Paper - The development of the islands of Langerhans in the human embryo (1903)

From Embryology
Embryology - 28 May 2024    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Pearce RM. The development of the islands of Langerhans in the human embryo. (1903) Amer. J Anat. : 446-455.

Online Editor  
Mark Hill.jpg
This historic 1903 paper by Pearce is a very early description of the development of the islands of Langerhans, the endocrine portion of the pancreas. The study include embryos and fetuses from the Carnegie Collection.

Internet Archive - 1903 THE AMERICAN JOURNAL OF ANATOMY Vol 2

Modern Notes: pancreas

Endocrine Links: Introduction | BGD Lecture | Science Lecture | Lecture Movie | pineal | hypothalamus‎ | pituitary | thyroid | parathyroid | thymus | pancreas | adrenal | endocrine gonad‎ | endocrine placenta | other tissues | Stage 22 | endocrine abnormalities | Hormones | Category:Endocrine
Historic Embryology - Endocrine  
1903 Islets of Langerhans | 1903 Pig Adrenal | 1904 interstitial Cells | 1908 Pancreas Different Species | 1908 Pituitary | 1908 Pituitary histology | 1911 Rathke's pouch | 1912 Suprarenal Bodies | 1914 Suprarenal Organs | 1915 Pharynx | 1916 Thyroid | 1918 Rabbit Hypophysis | 1920 Adrenal | 1935 Mammalian Hypophysis | 1926 Human Hypophysis | 1927 Adrenal | 1927 Hypophyseal fossa | 1930 Adrenal | 1932 Pineal Gland and Cysts | 1935 Hypophysis | 1935 Pineal | 1937 Pineal | 1935 Parathyroid | 1940 Adrenal | 1941 Thyroid | 1950 Thyroid Parathyroid Thymus | 1957 Adrenal
Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

The Development of the Islands of Langerhans in the Human Embryo


Richard Mills Pearce, M. D.

From the Pathological Institute, University of Leipzig, Prof. Marchand, Director, and The Pathological Laboratory of the University of Pennsylvania.

With 3 Text Figures. (1903)


The investigations upon the pathology of the pancreas carried out during the past several years have had, with few exceptions, for their definite object the determination of the existence of an internal secretion with which the islands of Langerhans and carbohydrate metabolism might be brought into harmony. The great importance to pathology of the studies carried out upon these subjects was t?) be found in the establishment of an anatomical and physiological basis for diabetes mellitus. That this undertaking has been followed by a good measure of success is shown by the important contributions to the pathology of the pancreas in diabetes by Opie, Ssobolew, Weichselbaum and Stangl, Wright and Joslin, and Herzog, as well as the experimental studies of Ssobolew and Schulze. Now that the relationship of a highly important physiological function to the islands of Langerhans seems to be established, it is desirable that the question of the histogenesis of the islands be put on a firm and definite basis. If the islands are Wholly independent structures, as conceived by v. Hansemann, evidence of this fact should be forthcoming from a study of their development; if, at their origin, they are indistinguishable from the proper glandular structure, evidence of this fact should be found in the same study. If the latter conception is the true one, as indicated, but not established, by the studies of Laguessc, Renaut, and Diamare, then the period and the manner of the differentiation should be open to demonstration. The study of the development of the islands in the human pancreas has been very imperfectly pursued, Renaut, as far as I can discover, alone having examined an early specimen.

I have been fortunate in securing during the past year a series of specimens of the human pancreas which were suitable for a systematic study of the development of the islands of Langerhans. This study has, I think, supplied convincing proof upon points which have hitherto been in much doubt, and it affords, in my opinion, a consistent and satisfactory explanation of the development of the islands. I have been able, in addition, to confirm the results of the study of their development in normal organs, with a study of the appearances which they present under the peculiar pathological changes which affect the pancreas in congenital syphilis.

The work falls into two categories: (1) the study of the embryonic normal pancreas, and (2) the study of the pancreas in congenital syphilis. Of the second portion of the study, only those parts which confirm or aid in explaining the normal development will be presented at this time.

For the materials for this study, I am indebted to Professor Marchand, of Leipzig, to Professor Mall, of Baltimore, and Dr. Longcope, of Philadelphia, to whom I wish to express my sincere thanks.

Review of the Literature

A comprehensive description of the development of the islands in the human embryo, I have not been able to find. Laguesse (1) in his description of the embryology of the pancreas of the sheep gives a very complete picture of the early development of the islands. At an early period, the end of the second month (embryo of 18.5 to 50 mm), the primitive gland structure of the pancreas, composed of solid cell masses, present here and there, generally along the outer border, cells which stain more darkly than the general mass. These cells proliferate, lose their granular appearance, and form protruding loops or swellings, spheroid or ovoid in shape. They remain in connection With the tubules for some time and constitute the “primary islands” (ilots endocrines). Later, in embryos of about 90 mm., these structures atrophy and are replaced by similar structures formed by proliferation of the fully formed secreting tubules of the acinus. These “ secondary islands” eventually become tunnelled and vascularized. Laguesse at this time, 1897, believed that they were formed not only during embryonic life but throughout adult life and represented a portion of the gland temporarily modified for a special function, an internal secretion apparently essential to foetal life. He believed, also, that it was possible for these modified structures to revert to the glandular type. Later (2), however, he modified this conception in that although he considered this transformation possible, yet he believed that some of the islands persisted as such throughout life. This change of view was due to the work of Diamare (3) who, supported by Massari (-1), insisted that the islands were definite, constant, and unchanging formations, formed early in embryonic life and persisting until death. Both of these writers attributed to the island an internal secretion. Their observations were made on the pancreas of fish.

Renaut (5) studied the development of the islands in a single pancreas of a human embryo of the third month (11 to 12 cm long), and found it practically to be identical with that observed by Laguesse in the sheep. He also believed that the islands furnish an internal secretion, of unknown nature, of more importance in embryonic than in adult life ; for in the embryo the islands reach their full development, While in the adult he considers them to be rudimentary. This is a radical change from an earlier view expressed by him (6) that the islands were lymphoglandular (points follvlculatres) structures.

Entirely opposed to the above theory of development is that of v. Hansemann (7), who states that the islands originate late in embryonic life by a proliferation of the connective tissue cells of the stroma. The capillaries of the stroma widen and the adjacent cells become richer in protoplasm; the vessels assume a glomerular form and the cells an investing arrangement similar to that of the perithelium of blood-vessels. He finds no relation between the islands and acini, but from the intimate association between the island and blood-vessels and lymphatics, he thinks the islands are concerned in some exchange of substance between the blood and lymph.

Further detailed studies of the development of the islands of Langerhans of the human pancreas I have been unable to find. Difierent writers mention their presence at various periods of embryonic life, thus Gartier (8), at. three months, Ssobolew (9), at six months, and Stangl (10), at about seven months, while Kasahara (11) merely states that they are numerous in the foetus.

Embryology of the Islands

My study of the development of the islands has been carried out on the pancreases of twenty-one human embryos. Some of the material was poorly preserved and unsatisfactory for detailed study, but fortunatelythe better preserved material supplied the most important phases of the diiferentiation of the islands. The process of development which I have observed would seem to be so satisfactory that I have not deemed it necessary to wait for more material before the publication of my results. The study was carried out on serial sections, the greater part of the material having been imbedded ir paraffin, but some were imbedded in celloidin. The sections were stained either with haematoxylin and eosin or with safranin and picric acid. The following list indicates the source of the material, and the length and probable age of the embryo. The length given is that from breech to vertex. The method of computing age is that recommended by Mall (12). Series from a catalogued collection are indicated by a number in Roman characters placed after the name of the possessor.

Number Source Length (mm) Age (probable)
1. Prof. Mall (No. 22), Baltimore 20 47
2. Prof. Mall (No 45), Baltimore 28 58
3. Dr. Longcope, Philadelphia 54 73
4. Dr. Longcope, Philadelphia 60 73
5. Prof. Mall (No. 44), Baltimore 70 84
6. Prof. Mall (No. 23), Baltimore 70 84
7. Prof. Mall (No 34), Baltimore 80 89
8. Marchand, Leipzig 90 94
9. Prof. Mall, Baltimore 94 97
10. Marchand, Leipzig no measurement 3 months
11. Dr. Longcope, Philadelphia 115 115
12. Prof. Mall, Baltimore 116 116
13. Dr. Longcope, Philadelphia 127 127
14. Prof. Mall (No. 48), Baltimore 130 130
15. Dr. Longcope, Philadelphia 145 145
16. Dr. Longcope, Philadelphia 160 160
17. Dr. Longcope, Philadelphia 160 160
18. Prof. Mall, Baltimore 170 170
19. Dr. Longcope, Philadelphia 200 200
20. Prof. Mall, Baltimore 205 205
21. Marchand, Leipzig 210 210
Data Source: Pearce RM. The development of the islands of Langerhans in the human embryo. (1903) Amer. J Anat. : 446-455.

The earliest stage studied (Nos. 1 and 2) was that in which the glands are represented by branching processes widely separated by a rich framework of connective tissue. The connective tissue is about equal in amount to the glandular portion of the organ, and is the most striking feature throughout the early stages of development. It disappears slowly but even in the fifth and sixth months of foetal life still arrests the attention. Its presence is of great assistance, for it marks off the anatomical units of the pancreas in a way that greatly facilitates their study. Each group of cells is thus distinctly separated from neighboring groups, rendering the study of such a group in serial sections possible and easy. This anatomical condition made it very easy, as will be shown, to trace the relation of an island to its acinus. In this early stage it was not possible to find any arrangement of cells suggestive of the formative stages of the islands, nor was it possible to find the peculiar cells with rich eosinophilic protoplasm which Laguesse and Renaut compare to the parietal or oxyntic cells of the gastric tubules, and which they consider to be the earliest diiferentiation of cells destined to form islands.

In the next preparation (No. 3), from an embryo 54 mm. in length, the early development of the islands can be easily seen. The freshness and excellent preservation of this specimen adapted it for a most careful and accurate study of the relations and differentiation of the cells. The glands are still represented by branching processes of cells, lying in an abundant stroma, but here and there a few processes which have become tubular are seen. The cells are closely crowded together, especially at the periphery or growing portion of the process. They have rather deeply staining, round or oval vesicular nuclei, and a very small amount of slightly granular protoplasm. The protoplasm is readily demonstrable in those cells forming tubular processes ; the nuclei are seen at the periphery, while the clear protoplasm at the attachment of the cells to the basement membranes forms a ring about the lumen. Here and there, and generally at the periphery, in either the solid or tubular processes, is occasionally seen a much larger cell with deeply staining chromatin and either with clear or eosinophilic protoplasm, but I have not been able to convince myself that they represent the primary differentiation leading to the formation of the islands. They always occur singly and the arrangement of the chromatin is suggestive of karyolcinesis. Entirely distinct from them are the small groups of cells lying at the side of a glandular process, but in direct continuity with it (Fig. 1). Each group is composed of ten to fifteen cells which have a round, lightly staining nucleus, and a comparatively large amount of finely granular protoplasm staining deeply with eosin. The cells are closely applied one to the other and form round or oval masses which can readily be distinguished from the aeinar process from which they arise. Sometimes they lie in a semilunar projection of the gland without distinct connections; but in following the group through a series of sections, the continuity with the gland can generally be demonstrated. The masses are generally distinctly separated from the surrounding connective tissue by a narrow space, which gives the effect of a capsule. This differentiation of cells represents, I believe, the earliest stage of the development of the islands, and is identical with that described by Laguesse in the sheep, and later by Renaut in the human embryo.

Alt=Human embryo 54 mm pancreas

Fig. 1. Pancreas of an embryo 54 mm in length. Early stage of the differentiation of an island of pond to those described by Langerhans. A round mass composed of cells rich In protoplasm is seen continuous with the periphery of the gland.

In another preparation (No. 4) of about the same age, although the preservation is also perfect, this differentiation is not so distinct. The study of this preparation added little to the above description.

Very little information was obtained from a study of preparations 5, 6 and 7. The sections had evidently been prepared for other purposes than the study of such minute details as this investigation required. The general conformation of the acinar processes can be readily made out, and in places groups of cells suggesting those described above are seen; but a study of their finer characteristics is impossible.

In preparation No. 8 (embryo of 90 mm.) the tubular character of the acini is more prominent, and the solid masses of cells representing the primitive islands stand out distinctly by contrast. The islands are much larger than in No. 3, being made up of from twenty to forty or more cells, and are still in close relation to the acinar processes from which they originate. The tendency of their protoplasm to stain intensely with eosin is more pronounced than in the earlier periods. In this specimen the islands exhibit the first stage of vascularization which is shown by the presence of two, three, or four red blood corpuscles lying here and there between. the cells. Occasionally in large islands a small capillary vessel filled with red blood corpuscles protrudes, generally from the acinar side, towards the center of the island. The cells of these larger islands show a tendency to form the intercapillary columns or groups characteristic of the later stages of development. About the islands the reticulum is somewhat differentiated so as to form a distinct capsule composed of one or two layers of cells. Besides this more advanced stage of development, the early stages described in No. 3 are still met with. In this early period no islands were found in the head of the pancreas, which is of interest in View of the observation of Opie (15) that in the adult, the islands are most abundant in the tail and body of the organ.

Preparation No. 9 shows greater vascularization but does not differ otherwise from that just described.

Preparation No. 10 is a perfectly preserved pancreas from a foetus believed to be of the third month; it presents many points of interest. Numerous primitive islands are scattered through the tail and body, while for the first time, a few are seen in the head. These last, however, represent the very early stages of development. Although most of the islands are in intimate connection with the acinus, an arrangement is frequently seen which indicates the manner in which the island eventually becomes separated from the acinus. As illustrated in Fig. 2, an island may be almost entirely separated from the acinus; the only evidence of continuity being a solid stalk-like process of cells. In the acinar portion of this process, a gradual transition of the cells from the type of the gland to that of the island may be traced. Most of the cells in the process are identical almost with those in the island. Followed through a series of sections, such an island has no connection with the acinus other than this solid process. The connecting process I is in some instances short and broad, in ‘ others long and narrow; occasionally it A may be constricted by the surrounding tissue as though about to be completely separated. Indeed, in this series completely separated islands are seen in the splenic portion for the first time.

A careful study of this stage of the development has convinced me that the ‘ separation of the island is brought about by an encroachment of the connected with the gland acinus by 9. tion of the connecting cells and their 9°11“ P1‘°°°33 0‘ 00113final disappearance between the island and acinus. The island then lies free to one side of the primitive acinus in a mass of connective tissue; a condition just the reverse of that in the adult pancreas. Later, in the fifth and sixth month, when the rapid development of the acinus occurs, the glandular elements surround and enclose the island, and it then occupies the center of the lobule. The isolated appearance of the islands at this period of separation recalls and seems to support v. Hansemann’s view that the islands develop from the cells of the connective tissue. Without the knowledge of the earlier stages of proliferation and temporary connection with the acinus which I have described, their isolated position would indeed be inexplicable. From v. Hansemann’s description, I cannot discover that he studied embryos of a period corresponding to that represented by No. 3 of my series. The advanced vascularization at this stage affords evidence of the manner in which the vessels enter the island. Unlike the glomerulus of the kidney, to which the island is somewhat analogous owing to its very rich capillary network, we have not one, but several afferent as well as efferent vessels. In those islands still connected with the acinus, a vessel does not accompany the connecting process of cells. By following an island through a series of sections, from four to seven branches, Without definite arrangement however, may be seen entering from the periphery. The vessels at the periphery anastomose freely and form a network much richer than that about a glandular structure of equal size. At this period the cells between the capillary network of the island assume a definite arrangement in columns or rows.

Alt=Human embryo 54 mm pancreas

Fig. 2. Pancreas of an embryo of about the third month. A fairly well developed island of Langerhans is seen connected with the acinus buy a solid process of cells.

connective tissue, causing an attenua-

If the development here described is compared with that observed by Laguesse and by Renaut, an agreement is found only in regard to the method of primary differentiation. These early investigators failed to observe the stage of the process in which the solid column of cells eonnects the island with its acinus and therefore failed to note the gradual separation leading to final isolation of the island.

The changes just described represent the last important phases in the differentiation of the islands. 1n the period represented by preparations 11, 12, 13, and 14, the increase in size, the progressive vascularization, and the appearance of a fine reticulum along the vessels may be studied. Occasionally an island may be seen in continuity with its acinus, but this appearance is now unusual. In preparations 13, 14, 15, 16, 17, and 18, the glandular elements may be seen gradually surrounding the islands, while in preparations 19, 20, and 21 this process is completed and the island is seen in the center of its acinus, the position it occupies in the pancreas of the adult.

A number of specimens of the pancreas of children born dead during the last month of pregnancy and at term, as well as of infants surviving a few hours or days, have been examined. Except for slight diiferences in the amount of granulation of their cells, the islands appear identical with those of the adult. Specimens of this period, the blood-vessels of which are injected with Carmine, show the extraordinary vascularity first described by Kiihne and Lea (17), but which is not more prominent than may be seen in the adult pancreas.

Syphilitic Pancreatitis of the New-born

Confirmatory evidence of the development above described has been furnished by a study of the changes occurring in that pancreas in congenital syphilis. This condition, first accurately described by Birch—Hirschfeld, consists of extensive sclerosis of the pancreas with simple atrophy of the glandular structures. The atrophy in the advanced stages is so extreme that the larger ducts, the islands, and a few fragrants only of glandular tissue remain.

The persistence of the islands in contrast to the extensive atrophy of the glandular structure is a striking feature of this lesion. In Birch Hirschfeld’s description the islands are not mentioned. Schlesinger, (14), Ssobolew, (9), and Opie, (15), however, call attention to their persistence in even the most advanced examples of sclerosis.

I have studied the pancreas of congenital syphilis for the purpose of obtaining evidence of arrested development of the islands. The increase of connective tissue isolates these structures so distinctly from the remnants of glandular tissue that their relations may be readily studied. In six cases examined, I found frequently a persistence of the solid mass of cells connecting the island with the acinus.‘ I believe that this appearance represents the arrested development of the island at the period corresponding to preparation No. 10. At this " ; a period (third month) the amount of connective tissue in FIG. 3. s philitic pancreatitis in a foetus of the seventh month. Arrested development of the DOTUI31 Pancreas 15 Very the island; the persisting solid process of great, and the rapid proliferation of the glandular portion of the pancreas characteristic of the fifth and sixth months has not commenced.

Syphilis of the pancreas has been observed as early as the fifth month (Muller, 16), but there are no observations to prove that it does not occur earlier. The increase of connective tissue, which is the characteristic lesion in the pancreas, causes not only an atrophy of existing glandular structures, but what is more important, prevents the further development of the gland. The glandular tissue is represented by small, irregularly scattered gland groups; the ducts are prominent; the islands are arrested at the stage in which they are still connected with the acini. The microscopic picture of the normal pancreas at the end of the third month is very similar to the syphilitic pancreas at the sixth or seventh month, except that the latter has a larger proportion of connective tissue. In other words, at the latter period, we have the development of the third month plus the connective tissue produced by the syphilitic process. This arrest of development i.s shown in Fig. 3, which is taken from a specimen of syphilitic pancreatitis in a foetus 21 cm. long.

  • 1 0pie, in his account of two cases of congenital syphilis, describes and pictures this condition.


The islands of Langerhans (embryo of 54 mm) originate through a proliferation and differentiation of the cells of the primitive secreting tubules. The differentiated cells characterized by a rich, finely granular, eosinophilic protoplasm lie as small round or oval masses in direct continuity with the cells of the tubule. Later (embryo of about the third month), the attached portion becomes constricted, and lengthening, forms a stalk-like, solid process of cells connecting the island with the acinus. At this period, a few entirely isolated islands are present. A separation takes place and apparently is brought about by the pressure of the investing connecting tissue. Vascularization has now occurred. In still later stages a progressive vascularization, increase of cells, arrangement of the cells in columns, and appearance of a fine reticulum are observed. The rapidly forming glandular structures finally surround the islands which then occupy the centers of the lobules.

In syphilitic pancreatitis of the new born, a condition in which the normal development of the pancreas is arrested by a rapid proliferation of connective tissue, confirmatory evidence of this mode of development is supplied by the presence of solid processes of cells connecting the islands and acini.

The demonstration of the differentiation and final independence of the islands of Langerhans as here given offers an anatomical basis for the theory, so strongly supported by pathological and experimental observations, that the islands have at physiology independent of the glandular portion of the pancreas.


1. LAGUEssE.—Jour. de 1’Anat. et de la Phys., 1894, XXX, 731; 1895, XXXI, 475; 1.896, XXXII, 171, 209; Verhandl. der Deutsch. Anat. Gese11sc11., 1897, XI, 43.

2. LAGUESSE.—C. R. de la Soc. de Biol., 1889, LI, 900.

3. DIAMARE.—InteI'na.t. Monatschr. f. Anat. u. Phys., 1899, XVI, 155, 177; Anat. Anzeiger, 1899, XVI, 481. Quoted by Oppel, Lehrb, (1, Ver. gleich. Mikr. Anat. d. Wirbeltiere, 1900, III, 800.

4. IWASSARI.--Rend. R. Accad. dei Lincei., 1898, VII, 134. Quoted by Oppel. 5. RE;NAU'r.-—Traité d’Histo1ogie~ Pratique, 1899, II, 1523, 1543. 1.0. 11. 12. 13. 14. 15.

RENAU'I‘.——C. R. de 1’Acad. des Sciences, 1879, LXXXIX, 247.

v. HANsmMANN. — Verhand1. der Deutsch. Path. Gese]1sch., 1901, IV, 187.

GARTIER.-—'. 1')issert., St. Petersburg, 1900. Quoted by Ssobolew, V. A., 1902, CLXVIII, 91.

SSOBOLEW.-—Ce11t1‘alb. f. Allg. Path. 11. Path. Anat., 1900, XI, 2013; V. A., 1902, CXLVIII, 91.

S'rANGL. — Wiener klin. Wochenschr., 1901, XIV, 964.

K.ASAHA.'RA.—V. A., 1896, CXLIII, 111.

MALL. — Johns Hopkins Hosp. Bull., 1903, XIV,

29. BIRCH-HIRSCI-IFELD.—Arch. der Heilkunde, 1875, XVI, 166.

Som.JcsINGmn.—V. A., 1898, CLIV, 501.

OPIE. — Jour. of the Boston Soc. of Med. Sciences, 1900, IV, 251; Jour. of Ex. Med., 1901, V, 397; Johns Hopkins Hosp. Bull., 1900, XI, 205.

Mt'ILLEn.—-V. A., 1883, XCII, 532.

K1'.'1mm and LEA.—UnteI'S11ch. aus dem Physiol. Institut. d. Universitfit Heidelberg, 1882, II, 448. Quoted by Oppel.

Table 1 as image

Pearce1903 table1.jpg

Cite this page: Hill, M.A. (2024, May 28) Embryology Paper - The development of the islands of Langerhans in the human embryo (1903). Retrieved from

What Links Here?
© Dr Mark Hill 2024, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G