VI SOME MODERN EPIGENETIC THEORIES T HE theories of the germ-plasm and the gene, and in fact all theories dealing in terms of hypothetical particles conditioning or determining development, belong, of course, to that same current of thought which produced the early theories of preformation. We have now to consider some modern epigenetic views, whose upholders naturally found themselves in more or less complete opposition to the modern preformationists, and we select as the main representative of this line of thought the distinguished French zoologist Yves Delage.
In his remarkable book on the major problems of biology 1 Delage gives a valuable historical account and critique of the main theories of development and heredity, particularly those of the nineteenth century, and indicates in broad outline a theory of his own which he calls the ‘theorie des causes actuelles’.
He opposes vigorously the Weismannian conception that all the potentialities of the germ are already present therein in material form, and he rejects completely the notion of representative particles. It is not necessary to follow in any great detail his thoroughgoing criticism of the numerous ‘micromeristic’ or particulate theories, which he contrasts to their disadvantage with the ‘organicist’ views held by W. Roux and himself; we may, however, briefly consider his treatment of Weismann and H. de Vries. To the general hypothesis that there may exist ultra-microscopic particles, intermediate between cellular units and molecules, and capable of assimilation, growth, and division, he raises no objection in principle, but he points out the futility of assigning to such particles special powers of determining the course of development. If they are assumed to represent and
1 La Structure du Protoplasma et les Theories sur V Her editi et les grands ProbUmes de la Biologie generale. Paris, 1895, 2nd edit., 1903; quotations are from the first edition.
SOME MODERN EPIGENETIC THEORIES 77 determine particular characters of the organism, two alternative logical possibilities arise. Either they represent the concrete characters of the organism, in which case their number must be infinite, or they represent abstract or subjective characters, in which case their own existence is equally abstract and subjective, i.e. they do not exist materially at all but are mere symbols. Weismann’s system is one of complete preformation — all his Ids must have been present in the ancestral Protozoa — and it does not allow for the influence of environment and the new-creation of form. He has got all that can be got from the concept of representative particles ; if his system is a failure it is because the fundamental hypothesis is false: ‘Let us therefore draw the conclusion: there do not exist in the germ-plasm distinct particles representing the parts of the body or the characters and properties of the organism’ (p. 719).
H. de Vries’ theory of intra-cellular pangenesis (1889) has in some ways so close a resemblance to the gene theory that it is of much interest to consider Delage’s criticism of it. De Vries held that the germ-plasm was present in the nucleus of every cell, germipal and somatic alike, and that cellular differentiation was due to the issuing out from the nucleus of particular kinds or groups of pangens which impressed upon the cytoplasm its special character. The theory is well summarized by Delage as follows:
‘The elementary characters and properties of organisms have as factors material particles, the pangens, intermediate between chemical molecules and cells; the immense diversity of organisms is based on the almost infinite variety of possible combinations of the pangens; the pangens are contained in a latent state in the nucleus, which serves to transmit them to the fertilized egg in reproduction, and from cell to cell in ontogeny; finally these pangens issue from the nucleus and spread through the cytoplasm to which they impart its particular properties, and cellular differentiation results from the fact that each cytoplasm receives only those that it needs for its particular development and for the functions it is destined to fulfil’ (p. 660).
Delage argues that the elementary characters which de Vries distinguishes are for the most part the product of subjective
78 SOME MODERN EPIGENETIC THEORIES selection by the observer, and that accordingly they cannot possibly be represented by material particles.
‘De Vries reasons’, he points out, ‘as one who should say: copper has a certain density, it is malleable, capable of taking a polish, it is yellowish red in colour and gives out a particular smell when rubbed, it oxidizes in certain conditions, &c., &c. All these properties are independent, for we see them all varying independently of one another. Thus colour is independent of density, for gold is yellower and less red than copper, it is also heavier; and lead is also heavier, though it is neither yellow nor red. Density for its part is independent of hardness, for lead, though it is denser is yet softer, and tin, though less dense, is softer also, while iron and platinum, though one is less dense and the other more dense, are both harder. From which we can conclude that all these properties, density, colour, hardness, smell, &c., are independent and must be supported by independent material factors. Metals then are not simple bodies, they are formed of particles of which some supply colour, others density, others malleability and so on, and the different metals result from different mixtures of these elementary particles’ (p. 662).
The analogy sounds a little far fetched, but, as we shall see later, it is from Delage’s own point of view quite a fair analogy. He raises the further objection to de Vries’ theory that it does not account for the activation at the proper time and place of the special sets of pangens responsible for cellular differentiation ; since by hypothesis all the nuclei have the complete set of pangens, the stimulus to differentiation, to the issue of the right pangens from the nucleus, can only come from the cytoplasm. This is, of course, a difficulty which the gene theory also has to face (see above, p. 70).
Delage’s own view of development and heredity may best be described as a theory of chemical epigenesis. His fundamental hypothesis is the mechanistic one that the structure of protoplasm is the mechanical cause of the phenomena of life. He holds that life is a resultant of the properties of living matter, and that these properties are the outcome of its physico-chemical constitution (p. 403). His theory is singularly free from speculation; it follows the facts closely, and is in effect a generalization of the observed data, con
SOME MODERN EPIGENETIC THEORIES 79 sidered from the physiological point of view. Therein lies its great value. Let us summarize its main outlines.
Protoplasm is a complex chemical mixture with, in addition, a certain structure. There is no reason to assume the existence of permanent vital units of a lower order than the cell. Nuclear division is always exactly quantitative and gives identical products; differentiation accordingly must be of cytoplasmic origin, for it is in their cytoplasm that two daughter cells first of all show differences, and if differences are later established in the nuclei, this can only come about subsequent to, and as a consequence of, the cytoplasmic differences (p. 759). The nucleus is thus not in any sense the dominating partner in the cell, as most other theories assume.
There is no rigid determination of characters in the egg. Differentiation in development depends to a large extent upon the relative position of the parts and upon environmental conditions. There is an actual new-formation of substances in ontogeny — a chemical epigenesis — and histological differentiation is in the main a consequence of this chemical differentiation. All development is based upon metabolic processes, and the nature of the food and of the products of dissimilation exerts a profound influence upon its course. The functional correlation of organs and cells is important. ‘In sum, ontogeny is not merely the development, separation, and accentuation of tendencies completely represented in some form, material or other, in the fertilized egg. It is partly that, and partly a progressive formation of parts and of properties actually new, and the initial constitution of the egg is only one of the indispensable conditions for this formation’ (p. 765).
Delage accepts in general the distinction between somatic and germinal cells, but does not regard it as absolute. For him the ovum is relatively simple in structure, essentially a slightly differentiated cell which has taken little or no part in somatic differentiation, and is thus capable of returning to its initial condition. The eggs of related species are much more alike than are the adults ; 1 the egg does not contain all
1 Contrast O. Her twig’s concept of the ‘Species-Cell’, Allgcmcine Biologic , 6~7th edit., Jena, 1923, p. 492.
80 SOME MODERN EPIGENETIC THEORIES
the elements of its development — the greater part it will find in the environment or make by the way. But while the egg has little or no preformed structure ‘its physico-chemical constitution is extremely precise, and the least difference in this respect is necessarily amplified to considerable proportions in the course of ontogenetic differentiation, and may lead to the considerable differences which exist between adults arising from different eggs’ (p. 772). Much of the apparent preformation is really environmental — to use a paradoxical expression. Since the egg is of precise and delicate organization it can develop only under very exactly delimited conditions.
‘It is therefore held between these two alternatives — to find from moment to moment the exact conditions necessary to it, or to die. Therein lies the whole explanation of heredity. For these conditions are precisely those which the egg of the parent has encountered at each corresponding stage. It is therefore inevitable that it follow the same development as the egg of the parent, since it has the same physicochemical constitution, and encounters, in the same order, a series of identical conditions which are rigorously determined. It is therefore not necessary that it should contain within itself all the factors of its development. It is sufficient that it contains one of the numerous factors indispensable to the identical reproduction of all the developmental phenomena, the other factors, no less indispensable, are situated outside of it, but the egg is certain to encounter them at the proper place and time, without which it dies, and development is not diverted but stops. To the inorganic object, the star or the river, there are open at every moment a thousand different ways which all lead to a normal goal; its development has nothing precise about it. To the organized being also many diverse paths are open every moment, but all lead to certain destruction, save one — that path which leads it to the goal its parents reached. Is it then necessary to suppose that it knows the way, or to be amazed that when it has succeeded in following a line right to the goal, this way has led it to the same goal its parents reached before it V (p. 777).
The orderly progression of development, its repetition of the parental ontogeny, and its steady course towards an apparent end or goal, finds thus a simple explanation which is independent of any elaborate hypothesis of preformation in the egg.
As applied to ontogeny, the theory of representative
SOME MODERN EPIGENETIC THEORIES 81
particles merely adds unnecessary complications to the simplicity of the facts. The fertilized ovum exercises certain simple vital functions; its descendants gradually acquire specialized functions, though no individual cell acquires any excessively complex or multiple functions. The cells therefore change in the course of ontogeny, but if one does not imagine them loaded with all the future properties of their descendant cells (as is done in the preformistic theories) there is no need to ascribe to them any excessively complicated structure. The fault of all preformistic and particulate theories is to translate these future possibilities of development into material predispositions — to hypostatize all potentialities as material actualities, preformed and held in reserve. But these ‘potentialities’ are purely virtual and conceptual — their appearance is entirely dependent upon future environmental conditions. ‘Latent’ characters of this kind are accordingly purely conceptual and do not actually exist. ‘It is the erroneous idea that the properties of the daughter-cell, and in consequence the physico-chemical aggregate which is the basis of these properties, must be found already formed in the mother-cell, that has led to the forging of so many hypotheses, as useless as they are improbable, about the structure of the idioplasm’ (p. 779).
The egg, therefore, regarded as it is and without hypothesis, has only the characters of the moment, and does not already possess those of its descendant cells ; there is accordingly in each ontogeny a real formation of characters not originally present in the egg.
‘ Latent or potential characters are absent characters. . . . The egg contains nothing beyond the special physico-chemical constitution that confers upon it its individual properties qua cell. It is evident that this constitution is the condition of future characters, but this condition is in the egg extremely incomplete, and to say that it is complete but latent is to falsify the state of affairs. What is lacking to complete the conditions does not exist in the egg in a state of inhibition, but outside the egg altogether, and can equally well occur or not occur at the required moment’ (p. 781). Ontogeny is not completely determined in the egg.
82 SOME MODERN EPIGENETIC THEORIES
In effect, Delage rejects the idea of germinal substance ; he states his theory throughout in terms of cells, regarded as physico-chemical systems.
A further important consequence of this objective, physiological interpretation of the constitution of the egg-cell is the rejection of the preformistic notion of a point-to-point correspondence between the structure of the germ-plasm and the structure or properties of the adult organism. From this point of view there is absolutely no foundation for the theory of representative particles. ‘From the physicochemical structure [of the cell] result properties and characters which cannot help expressing themselves, and are, taken together, the result of the totality of this structure’ (p. 782). 1
This general determination of properties is not peculiar to organic systems; the totality of properties belongs to the totality of the aggregate in chemical compounds as well. 2
It is quite possible, he argues, to harmonize this view of general determination with the undoubted fact that special peculiarities may be transmitted apparently as units. A slight modification of the germ-plasm may result in a specific and definite change in the adult organism, and may accordingly recur in future generations, but the modification itself is not necessarily particulate and localized; it must, from the physiological point of view, be regarded as a slight but definite change affecting the physico-chemical system of the germ-plasm or egg-cell as a whole.
When, for instance, two allied forms differ in some special characters their germ-plasms are extremely similar, but there must exist nevertheless a very slight difference of physicochemical constitution between the two germ-plasms affecting the whole and not merely the particular parts specially related to the characters in question.
‘All the parts of the egg . . . including the ultramicroscopic parts
1 ‘De cette structure physico-chimique r&ultent des propri£t£s et caract£res qui ne peuvent pas ne pas s’ exprimer, et qui sont tons ensemble le r6sultat de V ensemble de la structure/
2 Cf. his argument concerning the properties of metals (supra, p. 78).
SOME MODERN EPIGENETIC THEORIES 83 whatever they may be, correspond all and each to the totality of the organism. Certain chemical substances may have a special affectation or a localized destiny, but there is no representation of the parts or the characters of the organism by an equivalent number of special particles in the egg, having relations only with these, and remaining latent up to the moment when they arrive in the cells in which they are due to become active. It is therefore a matter of indifference whether the initial peculiarity affects the whole egg-cell or merely one of its parts, it will affect none the less the organism as a whole’ (p. 78 6).
It is interesting to see how in this brilliant analysis Delage arrives by mere dint of clear thinking at the result laboriously worked out by the Morgan school that each gene acts to some extent upon all characters. Delage took the further logical step that representative particles are superfluous abstractions, and the same conclusion applies equally to genes, in so far as they are regarded as independent and particulate units. Even had Delage known of the factorial basis for the gene theory he would not have required to modify his essential argument. If certain characters are actually ‘carried’ by certain parts of the cell, to wit, the chromosomes, these parts are still constituent elements of the physico-chemical system of the cell as a whole, and cannot be treated in isolation from the rest of the cell. Delage’s argument thus includes and allows for this special case.
Delage’s criticism of particulate theories is in fact unanswerable. If the cell and the developing organism are regarded as physico-chemical systems in constant metabolic relations with their environment, external and internal, there is absolutely no place for independent and isolated material units which represent or determine certain characters or groups of characters. No part of the cell can exist in isolation from the whole ; to imagine such is to create a conceptual fiction to which nothing corresponds in reality. To ascribe determinative, as distinct from conditioning, powers to such Active units is unscientific and explains nothing. As a purely provisional concept, the gene may have some practical utility, provided its fictitious nature be kept clearly m mind, but, as we have seen, it is doomed to be superseded
84 SOME MODERN EPIGENETIC THEORIES as soon as the physiological interpretation is discovered of the facts on which the hypothesis is based.
It is not necessary to follow out Delage’s general views on development, heredity, and evolution in further detail. He exhibits throughout a clarity of thought, fidelity to the facts of observation, and a sobriety of hypothesis that make his book one of the finest contributions to general biology ever written.
The physiological point of view is put very clearly also by Verworn, 1 who goes even farther than Delage in his rejection of preformistic doctrines. He points out that heredity and development are essentially phenomena of cells and organisms, and cannot be interpreted in terms merely of hereditary substance. They must be stated in terms of metabolism, for development and reproduction are an expression of changes in the metabolic relations between cell and medium, conditioned by growth, and growth is the cause of all development, both of the individual cell and of the whole cell-community. Like Delage he denies the necessity for assuming in the egg-cell an elaborate and complex organization; it is incorrect, he says, ‘from the fact that the small egg is differentiated into a cell-structure of astonishing complexity, to deduce the idea that the living substance of the former in comparison with that of every other cell, either every unicellular organism or every tissue-cell, must be distinguished by an inconceivably delicate and complex structure’ (p. 537).
The problem of hereditary transmission is met with in its simplest form in unicellular organisms.
‘It is seen here that the transference of the characteristics of the ancestors to the descendants, takes place by the transference of substance which possesses the characteristics of the ancestors. In order that this substance may possess all the characteristics of the latter, it must be a complete cell with all the essential cell-constituents. The characteristic peculiarities of the mother-cell are the expression of its metabolism. If, therefore, the peculiarities of the mother-cell
Max Verworn, Allgemeine Physiologic (1895, 7th edit., 19*2) English Translation, at General Physiology , London, 1899.
SOME MODERN EPIGENETIC THEORIES 85
are to be transmitted to the daughter-cells, its whole metabolism must be transmitted. But this is possible only when a certain quantity of all the essential constituents, i.e., of the protoplasm and nucleus of the mother-cell, passes over to the daughter-cell, for otherwise the metabolism of the latter would not be able to continue, and the cell would necessarily perish. In fact, it is seen not only in unicellular organisms, but everywhere in organic nature, that hereditary transmission takes place without exception by means of the transference of a complete cell with nucleus and protoplasm’ (pp. 544-5).
_/ The physiologist does not take kindly to the idea of a hereditary substance localized somewhere in the cell and transmitted intact in reproduction, for a substance that is to transmit the characteristics of a cell to its descendants must itself be alive. It must accordingly have a metabolism, and this is dependent upon its relations with all the other substances necessary to cell-metabolism, dependent, that is, upon the integrity of all the essential constituents of the cell. The singling out of a single cell-constituent as the special bearer of heredity is wholly unjustified; the cytoplasm is of equal value in this respect to the nucleus, and hereditary transmission is invariably mediated by the germ-cell as a whole.
The real vital unit is in fact the individual cell: ‘if the conception of the organic individual is not to be given up, it must be regarded as an unconditional requirement that the organism be characterized by the presence of all those vital phenomena that have to do with self-preservation. Only the cell fulfils this condition ; it is, therefore, the individual of the lowest order and the elementary organism’ (p. 64).
Verworn, it is true, elaborated a hypothesis that living substance was composed of biogens or elementary units possessing the metabolic characteristics of life, but the merit remains his of emphasizing the fact that life itself and the hereditary potentialities can be handed on only by the cell as a whole. It is the cell and not the germ-plasm that guards and transmits the flame of life — latnpada vital tradit.
The concept of individuality is, as we shall see later, of prime importance for biology. Verworn thought of the
86 SOME MODERN EPIGENETIC THEORIES organic individual as a physico-chemical system; E. G. Conklin in his admirable lectures 1 goes one step farther and associates both the physical and the psychical attributes of life in the common concept of organization. He considers that ‘the body or brain is not the cause of mind, nor mind the cause of body or brain, but that both are inherent in one common organization or individuality’ (p. 49). This applies also to the germ: ‘In the simplest protoplasm we find organization, that is, structure and function, and in germinal protoplasm we find the elements of the mind as well as of the body, and the problem of the ultimate relation of the two is the same whether we consider the organism in its germinal or in its adult stage’ (p. 50). Organization then at all stages includes both structure and function, mind and body, and continuity of organization implies not only persistence of structure but persistence of function. ‘Indeed the entire organism, structure, and function, body and mind, is a unity, and the only justification for dealing with these constituents of the organism as if they were separate entities, whether they be regarded in their adult condition or in the course of their development, is to be found in the increased convenience and effectiveness of such separate treatment’ (p. 81).
Conklin in his treatment of the facts of development lays stress upon the evolution of function, and adopts essentially the same view as Delage — that development is an epigenesis, predominantly chemical, leading to an actual new-formation of structure and function.
‘Development is not the unfolding of an infolded organism, nor the mere sorting of materials already present in the germ-cells, though this does take place, but rather it consists in the formation of new materials and qualities, of new structures and functions — by the combination and interaction of the germinal elements present in the oosperm. In similar manner the combination and interaction of chemical elements yield new substances and qualities which are not to be observed in the elements themselves. Such new substances and qualities, whether in the organic or in the inorganic world, do not
1 E. G. Conklin, Heredity and Environment in the Development of Men, 2nd edit., Princeton, 1916.
SOME MODERN EPIGENETIC THEORIES 87
arise by the gradual unfolding of what was present from the beginning, but they are produced by a process of “creative synthesis” ’ (pp. 88-9).
There is thus both in development and in cosmic evolution a real formation of new qualities, a real ‘emergence’ as Lloyd Morgan would say.
His own researches had shown him the importance of cytoplasmic differentiation in the egg, and he concluded that the hereditary potencies of the male and female germ-cells were not equal — that ‘the polarity, symmetry, type of cleavage, and the pattern, or relative positions and proportions of future organs, [are] foreshadowed in the cytoplasm of the egg cell, while only the differentiations of later development are influenced by the sperm. In short the egg cytoplasm fixes the general type of development and the sperm and egg nuclei supply only the details’ (p. 184). The chromosomes are certainly important, but in no case can the cytoplasm be regarded as merely serving as food or environment for the chromosomes. It is the entire cell, both nucleus and cytoplasm, that is concerned in heredity and differentiation.
In his book The Organism as a Whole 1 that redoubtable protagonist of the mechanistic point of view, Jacques Loeb, tackles the problem of the harmony and co-ordination of the living things, and expresses his dissatisfaction with the particularist conception revived by the researches of the Morgan school. He admits the importance of this problem, which, as he says, has always raised doubts as to the adequacy of the physico-chemical point of view.
‘The difficulties besetting the biologist in this problem have been rather increased than diminished by the discovery of Mendelian heredity, according to which each character is transmitted independently of any other character. Since the number of Mendelian characters in each organism is large, the possibility must be faced that the organism is merely a mosaic of independent hereditary characters. If this be the case the question arises: What moulds these independent characters into a harmonious whole’ (p. v).
The Organism as a Whole , from a Physicochemical Viewpoint. New York and London, 1916.
88 SOME MODERN EPIGENETIC THEORIES
The solution put forward by Loeb is that the unity of the organism is present from the beginning in the structure of the egg, particularly of its cytoplasm. The egg is the embryo in the rough, and the Mendelian factors in the chromosomes add only the finishing touches, probably by giving rise to special hormones and enzymes. Genus and species heredity would on this conception be determined by the cytoplasm of the egg, and would in the long run be dependent upon the chemical specificity of the proteins.
We shall at a later stage in this book come back to Loeb’s theory of the ‘embryo in the rough’; meanwhile, for our present purpose, the main point is his criticism of the particularist point of view, and his substitution for this of the conception of the organism as a harmonious physico-chemical system. His treatment of development is biochemical — it is a question, he holds, of the mechanism by which the ‘nonspecific building stones’ in the food are synthesized into the specific proteins of the species (p. 28). This purely biochemical point of view is very clearly put by R. S. Lillie, 1 who also emphasizes the importance of the specificity of the proteins. The following quotations give a good idea of his standpoint.
‘The problem of heredity is not a problem to be dealt with by itself; it becomes identical with the most fundamental problem of general physiology, the problem of how living protoplasm is synthesized from non-living matter’ (p. 70).
‘It is not a coincidence that living organisms, the most complex systems, in the structural sense, occurring in nature, are also the most complex in the purely chemical sense; and all of the evidence indicates that the structural complexity is the expression or consequence of the chemical complexity. The essential reason for this appears to be that a high degree of chemical specificity or individualization is the necessary prerequisite for structural complexity, and that chemical specificity depends largely upon peculiarities of stereochemical configuration. The number of individualized isomers in the case of any organic compound increases rapidly with increase in the number of asymmetric carbon atoms in the molecule. Hence the proteins,
1 R. S. Lillie, ‘Heredity from the Physico-Chemical Point of View’, Biol. Bull. xxxiv, 1918, pp. 65-90.
SOME MODERN EPIGENETIC THEORIES 89
formed of linked amino-acids, most of which are asymmetric compounds, exhibit the possibilities of chemical individualization to a greater degree than any other known class of compounds. It is further significant that proteins which are specifically distinct chemically, although otherwise closely similar — e.g. haemoglobins from different species — tend to form crystals, i.e., structural aggregates, which are specifically distinct in their form-characters’ (pp. 72-3).
Chemical specificity is the basis of morphological and physiological specificity,
‘and chemical specificity is primarily the property of the proteins. Other biochemical compounds appear to be chemically the same wherever found, but the proteins vary in their specific character from species to species. Moreover, physiologically corresponding or “homologous” proteins are more nearly alike in their chemical and physical characters the more closely related the species are from which they are taken. There is thus a general parallelism between the degree of chemical relationship exhibited by homologous proteins from different organisms, and the degree of biological relationship existing between the species’ (p. 74).
The chromosome theory does not fare very well at the hands of the biochemist, who fails to find any chemical basis in the chromosomes for the extreme complexity of structure required by the theory. This is clearly brought out by A. P. Mathews , 1 who writes:
‘The structure which the cytologist calls a chromosome and in which most see the bearer of all the hereditary traits and some even go so far as to imagine that each trait or character is represented by a distinct unit or gene, this structure as shewn in the fixed dyed section is nothing else than a salt of nucleic acid with the basic dye which has been used to stain it. Whether in addition to the dye there is present also some protein matter cannot be definitely stated’ (p. 75).
Our knowledge of the composition of chromatin lends no support to the hypothesis that the chromosomes are of highly complex structure. Where in theory we should expect to find the most complex compounds, e.g. in the sperm head, we find the simplest.
‘Our ignorance of the chemistry of all these complex compounds,
1 In E. V. Cowdry, General Cytology , Chicago, 1924.
9 o SOME MODERN EPIGENETIC THEORIES
however, is still so great that no chemist would be willing to make the affirmation that the chemical facts conclusively disprove the chromosomal theory of inheritance. At the best, or worst, they give little if any support to this view. In the author’s opinion, which is here given for what it is worth, the chromatin of spermatozoa is nothing else than the chromatin of spermatozoa: and that of an egg cell is the chromatin of an egg cell. They are not nerve, muscle, epithelial chromatin, in masquerade. While when united they may lead to the formation of all the other chromatins of the body, or at least play an important part in their formation, neither should be regarded as a museum containing samples of all the different products which it is capable of making. For since the number of these products is in fact infinite for each chromatin, as is shewn by the differences in cells produced by any change in the conditions of development, by the accidents of existence, such as galls on plants, &c., there is not room in the chromosomes for all these samples. But if the chromatin can make some other chromatins without having a sample to guide it, why not make them all ? Why have any samples at all ? Considerations of this nature will have different weight with different minds. And it must be remembered that the onus of proof is on those who assert that the chromosomes are such museums containing samples of all the chromatin of all the cells of the body, not only all the chromatins which develop during life, but all that infinite collection of old masters inherited from the past and all the infinite number of descendants yet to appear in the eons before us, and presenting qualities usually said to be dormant. They arc concealed no doubt in the chromosomal attic, ready to be produced when occasion arises’ (p. 90).
It would be easy to give many other examples of the modern physico-chemical or physiological treatment of the problems of development ; we shall conclude however with a brief reference to certain general views expressed by C. M. Child 1 which are based on a life-long study of regeneration and regulation. In his book on Individuality in Organisms , Child tackles the problem of unity and co-ordination which constitutes such a difficulty for all particulate theories of development. Though remaining a mechanist, he works out a dynamical and functional conception of the organism and its
1 Individuality in Organisms , Chicago, 1915. See also Senescence and Rejuvenescence , Chicago, 1915.
SOME MODERN EPIGENETIC THEORIES 91 development which is a distinct advance upon the ordinary static ‘machine-theory’. Most theories assume an ‘organization’ of some kind or other as a starting-point ; Child points out that the real problem is the way in which this organization is formed, and he attempts to show how the primary organization of the embryo may arise epigenetically through graded differences in metabolic rate along the primary axes. Theories of representative particles, theories of organic crystals, theories of a pre-established physico-chemical unity, all ‘proceed on the conception that a certain more or less complex “organization” is necessary as a starting-point; the machine must somehow be constructed before it can run’ (p. 25). But ‘to believe that metabolism results from structure and “organization”, as the activity of the man-made machine results from its structure, is to ignore the fact that metabolism is the formative agent in the organism’ (p. 25). The static view of the organism ignores the fact that ‘life is function. In no case does the organism begin to function only after its construction is completed ; it always functions from the beginning; it constructs itself by functioning, and the character of its functional activity changes as its structural development progresses. Structure and function are mutually related. Function produces structure, and structure modifies and determines the character of function’ (p. 16). What we see as protoplasm and structure is to be regarded not as the cause of metabolism and vital activity but as the relatively stable results of functional activity. (Cf. Roux, p. 104 below.)
Child’s own solution of the problem of unity is, in outline, as follows:
‘The organic individual, as a living entity possessing some degree of physiological — not merely physical — unity and order, consists in its simplest forms of one or more gradients in part of a cell, a cell, or a cell-mass of specific physico-chemical constitution. The process of individuation is the process of establishment of the gradient or gradients as a more or less persistent condition, and the degree of individuation depends upon the permanency of the gradient, the metabolic rate in the dominant region, the conductivity of the protoplasm, and probably on other factors as well. From this point of view
92 SOME MODERN EPIGENETIC THEORIES
the assumption of a mysterious, self-determined organization in the protoplasm, the cell or the cell-mass as the basis of physiological individuality becomes entirely unnecessary. The origin of physiological individuality is to be found, not in living protoplasm alone, but in the relations between living protoplasm and the external world’ (pp. 40-1).
It will be seen that Child’s position may best be described as a dynamical epigenesis. In this dynamical conception of the organism, the starting-point of ontogeny lies not in a certain organization but in a certain reaction-system. It is this reaction-system which constitutes the basis of existence, and it is in this system that differences in metabolic rate initiate the process of differentiation (p. 188). The organism acts as a unit in inheritance and development ; it is incorrect to treat of these phenomena in terms of germinal substance, or even of cells, or in terms of any form of pre-existing organization.
‘If the organism is fundamentally a specific reaction system in which quantitative differences initiate physiological individuation, development and differentiation, nothing can be more certain than that it acts essentially as a unit in inheritance. It is the fundamental reaction system which is inherited, not a multitude of distinct, qualitatively different substances or other entities with a definite spatial localization. Development is not a distribution of the different qualities to different regions, but simply the realization of possibilities, of capacities of the reaction system. The process of realization differs in different regions because the conditions are different. Neither characters nor factors as distinct entities are inherited, but rather possibilities, which are given in the physico-chemical constitution of the fundamental reaction system, but not necessarily localized in this or that part of it’ (p. 202).
Child’s dynamical and functional point of view enables him to shake free from the germ-plasm hypothesis altogether, with its absolute separation of soma and germ-plasm, and gives at least the possibility of a rational treatment of the problem of the transmission of acquired characters and the utilization of the mnemic conception (pp. 203-4). His criticism of the particulate theories of heredity is outspoken and pertinent, and we cannot resist a last quotation.
SOME MODERN EPIGENETIC THEORIES 93
‘These theories postulate in one form or another a multitude of specific material entities, each of which represents in some way some characteristic of the organism. The organism as we know it is the product of their combined and harmonious activity. Examination of these theories shows that these hypothetical entities, gemmules, determinants, physiological units, pangenes, specific accumulators, or whatever we prefer to call them, are themselves endowed,*# hypothesis with the essential characteristics of individuals, and that the organism as a whole is merely a composite of their orderly activities. Neither the problem of the individuality of the hypothetical units nor that of their orderly combination and unification in the organism receives any adequate consideration in those theories. They merely translate the problem into hypothetical terms which are beyond the reach of scientific method. The combination of these units into the individual is assumed to occur as the facts demand, and although the problem of the control and ordering of such units through all the changes involved in the development of a complex organism, say the human being, is one which staggers human intelligence, it is practically ignored’ ( P p. 22-3).
Essentially the criticism that Aristotle made of the Hippocratic doctrine of pangenesis! Child has fully appreciated the importance of the problem of ‘composition’.
Our survey of modern epigenetic theories, which are based for the most part upon a physico-chemical or physiological treatment of the problems of development, has necessarily been somewhat slight and superficial. Certain points nevertheless emerge with sufficient clearness. These theories represent a very healthy reaction from the static, morphological, particulate conceptions of the modern preformationists. They keep much closer to the facts of observation, and the purely speculative element in them is less prominent. They represent generalizations of fact rather than hypothetical constructions. They give full weight to functional activity and the influence of environment, and avoid the sterile view that all is predetermined in the egg, in some mysterious germinal substance. Several emphasize the importance of considering the cell as a whole in development and heredity.
The even more significant conception of the organism
94 SOME MODERN EPIGENETIC THEORIES
itself as a functional whole, in which no part or process can be properly understood in isolation from the activity of the whole, begins to emerge in these theories — whether in the form of a physico-chemical system , or in the more subtle but more accurate form of a dynamical or metabolic unity, as in the theories of Child. Conklin gives clear indication of a comprehensive conception of organization, which shall include both the structural and the functional, the physical and the psychical, aspect of living things.