Paper - The fate of the graafian follicle in the human ovary
|Embryology - 15 Apr 2021 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)
Shaw W. The fate of the graafian follicle in the human ovary. (1925) J. Obst. and Gynaecol. 679-689.
See also by this author - Shaw W. Ovulation in the human ovary - Its mechanism and anomalies. J Obstet. and Gynaecol. (1927) 34(3):
|Historic Disclaimer - information about historic embryology pages|
|Embryology History | Historic Embryology Papers)|
The Fate of the Graafian Follicle in the Human Ovary
By Wilfred Shaw, M.A., M.B., B.Ch.(Cantab), F.R.C.S. (Eng.),
Chief Assistant, Gynacological Department, St. Bartholomew's Hospital.
The foetal ovary at term is studded with primordial Graaﬁan follicles, and as the ovarian stroma is ill-developed at this phase the primordial follicles constitute the main bulk of the ovarian cortex. Their number has been estimated by Marshall to be 100,000 while Stevens gives the figure 10,000. lip to the present time a fair amount of attention has been paid t.o those follicles which rupture and shed their ova, but such follicles represent only a small proportion of the total number of follicles present at birth, and very little work has been done to determine the fate of the other follicles. There is reason to believe that usually only a single follicle ruptures each month during the child-bearing period, yet there is no evidence to show that ovulation occurs before puberty, and after the menopause Graaﬁan follicles have disappeared. If the child-bearing period is considered as the 36 years between the ages of 14 and 50, it follows that approximately only 500 follicles can be accounted for by the processes of ovulation and corpus luteum formation, and that the vast majority of the primordial ova present at birth have some other fate. An example is here afforded of what Starling describes as the prodigality of nature in the perpetuation of the type, for the young female starts life with an ample reserve of sexual cells.
It is the purpose of this paper to describe the results of investigations on young and degenerating corpora lutea, and to give an account of the atretic processes occurring in follicles which do not rupture. It appeared important that a description of these normal changes should be given, for variations occur in pathological conditions of the generative organs which cannot be distinguished unless the normal histology is known.
The material used was that employed in my work on the estimation of the date of ovulation, and on the relation of the corpus luteum to the premenstrual changes of the endometrium. The ovaries were obtained from patients who were operated upon for various pelvic conditions. They were ﬁxed in neutralized formalin and subjected to a complete histological examination. In this way an excellent opportunity was afforded of investigating the varieties of atretic processes occurring in degenerate Graaﬁan follicles.
Development of the Graafian follicle
The ovary is derived from the genital ridge of the intermediate cell mass. According to Beard, the primordial ova are found at a very early stage of development of the embryo, and migrate to the genital ridge. The neighbouring coelomic cells—Keith’s mesothelial cells—together with the primordial ova, constitute the germinal epithelium. VValdeyer‘~’ showed that by the. upward growth of the mesoblastic tissue beneath this layer columns of cells derived from the germinal epithelium and containing primordial ova were produced. At a later stage these columns become separated from the ge.rminal epithelium, and in this way the primordial ova, together with the surrounding mesothelial cells, which subsequently became differentiated to form the granulosa layer, attain their position in the ovarian cortex.
About the 36th week of intrauterine life the process of follicularization is first seen. A cavity containing clear translucent fluid—the liquor folliculi—appears in the granulosa layer, which partly separates the ovum from the wall of the follicle, although at one spot the ovum is attached to the peripheral ring of granulosa cells by a stalk, the discus proligerus, and is itself encircled by a layer of granulosa cells. At the same time two layers of stroma cells become differentiated to enclose the follicle—an inner layer of swollen cells, the theca interna or tunica propria, and an outer layer of stroma cells and vessels, the theca -externa or tunica ﬁbrosa. The structure formed in this way is a Graaﬁan follicle.
Before puberty, although a minor degree of follicular ripening frequently occurs, no ovum is shed and no corpus luteum is produced. The majority of follicles formed before puberty undergo atresia, but a state of dynamic equilibrium exists and other primordial follicles mature and replace them. There is no evidence to show that a further development occurs, and no ovum is shed, until puberty is reached.
Development of the Corpus Luteum
It is important to realize that the process of ripening occurs before rupture of the follicle and ovulation ensue. The features of this process of ripening can be seen in specimens of ovaries obtained in the postmenstrual phase, for at that time several. usually as many as ciglit or ten ripening follicles can be found. The size of the follicle increases through the distension of the cavity with liquor folliculi, and the granulosa cells proliferate and hypertrophy, The most marked change is seen in the cells of the theca interna layer. Extreme hyperaemia occurs, and the theca interna cells increase in bulk and in number. They assume a brown colour, and their protoplasm becomes granular. The whole follicle approaches the surface of the ovary. Strassmann3 has indicated how‘ this mechanism is brought about. The zone of theca interna cells on the side of the ovarian cortex proliferates eccentrically towards the surface, burrowing into the ovarian stroma and determining the direction in which growth of the ripening follicle takes place.
Further, the site of the attachment of the ovum to the granulosa layer by means of the discus proligerus changes. While in the -early stages of ripening the cuniulufs is directed towards the ovarian medulla, at the -end of this stage the cumulus has rotated to lie against the cortex.
After ovulation has occurred the follicle is converted into a corpus luteum. I have had the good fortune to obtain three specimens of very early corpora lutea, and as such specimens are rare and of great importance in determining the origin of the lutein cells a short acount will be given of the development of the corpus luteum.
The problem of the origin of the lutein cells of the human corpus luteum cannot be investigated with the precision necessary for most scientiﬁc determinations, for material is always very limited and extremely difficult to obtain. Theoretically, the problem can only be solved with mathematical accuracy by the study of a large series of specimens of known dates, and this ideal is not likely to be attained. It follows that at the present time most of the evidence is indirect. Two views hold the ﬁeld, that put forward by von Baer, who attributed the origin of the lutein cells to the theca interna cells, and that of Bischoff, who considered the granulosa cells to be responsible. it will probably be admitted by all, that a great deal of the work of the early investigators was incomplete, for their specimens were not accurately dated. Further, a great deal of confusion existed between the atretic follicle and the young corpus luteum. At the present time the problem must be considered from a very broad aspect, taking into consideration work done in a variety of different ﬁelds.
1. Comparative Morphology
It has been shown that in the rabbit ovulation takes place about nine hours after coitus. In this way, as the time of ovulation is known, a progressive series of specimens of early corpora lutea can be obtained if animals are killed at suitable intervals after coitus, and consequently the mode of development of the corpus luteum can be followed. For animal work this technique is the only one capable of accuracy. It was first emplowed by Sobotta,4 who Showed that in the mouse the large. lutein cells of the corpus luteum were derived from the granulosa layer. Marshall,5 following a similar technique, conﬁrmed this view for the sheep.
2. Fate of the thaw interna layer in the development of the corpus luteum
Van der Stricht,6 working with the ovary of the bat, pointed out that in the corpus luteum, in addition to the large lutein cells, another series of cells can be seen. These cells are found mainly at the periphery, are smaller in size than the large lutein cells and have different staining properties. Van der Stricht succeeded in demonstrating these cells in the human corpus luteum, and his observations were conﬁrmed by Bijhler, Seitz and others. The term para-lutein cells has been applied to them by Pinto and American authors. At the. present time I have examined some 50 specimens of early or recently retrogressive human corpora lutea. and I have never failed to ﬁnd this peripheral layer of cells. It seems clear that in the developed corpus luteum two layers of cells are represented, for there is no reason ‘to believe that the para-lutein cells are young forms of the large lutein cells. This would indicate that two of the layers of the ripening Graaﬁan follicle take part in the formation of the corpus luteum, and as no change is seen in the theca externa, the probability suggests itself that the large lutein cells are derived from the inner granulosa layer, while the smaller peripheral ring of para-lutein cells develope from the cells of the theca interna.
3. Specific staining methods
It was suggested by Miller,7 that as van Gieson’s stain possesses. the property of staining epithelial cells yellow and connective-tissue cells red, and as the lutein cells of the corpus luteum assume a yellow colour with van Gieson’s stain, the large lutein cells were probably epithelial in origin, and consequently arise from the granulosa layer of the follicle.
Differentiation can be obtained between the para-lutein and large lutein cells by appropriate staining methods. For example, it has been shown, by Solomons and Gatenbys and others, that after osmium tetroxide staining the two types of cell can be readily distinguished. It is not necessary to employ osmium tetroxide for Scharlach R, Sudan Ill, and Nile blue sulphate also bring out this differentiation, for at any phase of the menstrual cycle the cells of the two layers vary in their fat content. VVith paraﬁin imbedde.d sections, although with a little experience the cells can be distinguished fairly easily with the usual stains, Twort’s neutral red-light green method brings out the distinction very beautifully. It is probable that the para-lutein and lutein cells are morphologically distinct, for this difference in their staining properties would be difficult to explain in any other way.
4. The character of early human forms
The three specimens of very early corpora lutea that l have obtained display the early changes which occur in the formation of the corpus luteum. In the main the processes which take place in the ripening follicle continue. The extreme hyperaernia of the theca interna layer becomes better marked; capillary tufts project towards the grannlosa layer, and afterwards assist in the production of the convolutions of the corpus luteum. Very little. hypertrophy of the theca interna cells occurs, beyond the stage they have reached in the ripening follicle, and in very early forms of the corpus luteum a faint lipoid reaction .can be demonstrated in these cells. A capillary network becomes well-developed between the granulosa and theca interna layers which serves to produce a sharp differentiation between them. The greatest change is seen in the cells of the granulosa layer. Extreme hypertrophy takes place, and the cells soon attain a great size (Fig. I). As the theca inte.rna and the surrounding ovarian stroma present a dense unyielding periphery the development is centripetal, and results in the appearance of convolutions. This distortion is aided by the capillary tufts already mentioned, which arise from the theca interna, and which serve to vascularize the lutein layer. The form of these young specimens indicates that the granulosa cells give rise to the large lutein cells of the corpus luteum.
Proliferation of the corpus luteum continues until the 19th day of the menstrual cycle, i.e. 19 days after the beginning of the last menstrual period, and in ﬁve specimens of proliferating corpora lutea corresponding to the phase between the 16th and 19th days which I have now obtained the theca interna and granulosa-lutein cells can be distinguished. In the mature corpus luteum the theca interna cells can be identiﬁed as the para—lutein cells. They are found at the periphery, and enter into the septa between the convolutions (Fig. 2). Usually they are well differentiated from the large lutein cells, and I have never found them scattered amongst the large cells of the lutein layer. In the human corpus luteum they form a small almost negligible portion of the bulk of the corpus luteum. The histological form of the mature corpus luteum indicates that the main proportion of the internal secretion produced by the corpus luteum is derived from the large lutein cells. Moreover, in the mature corpus luteum of the late stage of the menstrual cycle, fatty acids can be found in the theca interna cealls. The comparison of the forms of young and mature corpora lutea, therefore, shows that the main part of the corpus luteum is derived from the granulosa layer, and that but a negligible portion obtains its origin from the theca interna layer.
Degeneration of the corpus luteum of menstruation
Degeneration of the corpus luteum of menstruation is Seen on the day prior to tl1e menstrual discharge. Organization accompanies this degeneration, and the two processes advance until the corpus luteum has been replaced by a structureless hyaline body, the corpus albicans. Organization occurs rapidly in the first ten days, but slows down afterwards, and it is not until several months have elapsed that complete replacement by hyaline tissue has occurred. The early signs of degeneration are as follows: Fatty acids are seen in the cells of the theca interna layer, and in the large lutein cells neutral fat now appears, and with paraffin imbedded sections the lutein cells are vacuolated and have lost the characteristic old-gold colour with van Gieson’s stain. Most important of all, however, is the deposit of hyaline tissue between the cells of the lutein layer. This tissue is well brought out either with l\rla1lory’s or van Gieson’s stain, and represents the most certain of the signs of organization of the corpus luteum. It is doubtful whether its origin is from the ﬁbro-blasts which invade the corpus luteum from the periphery, for it is difficult to trace any connexion between these ﬁbro-blasts and the hyaline tissue. It is possible that it is deposited from the capillary network of the lutein layer.
Within the cavity of the corpus luteum, adjacent to the lutein cells a lamina of hyaline tissue is deposited. This hyaline tissue differs from fibrous tissue in that it is not ﬁbrillar but homogeneous and not nucleated (Fig. 3).
During the stage of retrogression it is common to ﬁnd blood in the cavity of the corpus luteum. It is rare to ﬁnd haemorrhage in the cavity of the young corpus luteum, and there seems to be no doubt that the old view, that bleeding occurs into the cavity as the result of rupture of the follicle, is erroneous. On the other hand, haemorrhage can be seen inthe cavity of young corpora lutea in certain hyperzernic conditions of the ovary such as oophoritis, so that the presence of blood in the cavity is no evidence of degeneration, although it occurs most typically after retrogression has commenced.
The next point to consider is the colour of the corpus luteum. The mature corpus luteum before menstruation has started is grey. It is not until degeneration has begun and until fat appears in the corpus luteum that the vivid yellow colour appears. Unless this fact is appreciated the most yellow corpus luteum may be mistaken for the most recent corpus luteum present in the ovaries. The yellow colour is almost certainly produced by the fat contained in the lutein cells in their state of retrogression.
The degenerative changes described above continue hyaline tissue is deposited within the cavity and between the cells of the Iutein layer. The Iutein cells are compressed by the hyaline tissue, and undergo atrophy. More fat appears in them, with the result that with parafﬁn imbedded preparations the cells are completely translucent. The wavy convoluted form persists, but the bulk of the corpus luteum diminishes. The outer border of the lutein layer tends to be irregular at this stage, and the fattily degenerate lutein cells radiate among the neighbouring cells of the ovarian stroma.
In this way the corpus luteum is converted into a hyaline body with a convoluted outline. The cavity persists, but is greatly reduced in size, and contains a few strands of fibrous tissue. The lutein layer is. replaced by a thick tortuous ring of hyaline tissue. The structure produced in this way is the corpus albicans. It diﬂers from other atretic structures, to be described later, through the thickness of the h yaline layer and the persistence of the cavity (Fig. 4).
The period that elapses between the onset of degeneration of a corpus luteum and the formation of a fully-organized corpus albicans is capable of an approximate estimate in this way. With the technique I followed in determining the date of ovulation1 it was possible to calculate the number of old degenerating corpora lutea in the two ovaries of the cases investigated. If it is assumed that only one follicle ruptures each month, it follows that the total number of de-generating corpora lutea present in the two ovaries will give the number of months required for a corpus luteum to become converted into a corpus albicans. I calculated that on an average seven to ten months elapse before organization is complete. Incidentally this technique enabled an estimation to be made of the division of corpora lutea between the two ovaries. On an average it appeared that the distribution was even, and that usually ovulation occurred alternately from the two ovaries. This rhythm was not invariably regular, not more so than could be accounted for by the law of chance. Recently Riihl,1 working in Aschoff’s laboratory with a similar technique, has obtained approximately the same results.
There are two otheripoints about the retrogressive corpus luteum that require mention. The ﬁrst is the not infrequent presence of pigmented cells in the cavities of corpora albicantia. They are seen most frequently in hyperaemic ovaries and are easily recognized by their brown colour. They probably represent connective tissue cells engorged with pigment derived from the blood which was extravasated into the cavity of the corpus luteum from which the corpus albicans was derived.
The second point is the persistence of large fattily degenerate cells at the periphery of fully—developed corpora albicantia. These cells contain fatty acids, and tend to penetrate among the cells of the ovarian stroma. It is possible that they represent old paralutein cells (Fig. 4).
Degenerate forms of Follicles which have not undergone Rupture
It has been pointed out above that in the post-menstrual phase several —usually as many as eight or ten— ripening follicles can be demonstrated in the two ovaries. It is the general rule that only one of these ruptures and sheds its ovum. Exceptionally more than one may rupture, and then corresponding corpora lutea are found. My own ﬁgures are small, but I have met with the condition in three cases out of some 50 that l have examined. This point conﬁrms the views usually expressed as regards the etiology of binovular twins.
The ripening follicles which do not rupture undergo atresia. There is a certain amount of evidence to show that the rupture of a ripening follicle is inhibited by a young or mature corpus luteum. For example, Seitz” showed that during pregnancy, although follicle ripening occurs, ovulation does not ensue, and that the ripening follicles become atretic. Again, it has been shown by Pearl and Surface"-3 on fowls, and later by llermann and Stein” on rabbits and rats, that ovulation is inhibited by injection of corpus luteum extracts. Consequently it appears probable that the corpus luteum produced from the first follicle to rupture inhibits further ovulation, and the remaining ripening follicles undergo atresia. If two follicles rupture simultaneously, two corpora lutea are formed, and if the two ova are fertilized binovular twins result.
Up to the present. time very little attention has been paid to the phenomenon of follicular atresia. It is a subject of very great complexity, and, although it has not such theoretical interest as the production of the corpus luteum, it is important from the aspect of the origin of the interstitial cells of the ovary. In the atretic processes ensuing upon follicular ripening the following changes can be seen: The proliferation and hypertrophy of the theca interna cells continue and the cells assume the brown colour of lutein cells. The proliferation of the Cells of this layer is radial and eccentric, and the cells become scattered among the neighbouring ovarian stroma cells. In certain conditions, particularly in pregnancy as indicated by Seitz“ and in cases of uterine ﬁbromyomata as pointed out by C‘ol1n,14 this hypertrophy may attain a very considerable degree. On the other hand, the vascularity of this layer steadily diminishes, and fat can be recognized in the theca interna cells at a very early stage of the process of atresia.
Fig. 3. Degenerating Conrus Luteum. x 55 Fig L Early Corpus Luteum X :;(;0_ The lutein cells are translucent and structureless hyaline tlssue has been The cells of the granulosa layer are proliferating and some have already _d‘fP°_5‘ted b°t“:ee‘.1 them- A thmk rfng °f hyalme “S5119 has appeared reached a considerable size. The theca interna cells are much smaller and “"311” the °3V1tY adlacent to the lutem Cells are clearly differentiated from the granulosa layer. The dark protoplasm of the granulosa cells is characteristic of mature lutein cells.
Fig. 4. Corpus Albicans. x 60 The corpus luteum has been replaced by a thick shell of hyaline tissue which retains the general forn1 of the original corpus Iuteum. A few translucent lutein cells are seen at the periphery. The granulosa cells have atrophied and the granulosa layer has almost completely disappeared. A lamina of hyaline tissue has been deposited between the theca interna and granulosa layers. are large and have a radial arrangement.
Fig. 5. Atretic Follicle. x 400 The theca interna cells
Fig. 6. Conrus A'1‘Rl£'l‘lCUM. x 100 The granulosa cells have disappeared and the hyaline tissue has increased in thickness. A few theca interna cells still persist at the periphery. The convoluted form is well shown.
Fig. 7. CORPUS Cmnxcms. x 40 Collapse of the corpus atreticum has occurred. The structure resulting consists of thin bands of hyaline tissue matted together in an irregular manner. The hyaliue bands are thinner than in the case of a corpus albicans. No cavity is seen. Fig. 8. Conpvs FIBROSUM. X100 In this case the cavity of the original atretic follicle has been obliterated by the inward growth of the hyaline lamina. The result is the production of a solid hyaline body. I C1r.J. W2
of the ovarian stroma.
The granulosa cells undergo degeneration, fatty acids soon appear in their protoplasm, the cells shrink and at a later stage disappear. The most important feature of all the-changes or atresia is the appearance of a lamina of hyaline tissue between the theca interna and granular layers~ethe so-called glass membrane. It is readily recognized after staining with Mlallory’s or van Gieson’s stains, even in the early stages of the atretic process, and as atresia develops it increases in thickness. At tl1is stage the follicle is best termed an altrelic follicle, for, although degenerating, the general form of the Graalian follicle is retained (Fig. 5). It must be differentiated from a young corpus luteum first, through -the degeneration of its granulosa cells; secondly, by the appearance of the glass membrane; and, thirdly, by the centrifugal proliferation of the theca interna cells. At a later stage of atresia the ovum and the granulosa cells have disappeared, the liquor folliculi has been absorbed and the cavity of the follicle filled with a few connective-tissue cells. The structure now remaining consists of the hyaline ring surrounded by a layer of fattily degenerate theca interna cells which have a characteristic radial arrangement. The structure is commonly found in adult ovaries, and is best termed a Corpus Al1'eticum (Fig. 6).
As this structure has the form of an empty spherical shell, and is surrounded by dense ovarian stroma, it is unstable, so that it collapses and the opposite surfaces of the h yaline lamina come into apposition. Because of the original spherical shape, distortion occurs in the process of collapse, so that eventually an irregular hyaline body with wavy bands of hyaline tissue separated from each other by a few connective-tissue cells results. This structure, which is termed a Corpus Cand-loans (Fig. 7), is distinguished from a corpus albicans by the thinness of the hyaline bands of which it is composed, and through the absence of a central cavity. Now, as the size of the original atretic follicle varies so there is a variation in the size of the corpora candicantia in the ovaries. Some may equal the size of corpora albicantia, while others are small. Again, as in the corpus albicans, there is a tendency for fattily degenerate theca interna cells to persist at the periphery and to radiate amongst the ovarian stroma.
This usual method of atresia is sometimes modiﬁed. In certain cases collapse of the follicle does not occur. In these cases they are seen in the ovaries of patients suffering from adnexal inﬂamma tion and uterine ﬁbromyomata~—the main features of atresia, i.e. proliferation of the theca interna cells, atrophy of the granulosa layer and formation of the glass membrane, although present, are overshadowed by the irregularity of the development of the glass membrane. Usually in corpora atretica the inner edge of the hyaline lamina is regular, but in the cases now referred to thr hyaline tissue invades the cavity of the atretic follicle irregularly, so that the cavity becomes ﬁlled with liyaline tissue derived from the glass membrane. This structure is solid, and does not collapse, and has the form of an irregularly ovoid solid body with it.s centre formed of hyaline tissue, interspersed with a few nucleated connective-tissue cells. This atretic form of Graaﬁan follicle was described originally by Seitz and termed (ibvrpus Fibrosum (Fig. 8). It is found only rarely, but there is no reason to believe that it represents a pathological variety of the atretic process.
The last type of the liyaline bodies derived from the Graaﬁan follicle is the Corpus Restiforme (Fig. 9). This structure is found more frequently than any of the other degenerate forms of the Graaﬁan follicle. lt is much smaller than any of the types described above, and is derived from small follicles and even from primordial follicles by a similar atretic process to that responsible for the formation of a corpus candicans. A thin layer of hyaline tissue is deposited betvveen the theca interna and granulosa layers and autolysis of the ovum and granulosa cells follows. Later the theca interna cells disappear, and a thin wavy strand of hyaline tissue finally remains. Corpora restiformia are most numerous in the cortex of the ovary, but are also found in its medulla. They are described separately from corpora candicantia because of their smaller size, and because they may arise from primordial follicles. Again, only a single or double layer of hyaline tissue is present, the complicated form of the corpus candicans is never seen. Corpora restiformia arise most frequently during childhood; only rarely can their formation be folloxved in specimens obtained from patients after, puberty.
This array of degenerate forms of the Graaﬁan follicle may appear formidable, but each class has its own characteristics, and should, therefore, be considered separately. The nomenclature has been obtained from Aschoff’s textbook, for the individual types have long been recognized, although I am not acquainted with any accurate account of the process of follicular atresia.
It will be seen that in each type the atretic body is produced in a similar way, with the apparent exception of the corpus albicans. ln this case the ggranttlosa layer, instead of degenerating ﬁrst, proliferates to form the corpus luteum. Hyaline tissue, instead of being deposited between the granulosa and theca interna layers, is found between the cells of the lutein layer — also in the cavity of the corpus luteum — and gradually replaces the lutein cells, which eventually disappear. The difference between the atretic processes in the atretic follicle and degenerating corpus luteum, therefore, resolves itself into the additional mechanism whereby the previously hypertropliied granulosa layer‘ in the latter is replaced by hyaline tissue.
- Only a small percentage of the Graziﬁan follicles found in the ovary at birth undergo ovulation. The majority become atretic.
- The large lutein cells of the corpus luteum are derived from the granulosa layer of the follicles. The para-lutein cells develope from the theca interna layer.
- About eight months elapse for a corpus luteum to become converted into a corpus albicans.
- Numerous forms of atretic structures are derived from the Graalian follicle: they are, the Corpus atreticum, the corpus candiczms, the corpus ﬁbrosum, and the corpus restiforme.
1. Shaw, Wilfred. Journ. of Physiology, 1925, No. 3, I93.
2. Waldeyer. “ Eierstock und Ei,” 1870.
3. Strassmann. Arch. fiir Gyr-LEikol., I923, Bd. 119.
4. Sobotta. Armt. Anz., 1895, X, 482.
5. Marshall. Quar. Iourrr. Microsc. 501., I905, 49, I89.
6. Van der Stricht. Bull. Acad. Roy. Belgique, 1901.
7. Miller. Arch. fur Gymakol., 1914, 101, 568.
8. Solomons and Gatenby. Journ. of Obstet. and Gynecol. of Brit. Empir. Winter No., 19-24.
9. Twort. ]ou.rrr.. of State Medicine, 1924, xxxii, No. 8.
10. Riihl. Arch. far Gymikol., I925, 124, 1.
11. Seitz. Arch. fzlir Gy11Eik0‘l., I906, 77, 203.
12. Pearl and Surface. Journ. of Biol. Chem., 1914, xix, 263.
13. Herrmann and Stein. Wren. klin. Wochenschr., I916, xxix, 25. I4.
14. Cohn. Arch. frur Gynéikol., 1909, 87, 367.