further be argued that weighing is a very simple process, and it is, therefore, difficult to see why errors should arise which should reflect themselves in these rhythmic curves. A greater degree of scepticism would be justified if they were the results of a complicated estimation method for a chemical substance. But, as it is, these curves form perhaps the best evidence which at present exists for the applicability of the Ostwald-Robertson view to embryonic growth. In the absence of a really exhaustive statistical investigation of the growth of the chick in the egg, these rhythmic curves must be accepted for what they are worth. They would be more convincing if all the workers had found peaks in the same places, but the variation which exists is no argument against the reality of the phenomenon in view of the fact that different breeds of hen were used. Further work is greatly needed to clear up this question. If the peaks on the daily increment curve do turn out to be real, it may be possible to relate them to the peaks of normal mortality which Payne and others have studied, and which will receive further consideration later. (See Fig. 443, Section 18-2.) The autocatalytic curve has also been found by Robertson to fit the data of Stotsenberg already referred to for the growth of albino rat embryos, and a peaked curve is obtained when the daily increments are plotted against time. But, as will be seen, Brody's exponential formula also fits these data, and it is probably right to conclude, as McDowell and his collaborators do, that the figures are not sufficiently good to allow us to distinguish between the two formulae. They cannot be regarded as supporting, therefore, any particular theory of embryonic growth. 2-8. Instantaneous Percentage Growth-rate Brody introduced still another way of representing the facts. He defined the "genetic growth constants" of animals as being the same within each genetically identical group of animals, and as corresponding to specific velocity constants and equilibrium constants in chemical actions in vitro. It may be seen from Fig. 49 taken from Brody's paper that the mature weight of the animal in its life-span. A, is approached by successively decreasing gains in weight after the point of inflection of the sigmoid curve has been passed. The velocity of growth, therefore, declines in a geometrical progression with age. The normal animal reaches, as Brody puts it. SECT. 2] AND WEIGHT 421 under a given set of favourable conditions (much more constant, of course, in egg or uterus than outside), a mature weight which is characteristic of its own species, just as the product of a chemical reaction in vitro reaches under a given set of conditions a definite equilibrium concentration characteristic of its kind. The mature weight A was determined by Brody for a large range of animals Fig. 49 by a graphical method. Now, in this process of geometrical progression in which the increments in unit time are becoming progressively smaller, it is found that in each unit of time the gain made in percentage of the gain made in the previous unit of time is a constant. Thus, in the autostatic growth-phase of the rabbit, for example, the gain is, during each month, 78 per cent, of what it was during the previous month. Brody calculated out this constant, k (simple growth persistency), for a great many animals. It corresponds to the specific velocity constant in chemical equations. 422 ON INCREASE IN SIZE [PT. Ill Finally, B is the difference between the mature weight of the animal and the weight the animal would have had at conception (a minus quantity) if the whole of growth was representable by the curve for the autostatic or self-inhibiting phase. This genetic growth-constant also was found for many animals by Brody. From Fig. 50 it can be seen that, the higher the value of k, the more rapidly the mature value is approached (pigeon > mouse > rat > guinea-pig > sheep > pig > cow > man), and that the fact can be equally well accounted for on the assumption that a substance Yrs.cS^ Age (from blrbh) man 7 8 9 10 11 12 Mos.a' 60 64 68 72 76 12 16 20 24 28 32 36 40 44 48 52 66 Age (from conception) of animals OoJ Fig. 50. is used up during growth, or on the assumption — perhaps more likely in view of the work of Carrel and his collaborators on tissue culture (reviewed by Pearl) — that during growth a growth-retarding substance is produced according to the law of monomolecular change. The paper of Brody, Sparrow & Kibler was concerned with age equivalence. They showed that, with the aid of the formula previously established by Brody, W=^ A- Be-""^ (where W is the weight at age t, A a. genetic growth-constant, the mature weight, B another genetic growth-constant which increases in value with increase in length of the processes preceding the point SECT. 2] AND WEIGHT 423 of inflection, and k a third genetic growth-constant, the fractional decHne in the velocity of growth), it was possible to plot growthcurves for all animals to the same base, and so to determine their age-equivalence. Thus they found that i rat month was equivalent to ii-gi cow months, and i guinea-pig gram to 509-1 cow grams. They finally constructed a table (Fig. 51), in which the value of ^ was given for a great many animals, and a logarithmic graph, from 001 .02 Value of k. .03 .M .05 06 -07 .C8 -09.1 Z 5 .6 7 .8 .9 10 CotVC. Mos; Fig. 51 which can be read off the time in months (conception age) at which the animal with the constant k in question will arrive at 10, 20, 30 or 90 per cent, of its mature weight. Brody next considered the growth-constants during the autokinetic or self-accelerating phase of growth. He subjected the methods which had previously been used to represent growth to severe criticism, part of which has already been referred to. Thus, in the case of Minot's method, increments of growth are regarded as being added on discontinuously at the terminal points of arbitrary time periods, whereas growth is really a continuous process, and I dt_ \ w =-k 424 ON INCREASE IN SIZE AND WEIGHT [pt. iii the mathematical expression for it must take account of its smooth nature. Brody found the relationship between the relative rates of growth k ^j^ W to be k = log [R + i), where R is Minot's percentage growth-rate. The latter can, therefore, only be used when it does not exceed 10 per cent, for the period under consideration, i.e. never in embryological work. Brody also criticised the methods of Pearl, whose equation dW _ k dt t- a does not take account of the fact that part of growth is self-accelerating, and also Pearl & Reed's modification of the original Robertson equation. He maintained that the best way was to plot the log. weight against the age, when, if the result is a straight line, the rate of instantaneous growth, A;, must be a constant. This will not of course be confused with the fractional persistency constant referred to above, for the latter only refers to the autostatic or self-inhibitory phase. The autokinetic or self-acceleratory phase is ^ clearly the more important ^ and interesting for embryologists. Figs. 52 a, b, taken from Brody's paper, show the data of various workers for the wet weight of chick embryos treated in the way described, namely, the log. weight plotted against the age. It will be noted that a series of straight lines result, forming a system concave to the abscissa and rising rapidly but more quickly at first than later. The curve is thus the exact opposite of the Minot Cms ,.:?

  • «jy

6 10 12 14 16 Ifl 20 Incubation Age Fig. 52 a. Gms. 20 10 </^ o I -H vy I I ^ ^ /. \ 6 5 3 — = = E —^ ? V % h — — — — — 1^ — / 1 z t — — . A /. / ?i // ^ ( J // [■ / / V — — — ^ .a ~ 1 — i^ — 1 — / K/ i — — — — — — ' — — y. / j 5 'if / / / "l ^^ / J \ .3 .2 ! ^^ A / i ^ — ^ ^ — — z — 1 / / — r i— — — — — — — — / J r^' H as SQl Da ch 1 /A B— rtr: r* Days 2 4 6 6 10 12 14 16 Incubation Age IS 20 Gms., 22 20 19 16 14 12 10 8 6 4 2 10 12 14 16 15 20 Incubation Ag<^ Fig. 52 b. 426 ON INCREASE IN SIZE [pt. iii curve, which falls so markedly during the same time. Thus the growth-rate expressed in this manner also falls off as time goes on, or rather rises less and less rapidly, becoming eventually asymptotic to the mature value. This curve is undoubtedly a great improvement on Minot's, for it involves no arbitrary time period and depends on the differential calculus, which has as its special province the evaluation of instantaneous change. An infinitesimally small period dt, and an infinitesimally small increase of weight dW, are the basis of its operation. Then all the infinitesimal differences can be added together (i.e. integrated). Thus: .j^ "^ _ becomes, if the number of dfs, and dW's is infinitely large [n) , when integrated, W = Ae^*, for I + ^y = e^\ A being the weight at the beginning of the whole period. If this is turned into logarithms log W =\og A ^ kt, , _\ogW - log A or k , and, as where growth is being considered A is, to all intents and purposes, o, ^ ^ k = -^ — . t The instantaneous relative growth-rate for a unit time is the sum of all the instantaneous rates during the given unit of time, and may therefore be multiplied or divided, according to the time-unit in which it is desired to express it. The log. weight/age graph is, therefore, a measure of the instantaneous growth-rate, and the value of the constant k which can be calculated from the last equation will give the slope of the straight, or approximately straight, line. The log. weight/age graph could, of course, have been plotted by Minot, but Brody's use of the differential calculus was required to show that the slope of the curve gave an instantaneous growth-constant. Thus, in the example given, Lamson & Edmond's data, the constant is 56 from the 5th to the 8th day of SECT. 2] AND WEIGHT 427 development (very steep slope), 36 from the 8th to the 13th day (less steep slope), 24 from the 13th to the i8th day (still less steep), and 25 from then onwards. The higher numerically the constant k, the steeper the slope, and consequently the greater the instantaneous growth-rate. In Figs. 53 a, b and Table 55 are shown most of the weights and processes in the hen's egg whose constants have been calculated by Brody. Each system grows at a rate peculiar to itself. Murray, as we have seen, also plotted log. weight against age, but he did not get a straight-line relationship ; on the contrary, the resulting curve was concave to the age (abscissa) . McDowell, again, got a similar concave curve for the pre-natal growth of the mouse, and there is much point in his criticism of Brody 's work: "Brody draws a series of straight lines through corresponding exponential curves and concludes that growth-rate does not decline continuously but by abrupt drops between periods of uniform rate. Since any curve can be approximated by a series of straight lines, the critical significance, both of the specific number of straight lines, and of his general conclusions, seems somewhat questionable"*. Table 55 includes also a column in which the time taken for the embryo or a corresponding entity to double its weight or amount is shown. For, when the instantaneous percentage growth-rate is constant, the time intervals between doubling of weights are constant; therefore, from the expression W - Ae^\ at a certain time logs _ 0-695 k ~ k ' and, as k is found to be for the rat embryo 0-53 or 53 per cent., the time required for it to double its weight must be — ^ or 1-3 days. ^ 0-53 Further, if growth in weight can be taken as a measure of the increase in the population of cells in the body, a new cell-generation is produced every 1-3 days on an average, and the cell-division frequency is 1/1-3, i-^- 0-77 times per day. It is thus possible to determine, as Brody says, the mean life of a mother cell before it divides into two daughter cells.

  • Nevertheless, McDowell himself admits a discontinuity between pre-axial and axial

growth, as we have seen on pp. 394 and 396. n c i 55 Mom s.Gms 3.0 20 20 10 1.0 a .8 5 .5 3 .3 2 .2 1 .1 .001 .01 DaysO 6 10 12 14 16 18 20 Incubation A^ Fig. 53«' ^1 8 S88 S '^'^^ ^ ^ ':ind:^no'oO 430 ON INCREASE IN SIZE [PT. Ill Table 55. Instantaneous growth-rate {k). Brody's figures. Time in Growth which the Time rate % entity is Entity in question (days) per day {k) doubled {d) Investigator Chick Wet weight . 5-8 56 1-2 Lamson & Edmond ,, 8-13 36 1-9 JJ 5> 13-18 24 29 5> 55 6-10 56 1-2 Hasselbalch 10-14 29 2-4 55 14-19 19 3-6 35 6-10 47 1-5 Murray 10-14 33 2-1 35 14-19 21 3-1 33 CO2 excretion 0-4 98 07 Atwood & Weakley >> 4-14 31 2-2 35 35 5> 14-19 Pause — 33 33 3) 19-21 31 — 33 55 0-16 36 I '9 Hasselbalch 5) 0-16 32 2-2 Murray Urea excretion 5-7 76 — Needham ,, 7-14 34 — 53 Glutathione content 6-9 54 — Mvirray 5) 9-15 30 — 55 Total CO2 content 7-9 56 — 55 ,, 9-16 37 — • 35 Chloride content 12-15 32 — ,j Calcium content 12-16 85 — Plimmer & Lowndes Creatine content 14-21 28 — Mellanby Nitrogen content 6-9 60 — Murray jj 10-15 47 — ,, 15-20 23 — Total solid content 5-10 57 — ,, 10-16 46 — Ash content ... 10-14 43 — ,, 14-19 25 — Calorific value 7-9 56 — ,, 10-15 47 — Rat Wet weight ... 13-22 53 1-3 Stotsenberg Guinea-pig Wet weight ... 17-20 100 0-7 Ibsen & Ibsen; ,, 20-35 25 2-8 Draper; Hensen ,, 35-52 9 7-8 S3 55 ,, 52-70 5 15-1 33 33 Man Wet weight ... 60-110 8 8-7 Streeter ,, 110-160 Not straight 55 ,, 160-240 1-7 41-0 33 , 240-280 1-3 55-0 „ Brody himself did not omit to make suggestions as to possible correlations between his abrupt breaks in growth-rate and other phenomena known to be taking place during the embryonic development of the chick. In the first place, he associated the breaks in the growth SECT. 2] AND WEIGHT 431 rates of carbon dioxide production at the 14th day with the change in mode of respiration from aquatic to terrestrial which takes place late in incubation. This is quite a convincing correlation, but his suggestion that the first break (at four days), before which the instantaneous growth-rate is about 100 per cent., and the Minot growth-rate 1000 per cent., is associated with a general critical period occurring at that time is not really so satisfactory. For almost any process has its critical moments during development — for example, the peak in protein metabolism at 8-5 days. In cases where there is no a priori reason for assuming correlations except the one fact that their peaks coincide or are converse to each other, the utmost caution should be used in so correlating them. Wholesale correlations of apparently unrelated phenomena may be chemically misleading. Thus Brody cites Tomita's peak in total lactic acid content at the 5th day (see Fig. 292) as evidence of a critical period corresponding to the abrupt break in his growth-rates of carbon dioxide production and to the peak in Payne's mortality curve (see Fig. 443). Brody is not the only investigator who has occupied himself with the growth-rates of different chemical processes and amounts in the embryo, but, before passing on to discuss these points, which will lead naturally to the question of the growth-rates of parts of embryos, a further word must be said about Brody's work. At present it is not possible to tell much from the comparison of embryos of different kinds, though it is obvious that an immense field of research is opened up here for the comparative embryologist of the future. Thus the equation for the development of the chick embryo in weight according to Murray is W^=o-668^^^, corresponding to instantaneous growth-constants of 0-47, 0-33, and 0-2 1 successively,* while the equation for Stotsenberg's rat embryo figures, according to Brody, is W^ = 0-000065^°^^*, corresponding to a steady rate of 53 per cent, per day instantaneous. On the steadiness of this rate Brody says, "If there is no fallacy in this reasoning we have reached a new and an extremely important conclusion. While all investigators of the time relations of growth have reached the conclusion that the percentage growth-rate continuously and rapidly declines with age, our conclusion is that the instantaneous percentage growth-rate remains constant for the relatively enormously

  • A later value, due to Vladimirov & Danilina, is W=o-'^2^fi'^.

432 ON INCREASE IN SIZE [pt. iii long period between 14 days and birth. The cause of this difference in results is due to the fallacy in the method of analysis employed by Minot". This is only true, subject to confirmation of the fact that the foetal log. weight/age graph gives a straight line over definite periods, and this is just what is not certain. Decision on the matter cannot yet be made. Another interesting point which emerges from Table 55 is the long embryonic stage in the guinea-pig. The chick hatches when its k is about 0-21 and the rat is born when its k is even higher— 0-53, but the guinea-pig stays inside the uterus until its instantaneous percentage growth-rate has dropped to 0-05. Brody succeeded, indeed, in raising guinea-pigs by feeding them on hay and grain immediately after birth, so that they tasted neither colostrum nor milk. It is interesting, again, to note that there is only one break in the instantaneous growth-rate of carbon dioxide production, whereas there are at least two in the instantaneous growth-rate of wet weight. This must mean either that the respiratory function develops at a rate quite independent of the growth in mass, or that the weight of the body cannot be taken as an index of the growth of the metabolising tissues. This point will be referred to again, for it is of much importance in chemical embryology. On the other hand, the respiration k does show a break about the 17th or i8th day, which is duplicated in the wet weight k, or, at any rate, in the log. weight curve constructed from Lamson & Edmond's data — for it is not so apparent in those of Hasselbalch and of Murray. This may be associated, as we have seen already is Brody's suggestion, with the change in form of respiration occurring then (chorio-allantoic to pulmonary). There is no doubt that some obscure events are associated with this late stage in the chick, e.g. the mortality peak of Payne, which can be greatly intensified if a certain lethal gene is present, and the sudden immunity to implanted rat sarcomata (Murphy), which the chick then acquires. Brody suggests that the chick embryo passes at this stage through a "metamorphosis" similar to those hidden ones which exist, according to Davenport, in the development of man. The extremely small values of k for the embryonic period of man are worth attention. The human embryo grows a great deal more slowly than any other. Five months after conception the instantaneous percentage growth-rate is only 1-7 per day, while, during the week preceding birth, the rat embryo grows at the rate of 53 per SECT. 2] AND WEIGHT 433 ■ whole embryo (Murray and Needham) Bcalorlfic value (Murray) Ddry solid (Murray) Ocarbohydrate (Needham) ® protein (Murray and Needham) e fat (Murray) cent, per day. The lowest rate of growth ever reached by the rat after birth is 3 per cent, per day. Given percentage rates of growth, therefore, do not signify equivalent stages of development irrespective of the species of animal. Calculation of the rates of growth for various processes and individual components in the development of the embryo has also been done by other investigators using the Minot method. In 1927 I calculated the percentage growth-rate for the total carbohydrate content of the chick embryo; it fell from 56 to 22 per cent. In Fig. 54 is shown the fall in the Minot curves for the wet weight of the whole embryo, the calorific value, the dry weight of the whole embryo, the sugar, protein, and fat content of the embryo. All of ^° them fall, but we have here an instance of the limited but real value of the Minot curves, which, although no absolute conclusion can be drawn from them, do show that the tissue constituents and the dry weight have a different behaviour from the wet weight. It can easily be seen that they form a plateau between the loth and the 15th day, during which they grow at a constant rate while the wet weight is falling all the time. This plateau also appears on the curves for carbon dioxide output calculated in the same manner as percentage growthrates from the figures of Bohr & Hasselbalch; Atwood & Weakley; and Murray in 1927 by Brody. The plateau must be due to the fact that the growth of dry substance is specially rapid during the middle phase of development; it is then that the embryo makes the most rapid strides from wetness to dryness. It is interesting to see that the growth-rate of carbohydrate is never as high as some of the others, and never drops so low. It is significant, moreover, that on the 1 9th day the Minot growth-rate of the protein has dropped below that of the whole body, while the growth-rate of fat remains well above it. This is an illustration of the "relative" use of Minot's method. Fig- 54 88 434 ON INCREASE IN SIZE [PT. Ill 2-9. Growth Constants Brody is not the only worker who has applied the differential calculus to embryonic growth-curves. Teissier and Lambert & Teissier suggested simultaneously that this should be done, but their work was quite theoretical. However, Schmalhausen published independently at almost exactly the same time a paper in which it actually was done. He criticised Minot's method of calculating Per cent per ^"y k=.53 50 ^ k- ^ \nW2 IntVi 5 40 _ ^ V. ^ i. 0,30 — ct JO J3 (0 3 t. 0. 0> I JS20 C c 'i3 Q) w U JP Q. k=.11 c ID — c. k=-047 U=-031 ^ .^^ Rat,?, — 1 1 h unmated 1 1 >== l_ 1 1 1 1 1 1 1 1 1 Dayso 10 20 30 40 50 60 70 so 90 100 no 120 iso uo i50 16O i70 18O c o O 8 CO 18 28 38 58 68 78 Age Fig. 55 a 108 118. 128 138 U8 158 " mittlerer prozentualer Zuwachs " in arbitrary time-units from exactly the same point of view as Brody. Thus, he says of the Minot method: "the larger time intervals we take, the bigger the error will be. With equal time intervals, the error will be bigger the bigger the rapidity of growth, and this will in fact lead to altogether misleading figures for the early periods". We need not follow Schmalhausen's reasoning, which led him to adopt the calculus as a better assistance in studying growth-curves, for we have already examined and approved the SECT. 2] AND WEIGHT 435 C JO J3 a 10 ll f \ / \ ^ / / 1 \ \


^ c r c=— ,^ arguments of Brody. Schmalhausen speaks of the "wahre Wachstumsgeschwindigkeit " instead of the instantaneous percentage growth-rate, and of r instead of k. His equation relating the instantaneous percentage growth-rate to the Minot growth-rate is exactly the same as Brody's, and he points out that, when, according to the old method, the growth-rate would be 700, the instantaneous method would give a result of 207, though 50 per cent. (Minot) would be equivalent to 40-5 (instantaneous). These figures might have been read off from the graph of relation given by Brody, and it is surprising that the two workers, one at Kiev and the other in Missouri, should have been thinking on such very similar lines. It is still more surprising that embryologists had not thought on such lines long before. Schmalhausen and Brody diverge, however, upon one important point, namely, the shape of the line given when the log. weight is plotted against the age, for Schmalhausen regards it as a curve — just as Murray and McDowell do — while, as we have seen, Brody lays great stress on the representation of it by a series of straight lines having abrupt breaks between them. Thus, the instantaneous percentage growth-rate, which with Brody remains constant over certain definite periods, with Schmalhausen continually declines in value. In other words, Brody's diagram which shows the instantaneous percentage growth-rate dropping in a stepped formation from fertilisation to hatching is replaced in Schmalhausen's work by a regular curve passing downwards to become asymptotic to the abscissa (Fig. 55 h), just as the Minot curve does, only, of course, plotted from a set of figures having a real meaning as against the abstractions of Minot. This difference of outlook leads naturally to very wide differences in conclusions; thus Schmalhausen has nothing to say about critical points or hidden metamorphoses. Having diverged from Brody in this direction, 28-2 10 n n 16 20 Zl Age in days P=wet weight; Cv = instantaneous % growth-rate. Fig. 55 *• 436 ON INCREASE IN SIZE [PT. Ill he proceeded a good deal further along it by observing that the graph relating instantaneous percentage growth-rate to age was practically identical with a rectangular hyperbola, and that there was a simple relation between the values of r or Cv (Brody's K) and the age, for the product of the two was always roughly equal to 300.* Table 56. Embryonic growth: Schmalhauseri's '■^Wachstumskonstante" {^^wahre WachstumsgeschwindigkeW^ x time). Cvl Cut length age Investigator Man Whole embryo wet weight 193 369 Friedenthal Mouse 95 5> — 337-5 McDowell et al. Rat 35 J5 441 518 Stotsenberg Chick J> » — 318-5 Murray 35 55 518 321 Schmalhausen Liver 329 Lung ... — 321 Fore limb (whole period) — 293 (3-18 days) ... — 329 Hind limb (whole period) — 347 (3-18 days) ... — 395 Stomach ... — ■ 374 Brain — 210 Lens — 210 Whole eye (2-10 days) ... — 317 ,, (i 1-21 days) ... — 99 Heart — 276 Mesonephros (4-13 days) — 224-5 Metanephros (7-17 days) — 359 (17-21 days) — 196 Ovary — 145 Guinea-pig Whole embryo wet weight — 347 Draper Trout 55 55 (30-51 days) — 206 Kronfeld & ScJ (51-99 days) — 207 S3 Schmalhausen gives no explanation of the breaks in the cases of those organs which have two values of Cvt, but calls attention to the fact that the organs of early differentiation have low Cvt and vice versa. If the curve obtained by plotting Cv (Brody's k) against time is a regular hyperbola, then the product Cvt should be 300. If it exceeds this figure, the curve is descending and becoming asymptotic less rapidly, i.e. the rate of growth (instantaneous) is not falling off as rapidly as it will be if the product is less than 300 at any given moment. This constant he calls the "Wachstumskonstante", and its values, calculated by him for a number of embryonic processes, are seen in Table 56. It is perhaps the least convincing part of his exposition, for when during a certain series, e.g. the growth of the human embryo, the constant Cvt oscillates between 899 and 93 as extreme limits, one may legitimately doubt whether great stress can be laid

  • Brody himself does not find this to be so.

SECT. 2] AND WEIGHT 437 on the average. Moreover, Janisch treats the same curve as a catenary exponential one. And, ahhough the " Wachstumskonstante " for the various organs and parts of the embryo show differences which might well be regarded as characteristic for the tissue in question, it is disturbing to find so wide a difference from the predicted value in the case of the rat embryo, explained though it is by Schmalhausen as due to variable factors in the food of the maternal organism. The reason why 300 is the number to which these figures approach is, of course, because, according to Schmalhausen's formula, the increase of the embryonic weight can be expressed by the equation W= k [atf, where W is the weight, a the "Lineargrosse", t the time and k a constant. This agrees with the hyperbolic nature of the Cvjt curve. Table 57. Instantaneous percentage growth-rate [Chick). Schmalhausen Day of Brody j^ development (Smoothed) (Raw) (Smoothed) O-I 1-2 2-3 — — — — 190 190 3-4 — 119 140 4-5 — 139 107 5-6 — 83 87 H 47 79 70 7-8 47 38 60 8-9 47 36 50 9-10 33 53 '^2 lO-II 33 22 38 11-12 33 ^\ 33 12-13 33 48 30 13-14 33 30 27 14-15 21 20 25 15-16 21 32 23 16-17 21 23 21 17-18 21 22 19 18-19 21 II 17 19-20 — ^l 15 20-21 — 16 12 for the equation of an equal-sided hyperbola is j; = 3/x. Schmalhausen does not derive his Cv directly, but calculates it in each case from the Minot percentage growth-rate figures. It is instructive to place side by side the instantaneous percentage growth-rates of Brody and Schmalhausen for the chick embryo, as is done in Table 57. That of the former has three constant periods, that of the latter 438 ON INCREASE IN SIZE [pt. m shows a gradual decline, and the figures illustrate what has already been said, namely, that, until we possess much better statistical data than is actually the case, we cannot differentiate between the Brody position and the Murray-McDowell-Schmalhausen position.* As the matter is fundamental in view of the important theoretical issues involved, the accumulation of more data is urgently to be desired. It may be mentioned that Cohn & Murray, plotting log, weight/age curves for the growth of embryonic heart cells in tissue culture, obtained curves concave to the age axis and not straight lines. Schmalhausen also studied the growth in length of the chick embryo, calculating it from the weight by the formula L = VW. The daily gain in length a he found to be variable around a constant value of 1-47 for the first half of development and 2-00 for the second half. But when the weights were corrected by the estimation of the embryo's specific gravity (average for first half 1-025, average for second half i -06) the corresponding daily gains in length worked out at I -Go and 1-79. A further correction made necessary by the presence of the feathers during the last half of development brought the figure down to 1-64, so that throughout incubation the embryo apparently grows in length at the same average rate. The duck embryo, according to Schmalhausen, has a daily length increment of i*io mm., and this value he regards as constant for the species. He went on to calculate a for the human embryo, using the weight data of Friedenthal and Zangemeister, and for the embryo of the white rat, using the data of Stotsenberg. All these results are shown in Table 58, together with his further assessments of a calculated from the " Normal tafeln" of Minot & Taylor for the rabbit and Keibel for the pig. The daily size ("Lineargrosse") increments of his own measurements of separate organs and parts of the chick embryo are also given. During the course of development the value of a rises and falls according to the rate of growth. If the value a/ 1 is calculated, where a is the daily increment in length and / the length of the part in question at the beginning of the period, the absolute size of the part will cease to affect the result, and the organs will be comparable with themselves and with the whole embryo. When this is done, the ratio is found to be fairly constant, rising as high as 10-3 per cent, for the stomach and falling as low as 6-3 per cent, for the lens. These figures are also given in Table 58. Schmalhausen concluded from

  • Recent work by Byerly supports that of Brody.

SECT. 2] AND WEIGHT 439 them that organs which reach a high state of differentiation early grow the most slowly (brain and lens), while less differentiated organs grow most quickly (liver and limb-buds). The growth of the body as a whole is the average practically exactly of the rest, and it is interesting to note that the organ which most nearly approaches it is the heart. The heart would seem to grow in size at the same rate as the entire body. But, as Schmalhausen says, this growth in size seems to have no simple relation to the growth in weight as shown by the percentage growth-rate. Table 58. SchmalhauserC s values for a, i.e. daily increment in size or '^ Lineargrdsse^\ Chick Duck Man )> Rat Rabbit Pig Guinea-pig Whole embryo (ist half) (2nd half) Brain Lens Spleen Heart Lung Liver Testis Metanephros Stomach Fore limb Hind limb Pectoral muscles Whole embryo Millimetres A r a a/l Investigator 1-6 Schmalhausen 1-64 7-65 05 6-50 o-i 6-30 o-i 6-62 03 7-50 0-29 7-82 0-55 883 0-09 8-82 0-32 ID- 10 0-64 10-30 0-44 7-33 0-76 885 0-43 909 I-IO — 0"55 — Friedenthal 0-55 — Zangemeister 1-47 — Stotsenberg 1-2 — Minot & Taylor i-i8 — Keibel 0-75 — Read In a later paper Schmalhausen studied the relation between initial weight and end weight in a number of animals, wishing to obtain some means of comparing their " Wachstumsertrage " or Growthyields, on a basis independent of their size. He found that u, or the mass of substance added on to itself by the organism between times ti and tz, could be calculated by the formula u = t h *^i 0-4343 u u post u embryonic embryonic whole period period life-span 2-2 17-2 19-4 1-8 15-3 I7-I 9-5 3-8 13-3 8-9 2-3 II-2 13-6 4-3 17-9 13-6 2-3 15-9 15-6 4-8 20-5 20-9 3-3 24-3 440 ON INCREASE IN SIZE [pt. iii the initial weight being taken as unity {k = Cvt). His results for various organisms show interesting differences, thus : Sturgeon Pike Hen Mouse Rat Guinea-pig ... Pig Man Here we observe the effect of early hatching in the two aquatic forms, which have the greater part of their growth still before them at the time of leaving the tgg. The other figures demonstrate quantitatively what is apparent to common sense, namely, that the embryonic period is the time of greatest growth in terrestrial animals. 2-10. The Growth of Parts We must now turn to the relative growth-rates of parts of the embryonic organism. This is a field which has mainly been tilled by anatomists, but it is of the greatest importance to the chemical embryologist. For the increasing and decreasing intensities of physicochemical processes cannot be intelligently studied in the absence of a knowledge of the distribution of the whole mass among the different organs and tissues. The investigation of the relative growths of endocrine glands, again, cannot but throw much light on the development of the adult metabolism in the embryo. D'Arcy Thompson sees an appreciation of this in the eighteenthcentury preformationists. "It was the apparently unlimited extent", he says, "to which, in the development of the chick embryo, inequalities of growth could and did produce changes of form and changes of anatomical structure that led Haller to surmise that the process was actually without limits and that all development was but an unfolding, an 'evolutio' in which no part came into being which had not essentially existed before. In short the celebrated doctrine of preformation implied on the one hand a clear recognition of what, throughout the later stages of development, growth can do, by hastening the increase in size in one part, hindering that of another, changing their relative magnitudes and positions, and altering their forms ; while on the other hand, it betrayed a failure — SECT. 2] AND WEIGHT 441 inevitable in those days — to recognise the essential difference between these movements of masses and the molecular processes which precede and accompany them and which are characteristic of another order of magnitude." The papers of Schmalhausen are of much importance in this matter. Inspired by the views of His, who declared in 1874 that all the development of shape could be ascribed to unequal growth in various component parts of the embryo, he set himself to weigh and measure a great number of these individual sections. He first studied the relative growth-rates of the brain and eye of the chick embryo, together with the liver, lung and stomach, representing the organs of endodermal origin. In each case, he calculated the % growth-rates and the percentages formed of the weight of the whole body. For the organs of mesodermal origin, he chose the heart, the mesonephros, the metanephros, the ovary, and testis. These figures he treated in the same way. In many cases his weights were not obtained directly but by reconstructing from serial sections and then weighing, proper allowance being made for complicating factors such as specific gravity. Fig. 56 shows one of his graphs — it is specially interesting as showing the definite decrease in weight which the mesonephros undergoes after the 15th day in giving place to the metanephros or adult kidney. It also includes % growth-rate curves for the fore and hind limbs. Lastly, he ascertained the growth-rate of the feathers. In general, he found that the changes in the growth-rates of organs were synchronous. The percentage growth-rate (see Fig, 57) seemed to have peaks in its descent, each one less marked than the preceding one. In each case, the growth-rate of every organ shows a certain rise, but the amount of the rise differs in different cases — thus the lung is the organ which is growing fastest about the 6th day, the hind extremity about the loth day and the stomach about the 13th. On the whole, the periods of depression of the growth-rate of the majority of organs are from 7 to 9 days, from io| to ii| days, and from 14 to 16 days. When the weights of individual organs, however, were arranged plotted against weight of embryo, not age, the peaks disappeared, as would be expected, for the total weight is the sum of the weights of the organs. It would be interesting to plot the logs, of Schmalhausen's organ-weights against age in order to obtain the instantaneous growth-constants of Brody for each one. Schmal 442 ON INCREASE IN SIZE [PT. Ill hausen's general results, however, were as follows: on the 5th and 6th days, the growth-rate of all organs is falling, with the possible exception of liver and hind limb. This continues till the beginning of the 7th day, save that the eye and the lens may show a slight rise. At the beginning of the 7th day, however, the growth-rates of all organs rise, firstly the mesonephros, the liver and the lung, and, to a ^17 S. 16 x5 10 1.4 o^l3 o\i2 <n clO ^ 9 c m 8 2 7 (0 5 X3 E (E) = Fore limb G = Gonad M = Mesonephros Abs. web wei ghb 12 3 4 5 6? 9 10 11 12 13 14 15 16 17 18 19 20 Days Fig. 56. less extent, the hind limbs, then these are followed by the rest. After a peak all fall until the 1 1 th day, when all again rise, only to fall on the 14th, with the exception of the feathers, which maintain a rise. Later a more gradual rise in growth-rate takes place throughout the body, begun by the lens and liver and, to a less degree, by the eye and the brain, and continued by the stomach and the mesonephros. After the subsequent fall, only small variations take place, which are found to be synchronous for groups of organs such as kidney-liver SECT. 2] AND WEIGHT 443 stomach until the end of the embryonic period. It is certainly interesting that organs so different in origin and nature as the eye and the mesonephros should be similarly affected by spurts of growth at various stages, and Schmalhausen concluded that this was an argument against the hypothesis of specific organ-stimulating substances, the presence of which would from time to time cause more 1 Z 3 V 5 6 7 a 3 10'^ 1Z ^3 1'^ -75 16 17 13 13 20 Zi Days Fig. 57. C = brain; £; = hind limb; A" = whole body; Z-t = liver; Li = lens; Af=mesonephros; Oc = eye; P = lung; Af /n = metanephros ; 5^ = stomach; G= gonads. rapid growth in one place of the embryo than in another. It looks much more as if growth-promoting substances were passed round in the embryonic circulation at certain definite intervals, and so exercised an effect on a large number of different organs. In this connection, the recent work on the growth-promoting substances of egg-yolk should be borne in mind, and the experimentally determined cycles of varying permeability to fat-soluble and watersoluble substances in the walls of the vitelline blood-vessels. One relation which seems clear from Schmalhausen's work is that growth 444 ON INCREASE IN SIZE [pt. iii of fore and hind limbs does not accomplish itself in the same spurts as the viscera do, for during the 8th day depression in the growthrates of the latter the skeleton is growing vigorously, and during the loth day peak it rather falls off. Very similar remarks apply to the hind limb growth-rate. Schmalhausen concluded that very young organs can respond to a given intra-embryonic environment by increase in growth-intensity, while more differentiated organs can at the same time respond by depressions in their growth-rates. "If one and the same influence", he says, "can act in a stimulatory manner on the growth of some parts or organs, and inhibitorily on the growth of others, we can see how unequal growth can take place and hence a change in form." It would also appear that the more development goes on, the more different the rates of growth of different organs are. Three factors seem to control the growth-rate of a single organ: (i) the age of the embryo, (2) its own degree of differentiation, and (3) growth-promoting substances or embryonic hormones present in the circulation. Under (2) would be included the time of origin of its "anlage" and the intensity with which its preliminary growth would take place. These views are not compatible with Mehnert's "laws of organogenesis", the main one of which was that the growthrate of an organ in the embryo was proportional to its degree of development at the time in question. The only criticism that can be levelled against Schmalhausen's work is that the number of embryos employed was perhaps rather few. In conjunction with Stepanova, Schmalhausen made further investigations on the growth of the embryonic skeleton in the chick. Similar fluctuations in pre-natal growth-rates of parts have been discussed as regards the primates by Schultz. Schmalhausen has attempted to give an explanation of these spurts in terms of metabolism. Summarised again there are, in the case of the chick, three or four periods, in each of which the growthrate first rises and then falls, as follows: Days 0—4 1st period, great fall from a high value 4-9 2nd period, rising to the 6th day then falling 9-12 3rd period, rising to the loth day then falling 12-15 4th period, rising to the 13th day then falling 15—21 5th period, rising to the 17th day then falling He has suggested that these periods may partly correspond to the periods which can be distinguished in the development of the chick SECT. 2] AND WEIGHT 445 embryo, during which one type of chemical molecule is predominantly burned to furnish energy for the growing organism. This subject will be handled fully later (Sections 6-8, 7-7, and 9-5) ; here it suffices to say that the beginning of development is in many ways closely associated with an important carbohydrate metabolism, and the latter part with the metabolism of fatty acids, while the intermediate part would appear to have an association with catabolism of protein, in view of the fact that the point of maximum protein catabolism occurs when 8*5 days of development have been completed. These periods, in Schmalhausen's view, may be identified with those in which he finds spurts in the growth-rate. A certain amount of scepticism about this identification would seem justifiable until we have irrefragable proof that the spurts are more than chance variations in a curve composed of too few data. His suggestions involve the view that "Abbauprodukte" accumulate from time to time in the developing embryo, and so hinder its growth (essentially the same theory as those of Jickeli and of Montgomery) . Thus his first depression of the growth on the 4th day corresponds to an accumulation of lactic acid and ammonia (see further on for the detailed references) and his second depression of growth on the gth day corresponds to an accumulation of urea. Finally, his third depression of growth about the 12th day corresponds to an accumulation of uric acid. He admits that there is nothing chemical which obviously coincides with the later depressions of growth, but supposes that they depend on the decreasing excretory power of the mesonephros. After the i6th day the metanephros would be undertaking the duty of excreting waste products, and growth accordingly begins again. Ingenious as these correlations are, they cannot be said to be convincing, in view of the fact that many other processes besides the excretion of waste products may be supposed to be exercising an eflfect on the growth-rate. More interesting is Schmalhausen's attribution of great importance to the surface of the blastoderm, the blastodermal capillaries, and the active surface of the excretory organs. Measurement of these during the course of development would throw a bright light on these problems. Schmalhausen did himself compare the growth in weight of the embryonic kidneys with the daily increment of the whole body, and, although the figures were rather erratic, he felt able to conclude that, owing to the slow growth of the mesonephros and metanephros, the excretory surface 446 ON INCREASE IN SIZE [pt. iii was only just keeping pace with the growth of the embryo. In these circumstances, it was not surprising to find now an accumulation and now a flushing out of waste products from the embryonic body. In a later paper, however, he modified considerably his views on this subject. As regards the growth of individual parts, Schmalhausen later introduced several further expressions. "Homonomic growth", in his terminology, means growth of an organism in which all the parts and organs have the same growth-constant, "heteronomic growth" — the more usual form — is the growth of an organism composed of organs each with its own characteristic growth-constant. Then the growth-directing force may exist either inside or outside the anlages of the separate organs — in the chick it apparently does not exist inside — and in the former case it would be called "autonomic growth", in the latter " automorphic ", while, if the influence was directly the growth of another organ, it would be termed "heteromorphic". Schmalhausen found that, although the organs in the chick embryo taken at any one moment had very different rates of growth, yet, if they were all dated, as it were, from the time of formation of their anlages, they showed very similar rates of growth. Thus an anlage developing late would be growing much quicker than the whole body, while, at the same time, if its instantaneous percentage growth-rate curve was plotted, it would be found to have a shape very like that of the organism as a whole. Thus organs can only be compared as to their growth-rates if they are taken from their own particular origins and not from the origin of the body as a whole. In homonomic growth, of course, one is dealing with organs originating at the same time and having identical growth-constants. In this case, the definite proportions of the resulting organism can be deduced from those of the anlages ; in other words, a kind of preformation holds good. If Table 56 be again referred to, it will be seen that, on the whole, it takes functioning organs longer to grow than functionless ones. Thus the metanephros, which at first has a Cvt of 359, drops to 196 after it has begun to excrete actively by about the i6th day. In Fig. 58 is shown the relation between the weights of the organs in the embryo chick expressed as percentages of the total weight of the body. The heart and mesonephros are seen to have their maximal relative size very early in development, after which the former declines slowly and the latter more rapidly. The first four days of SECT. 2] AND WEIGHT 447 development see also the maximal relative size of brain and lens, but these fall very rapidly away from their pre-eminence. Towards the end of the developmental period, the fore limb gains much in importance, and about that time also the metanephros reaches a maximal point of growth. For further comparison further calculations are necessary. The relative instantaneous percentage growth-rate could be obtained from the equation ^ , 1 ^ (^v _ log v-^ — log V Cw log Wi — log w ' 1 "\ \ / \ zv 22 20 18 IB n n 10 / \ \ \ I \ / \ s ' / A \ ,^ / \ \ / ^ •^ ^ 1 \ / \ \ 1 / \ s H 1 1 f ^\ ^A \ / 1 1 ""■- ^ ^ 8 6

I. /u ^ L^ ^ — — - — — ..^ ^ ^ — -^ ^


z £ ■ — xC — ' '^ =^ =; N ^--M "^ 1 I 3 V J 6 7 9 10 Tl 12 13 n IS 16 1? 18 IS ZO 21 Days Fig. 58. £ = fore limb; G = brain; ^=heart; Z,=lens; jV= metanephros; f/ = mesonephros. where Cv is the instantaneous percentage growth-rate for the organ or part in question, v-^ and v the weights of the organ at the beginning and end of the period in question, Cw the instantaneous percentage growth-rate of the organism as a whole and w-^ and w the weights of the organism as a whole at the beginning and end of the period in question. But this would not take into account the time of formation of the various anlages. More complicated expressions have, therefore, to be found, but as they do not at present seem to have any direct importance for the chemical embryologist, a reference to the original paper must suffice. They involve the computation of an "extensity factor" which is usually the same as the 448 ON INCREASE IN SIZE [pt. iii constant a, already referred to, and an intensity factor which is the corrected product of the instantaneous percentage growth-rate and the time, i.e. Cvt. The relative extensity factor is obtained by determining the time which is taken by the organism or the organ to grow i mm. in length, thus reversing the process by which a was originally found. The size of the anlage is also included, and called the mass factor. By the aid of all this apparatus, Schmalhausen compares organs on a common basis, i.e. the time taken for i mm. increase in length to be made. Thus the extensity factor of the chick embryo brain [r) is 1-27 and that of the duck embryo brain 1-26, which means that the sizes of the respective organs are in their earliest stages almost identical. But the relative instantaneous percentage growth-rate (intensity factor, k, or Cv) differs considerably, for in the chick it is 1-87 and in the duck 2-01, which means that the duck embryo brain grows distinctly more rapidly than that of the chick embryo and finally attains a larger size. Again, for the stomach of the chick embryo the extensity factor, r, is 0-244 ^^^ ^^^ the duck 0*324, but the intensity factor is 3-59 for the chick and 2-86 for the duck, or, in unquantitative terms, the stomach is rather bigger to start with (relatively) in the duck than in the hen, but the chick stomach grows faster and reaches eventually a larger proportion of the body. This work on disproportionate or heterogonic growth led Schmalhausen into a field which had been in course of investigation by Huxley and others. Schmalhausen was able to obtain Huxley's formula from his own, and concluded with some justice that his own were of fairly general validity and did not hold only for embryonic growth. On the other hand, owing to the absence of a true extensity factor in Huxley's formula, the latter could not be applied to the embryo; for, although in post-embryonic growth-curves the ages of all the organs can be taken as approximately identical with the age of the animal, this is by no means the case in embryological work, where the time of formation of the various " anlages " is of the greatest importance. The investigations on heterogonic growth are not immediately germane to the theme of this book, but they may at any moment become very important for the chemical physiology of the embryo, and it is necessary therefore to be aware of them. Schmalhausen's work is really an extension to the embryo of the conceptions of Pezard and Champy, as worked out in recent years by Huxley SECT. 2] AND WEIGHT 449 and his associates. A rich harvest awaits the investigator who discovers the relation between chemical constitution and the differential growth-ratios. Perhaps a fruitful line of work will develop from the finding of Robb that the log. weight of an organ plotted against the log. body-weight often gives a straight line. He has suggested that organ-growth may depend on a kind of partition-coefficient, organs competing, as it were, for the building-stones in the blood-stream, and securing now a greater now a lesser proportion, according to the changing permeability of their cell-walls. The changes which occur in the chick's relative growth-rates of parts at hatching have been studied by Latimer, who combined together the data collected for pre-natal stages by Schmalhausen and those for post-natal stages by various American workers. His results lead to the conclusion that the organs and parts fall into three groups : (i) Those in which no change in relative growth-rate is found, e.g. liver, gizzard, feathers, ovaries. (ii) Those which show a brief post-natal retardation, e.g. total body-weight, brain and heart. (iii) Those which show a marked post-natal acceleration, e.g. kidneys and spleen. As will be shown below, the brain and eyes, so prominent in the embryo, fall consistently throughout life when expressed as per cent, of the whole weight, while the gizzard, liver, kidneys, spleen and heart have a maximum in early post-natal life. Other work on relative sizes of parts has been done by Jenkinson on embryonic trout, by Keene & Hewer on man, and by Jackson who gives a graph (Fig. 59) showing the relative proportions in the human embryo at different stages of its development, collected from all the available data. Boyd (on man) and Welcker & Brandt (on the chick, salamander and man) made earlier attempts at the same thing, but the ages of their embryos were unknown. The graph demonstrates the relatively large size of the brain in the early sizes, and in many ways resembles the graph for the organs of the chick given by Schmalhausen. "In general," says Jackson, "the period of maximum relative growth passes in a somewhat wave-like manner over the body from the head towards the foot. The head reaches its maximum relative size about the 2nd month. In the trunk, the upper portion, including the thorax and the upper abdominal viscera, is relatively largest throughout the earlier half of foetal life. NEi 29 450 ON INCREASE IN SIZE [PT. Ill The lower part of the abdomen becomes more prominent towards the end of the foetal period, due chiefly to the rapid expansion of the intestines at this time. The pelvis and lower extremities do not reach their greatest relative size until early adult life, although the upper extremities have reached their maximum relative size at birth. It may also be noted that the organs lying dorsal to the body axis grow at first far more rapidly than those ventral to the body axis, for, 20 40 60 80 100 120 140 160 180 200 220 240 260 280 Days Fig- 59 while in the 2nd month the former are three times the size of the latter, at birth they are equal, and in the adult the latter are six times the size of the former." Jackson's data on the growth of such organs as the suprarenal gland should also be of much service to chemical embryologists, and his paper as a whole is of great value, as it summarises the results of all the earlier workers — Welcker & Brandt; Brandt; Anderson; Lomer; Meeh; Liman; Thoma; Oppenheimer; Collin, Lucien & Beneke; Devergie; Schmidtt and Elsasser. The general results of all the workers who have occupied themselves with the weights of foetal parts are tabulated in Table 59, which gives the point in development at which the maximum percentage of the total body-weight is reached, what that percentage is, what it becomes at birth and what eventually it is in the adult animal. SECT. 2] AND WEIGHT 451 Table 59. Relative maximal weights of constituent parts of the embryo. Point at which m ^ v I m 1 1 m 1 <i % of total body-weight Part or LLL€XJ\.LLkL\jixL\. 13 attained in At maxi At birth or At adult organ development mum hatching stage Pig Total viscera 1 5 (length in 38 160 8 (Lowrey) Head 1 8 mm.) 30 22-0 6 Brain i8 9 4-0 0087 Spinal cord i8 1-87 0-33 004 Eyeball 86 1-15 04 o-oi Heart 15 4-64 10 0-32 Lungs 86 3-9 20 07 Liver 25 15-9 31 1-38 Kidneys 58 2-59 I-OI 025 Mesonephros 15 120 — Spleen 15 — 0-17 0-13 Pancreas 15 — o-i6 0-14 Thymus 15 — 0-37 — Thyroid 58 — 0026 0-004 Suprarenal 58 0-13 O-OIQ 0-005 Intest. and stomach 15 3-6 4-79 Man Head 2 (months) 45 26-0 (Jackson *) Trunk I 65 40-0 — Fore limb 10 10 lo-o* Hind limb 10 20 200 i 72-4 Brain 2 20 135 — Spinal cord I 5 0-15 — Heart I 5 0-7 — Liver 2-5 7-5 5-0 — Lungs Spleen 4 ID 33 0-4 2-0| 0-4 i 0-94 Thymus 10 0-3 0-3 — Thyroid ID 0-I2 0-I2 — . Kidneys 7 i-o 1-05 Suprarenal 3 0-45 024 0-90 Chick Heart 4 (days) 1-5 0-56 — (Schmalhausen) Mesonephros 4 o-6i 0014 — Metanephros 16 0-39 023 — Brain 4 300 2-6 — Lens 6 0-17 0025 — Fore limb 16 32 2-1 Stomach 20 3-59 3-56 — Dogfish Head 0-09 (wt. in 40-0 17-5 12 (Kearney t) Skin (birth) 75-0 gm.) 11-3 I I -3 7 Skeleton 85 100 8-6 9 Muscles (adult) 630 45-0 63 Total viscera, (a) c. o-i 19-0 I2-0 — two maxima (b) c. 350-0 14-3 — 9-8 Brain 01 15-0 1-6 0-9 Spinal cord 01 1-76 0-5 0-17 Eyeballs o-i 90 2-0 064 Heart O-I 4-0 0-15 0-20 Pancreas 350-0 0-14 006 0-14 Liver 20-0 7-0 4-8 59 Spleen 350-0 0-38 0098 0-2 Rectal gland 01 01 05 0032 — Mesonephros 1-8 4-8 i-i 0-38 Testes and ovaries 200-0 0-9 0-4 0-28 Stomach and intest. 350-0 5-5 2-9 4-0

  • Including all previous work on man.

t "The various organs and parts in dogfish [Mustelus cams] embryos and adults show relative growths strikingly similar to that which has been observed among the higher vertebrates, including mammals and man". 29-2 452 ON INCREASE IN SIZE [pt. m From this it can be seen that the organs which reach their maximum relative weight early in development are the heart, spleen, pancreas, thymus, brain, spinal cord and head. The mesonephros also, of course, reaches its maximum fairly soon and declines more or less rapidly afterwards. The muscle masses, shown especially in the figures for fore and hind limbs, increase steadily in relative weight and reach their maximal relative size at or shortly before birth. The suprarenal gland, the stomach, the lungs, and the thyroid are variable in their point of maximum. But, as can be seen from the table, the data on these matters are not very numerous, most of the attention which has been given to the relative weights of parts and organs having gone into the study of post-natal life. W. Schultze has made an interesting investigation on the effect of hormones on the developing parts and organs in the tadpole. It is interesting that the only conclusions to which Jackson would commit himself were ( i ) that the embryo grows much faster in the earlier stages than in the later, and (2) that, at any rate as far as vertebrates were concerned, pre-natal growth is relatively much greater at the cephalic than at the caudal end. These points had already both been stated by Aristotle, and the whole advance lay in giving them a quantitative backing. Jackson did not consider that his figures supported the view of Preyer that those organs grow fastest in the embryo which will afterwards first come into functional operation. A great mass of such data has since Jackson's time been collected by Calkins & Scammon and many investigators working under their influence. We heed not do more than mention their work on the growth of the spinal axis in the human embryo, that of Scammon on the height-weight index, Nafiagas on anencephalic embryos, and Brody and Hammond on proportions in the cow, for these and many others only indirectly concern us. But it is important to note that for the human embryo Calkins & Scammon found that from 3 months onwards the growth in length, girth and diameter of the various external divisions of the body was directly proportional to the growth in total body-length. While each dimension has its own growth-rate with respect to the total body-length, this characteristic rate does not alter during the period under consideration. All these entities then may be expressed by the Calkins equation D =^ aL±b, SECT. 2] AND WEIGHT 453 where D is the dimension in question, L the total body-length, and a and b constants. The constant ^ is a measure of the amount of growth that has gone on prior to the period in question, and, as it is negative for the limbs but positive for all head and neck measurements, the conclusion is that while the latter have been growing extremely rapidly before the period began the former have not. This is as would be expected. The same conclusion emerges from the data of Corrado on the weights of head, trunk and extremities as analysed by Scammon. Many organs have been examined by the investigators of this school. The cerebellum, for instance, was found by Scammon & Dunn to increase in absolute volume and weight first slowly and then more rapidly during the foetal period; thus, its percentage growth-rate rose for the first six months of pregnancy, only to fall sharply afterwards. The pancreas, studied by Scammon, grows at a rate very like that of the whole embryo, but the relative weight of the organ with respect to total body-weight undergoes a reduction from 0-3 per cent, at the 4th month to o-i per cent, at birth. The uterus, on the other hand, passes through two definite phases in pre-natal life. Until 7 months the organ shows a lineal increase with respect to body-length which is comparable to that of most lineal body-dimensions, but after 7 months it grows much more rapidly. In early post-natal life, however, the organ goes through an involution stage which has long been known, actually decreasing in size by hypoplasia and hypotrophy until it reaches the level it would have attained had the early foetal growth-rate been continued. "This suggests", says Scammon, '^that the growth of the uterus in the latter foetal months consists of a substratum of typical foetal growth plus a secondary increment due to an extra stimulus furnished by a hormone of placental or possibly ovarian origin." Here is an excellent illustration of how an organ can act as an index registering obscure physico-chemical changes in the internal environment of the embryo. The other organ which undergoes a reduction in size following birth in man is the suprarenal gland, and it also has been investigated by Scammon. But, unlike the uterus, its growth when observed in the foetal period shows no increased intensity towards the time of birth, so that the involution which occurs afterwards by degeneration of the two inner cortical layers decreases its size far below what it 454 ON INCREASE IN SIZE [pt. iii would have reached had it gone on growing at the same velocity as before birth. This leads on to the general problems raised by the growth of the ductless glands in the embryo, problems of the greatest importance in view of the regulating influence which the foetal endocrines probably exercise. The pituitary gland, according to Covell, grows proportionately to the total body-weight in human pre-natal life, i.e. slowly till about the 5th month and more rapidly thereafter. The thyroid also shows no outstanding variations from the normal curve. The growth of the thymus, however, is characterised by high variability, while the pineal gland grows at nearly the same rate as the brain. The results of studies on the weight of these glands, therefore, do not reveal any striking correlations, and they must be supplemented by histological evidence. This will be presented in the section on hormones. Scammon; Scott; and Scammon & Kittleson have studied the growth of the intestinal tract and the stomach in the human embryo. While this work does not give us any help in evaluating the active absorptive surface during embryonic life, its main conclusions are of interest. Thus the growth of the gastro-intestinal tract follows the law of antero-posterior gradient or direction, for the more cranial portions grow relatively more rapidly in the early part of foetal life, while the successive caudal portions show smaller amounts of growth at the beginning and larger ones later. The number of crypts and glands in the stomach mucosa seems to increase per sq. mm, very regularly during the progress of foetal growth. Watkins' study of the growth of arteries is also interesting, for it shows that the vessels which supply the foetus only have a rate of growth similar to that of the body as a whole, i.e. slow for a short time at first, and then for a long time rapid, while the arteries which supply the placenta as well as the embryonic body have a long period of slow growth followed by a short period of rapid growth. These facts throw a certain light on the metabolic needs of the developing organism. Much valuable information is contained in the papers of Scammon and Armstrong on the foetal growth of the eye, and in that of Noback on the respiratory system, but it cannot be given here. Davenport has recently considered the growth-curves of man in the light of the work of Scammon and his associates, and Gunther has written on these subjects especially in relation to sex. SECT. 2] AND WEIGHT 455 2-1 1. Variability and Correlation Enough has now been said about the growth of the parts and organs of the whole considered in isolation, and we must consider the relation between the growth-rate and two other factors, namely, variability and correlation. The population of cells in the metazoal embryo may no doubt be compared with the populations of protozoa in cultures, but, whereas the functions of the whole in the latter case are very limited, those of the whole in the former case are highly complex. In other words, one may enquire to what extent there is variability between different embryos of exactly the same fertilisation age. Closely allied to this question is what is, to all intents and purposes, its converse, namely, at what point in development is the correlation coefficient greatest, i.e. at what point is the swing of variation among embryos away from the mean least? It is to be regretted that these enquiries have not been very deeply carried on in embryology, but there are some rather significant observations which need attention. So far only the mean values for weights and measures of embryos have been under consideration. But obviously no statistical study of these is complete without a consideration of the amount of variability among the individual cases from which the mean value is derived. The variability coefficient is defined as the standard deviation X 100, mean the standard deviation being a measure of the spread of points around the mean, i.e. a measure of the point upon the frequencycurve where the change takes place between concave to the mean and convex to it. Fig. 22 showing McDowell's points will explain the meaning of this. It has long been known that the variability coefficient decreases with age in man, and it is always stated that it follows the changing growth-rate quite closely, but some confusion has been caused in the past by a doubt as to what manner of representing the growth-rate is being referred to. The fact is, however, that the variability coefficient follows the simple increment curve. Thus, if for absolute growth a sigmoid curve holds good, the greatest daily or monthly increment will occur as we have seen at the middle of the period, and this peak will coincide with a peak in the variability coefficient. This was found to hold in 456 ON INCREASE IN SIZE [pt. m actual fact by Boas & Wissler, by Boas, and by Bowditch, who studied exhaustively the growth of Toronto school-children. The variability coefficient in that case followed exactly the curve of yearly increment and reached an exactly simultaneous maximum at the age of 15 years. The correlation coefficient behaved in the same way. Boas & Wissler explained their results by saying that correlations between measurements in one individual ought naturally to be greater during periods of rapid growth than at other times, because the variations in responsible factors will affect them all to an equal extent. Variability is what governs correlation, so that would also be expected to rise and fall in the same manner. But there is a proviso that must be made here, for Boas & Wissler's use of the term "growth-rate" is not the same as that of Brody, for instance. Boas & Wissler mean by the time of greatest growth-rate the time at which the largest increments are being added on to the organism in unit time, i.e. the half-way point in the curve of the autocatalytic equation. Brody means by the time of greatest growth-rate the time at which the organism is adding on to itself the largest relative increments, i.e. the earliest stages of embryonic life, when the slope of the log. weight/age curve is extremely steep, and the embryo doubles its weight in an exceedingly small lapse of time. The variability coefficient shows, therefore, that it is at any rate true to say that, during the phase when the largest absolute increments are being made, the widest variations from the mean tend to occur. This seems very reasonable, but there is also evidence which shows that, when the period of most rapid growth in Brody's sense is occurring, the variability coefficient is also large. Other examples are numerous. Buchem found a coefficient of variability of 0-4 for the early stages of the embryo cow and o-i later. Edwards found a variability coefficient of 0-1347 for unincubated chick blastoderms, but of 0-1087 for those incubated 24 hours. Jenkinson gives a graph exactly analogous to those of Boas & Wissler computed from Roberts' measurements of English artisans. Jenkinson also worked on the trout embryo (or rather the alevin, for his first point was 2-3 weeks after hatching, by which time the yolk was not completely absorbed). He found that there was a close general agreement between weekly increment and variability coefficient for the first 10 weeks after hatching, true not only for the growth and variability of the body-length but also for some of the parts such as SECT. 2] AND WEIGHT 457 the diameter of the eye, the length of the head, and the length of the caudal fin. Fig. 60, taken from Jenkinson, shows how, although the curves do not run absolutely parallel, they certainly rise and fall together. On the other hand, the correlation coefficients between several pairs of organs show that in many cases — total length and breadth of caudal fin, total length and length of anterior dorsal fin, total length and length of head, head length and eye diameter — there is a significant diminution in value during the time that there 70 a .^ 60 iB 50 40 C « 30 c y 20 o 10 < Tot \ \ al lar AX /ae. Total 1 lengt h At \ \ ^ ^^V, \ \ \ V V J ^ \

  • %


\ X^^^ ■» m -, .3 1'3 2'3 3*3 4*3 5-3 6*3 7'3 8*3 9*3 10-3 Weeks after hatching Fig. 60. is a decrease in growth-rate. Thus, when the growth-rate is highest, the variations between individuals are greatest, but the correlation coefficients between various organs or parts in the same individual are also greatest. It is easy to see why this should be so, but Boas has given a mathematical proof of the relation between these coefficients and the growth-rate. Expressed differently, it could be said that the faster the growth-rate the more proportional the growth but the greater the variation as between individuals. Turning now to the early part of the embryonic period, the first complete investigation was that of Fischel, who studied the individual variations between duck embryos at the primitive streak stage. Von Baer 458 ON INCREASE IN SIZE [PT. Ill had already averred qualitatively as early as 1828 that variability was much more pronounced in the earlier stages than in the later ones. Kupffer & Benecke; Keibel & Abraham; and Assheton afterwards drew special attention to it. Fischel, however, measured the length of the embryos, and found that the older they were the more regularly they agreed together. This led him to conclude that regulating influences came into play during development which brought about a more synchronous course of growth and differentiation, and made 6 7 8 9 10 n •^ Number of Somites Fig. 61. the individual variations less and less obvious. Such a standpoint would accord well with frequency-curves such as that of McDowell and his collaborators for the mouse embryo (see Fig. 22), where the range of weights on a given day is 4 or 5 times as large at the beginning of development as it is at the end. His and Levi, working on the development of the chick, came across the same phenomenon. Fischel divided the total length of the embryo into a number of constituent lengths, e.g. from the cephalic to the caudal end of the somites and from the extreme cephalic end to the anterior blastopore ("Darmpforte"). He distinguished 13 such lengths, and these he measured in a large number of embryos, judging the age in each case by the number of somites formed. As will be seen from Fig. 61, the limits of variation considered as absolute maxima and minima SECT. 2] AND WEIGHT 459 are much the same when the embryo has o somites as when it has 20, but, in view of the increase in total length during this time, it will be seen that the maximum variation is much less important as time goes on. "The relative differences", said Fischel, "are certainly less in the later stages than in the earlier ones." Fischel's measurements of the lengths of the parts all showed the same relation as regards variability, but, though the length of the body is increasing regularly all through this period, the length of the part between the anterior end and the ist somite remains practically stationary, as does the length of the part between the last somite and the posterior end of the embryo. In other words, the increase in length is entirely due to growth of the middle region in which the somites are being produced.* The size of the individual variations can be large; thus an embryo may be more than 50 per cent, longer than another one of the same stage. Fischel's examination of the work of other authors, such as that of Bonnet on the sheep and of Keibel on the pig, induced him to suppose that very similar effects were seen in mammalian embryos. Philiptschenko carried the question into insect development by investigating an apterygote, Isotoma cinera, and found that the older stages showed the greater variability. On the other hand, Zuitin, who has studied the development of Dixippus morosus from this point of view, found that, just as in the birds and mammals, the earlier stages were the ones which showed most variations from the mean. Schmalhausen has also considered the question on the basis of the figures he obtained in his studies on the growth of the chick embryo, already referred to. He points out that the early stages in embryonic development are those when the anlages are being formed. At that time, every few hours, as it were, are marked by the start off of one or more parts or organs on the long declining hyperbola which represents their instantaneous growth-rate. Thus a cross-section through an embryo in those early periods would demonstrate, if some method of the future made it possible, a series of growth-rates, some low, appertaining to the more senior organs, some high, appertaining to the more junior ones. Moreover, the mass of each anlage is different for each organ, so that extremely complicated effects will be observed if the weights of the organs are described in percentage of the total body-weight. Thus

  • As Levi has shown, embryo size in the somite stage is very similar no matter what

the size of the adult bird, but soon the larger animals grow longer. Also the size of the first somites of all sauropsida is between 5000 and 8000 /x^ surface. 460 ON INCREASE IN SIZE [PT. Ill 10-0 the earlier the stage the more chance there will be for individual differences in growth-rate to reveal themselves, while in later development these will be equalised, adjusted, and compensated by a process of self-regulation. Schmalhausen does not say, however, what this process of self-regulation is, and its nature, indeed, offers one of the most interesting problems in embryology. We shall refer to it again in the chapter on hormones. Schmalhausen suggests that Philiptschenko's results might be explained by the extremely small time elapsing between the formation of the various anlages in animals with ultra-rapid incubation times such as some insects. Owing to the high i