New Developments in Ovarian Dynamics and the Law of Follicular Constancy

In view of the great significance of the problem of ovarian dynamics, it seemed necessary to continue the comparative experiments with ovarian fragments with and without simultaneous partial castration, under more exact conditions than was the case in our former experiments.

The experiments lasted from October 1923 to March 1924. Dr H. E. Voss assisted in the operations and dissections. Mrs D. Švikul and Miss E. Kropman made the histological sections, counted the number of ova and measured the follicles. Mr S. Vešnjakov made the photographic records. To all these associates warm thanks are due.

All the experimental animals were fully grown rabbits weighing from 1·7 to 2· 7 kg. In this they differed from the rabbits of the series in the previous papers. Several months before operation they were isolated, this being of great importance in respect to the formation of corpora lutea. The operation was performed in the experiments of group A (ovarian fragments with simultaneous partial castration) by the ventral method; in group B (ovarian fragments without partial castration) by the dorsal method. In group A one whole ovary and the greater part of the second ovary were removed, only an upper fragment remaining in the body. The removed ovaries were weighed. No ligatures were made on the side on which the fragment remained ; this precaution was taken in order not to interfere in any way with the blood supply of the remaining fragment. In group B, one ovary was left intact and the second ovary was cut into two fragments, a small upper fragment and a larger fragment below. The animals remained isolated for the whole duration of the experiment.

The nuclei of the young ova were counted in the ovarian fragments in serial sections 7·5 μ thick, as formerly done by Arai(6). The counting was done twice, once by Mrs Švikul and a second time by Miss E. Kropman. Both, figures are given in Table I. There was a significant difference only in case VI, where the number of ova was a small one. As already insisted upon (2), many ova are counted twice, but the error introduced by this can scarcely be of any significance. The size of the Graafian follicles was measured in the usual way with the help of an ocular micrometer. In the small fragments of groups A and B all the Graafian follicles were counted and measured. In the ovaries the maximal diameter of the follicles was measured and the large follicles only were counted. The diameters of the follicles were measured, including the theca interna; the method is not a very exact one, since the follicles are not always spherical. The diameter of the follicles can be also estimated by counting the sections of one and the same follicle. The results are given in Tables I to III.

Table I reveals two very striking phenomena. Firstly, the ovarian fragments in group A have an average weight of 43 mg., while the ovarian fragments of group B only average 15 mg. At first sight one might think that this indicates an hypertrophy of the ovarian fragments in those cases in which the total ovarian mass had been reduced, and a lack of hypertrophy in group B in which the total ovarian mass had remained unchanged. Indeed this was the case in all previous experiments made by numerous authors including ourselves who worked with partial castration in rabbits and cats. But we must emphasise that in this new series there was no pronounced hypertrophy as expressed by the total weight of the ovarian fragments. If we compare the calculated original weight (see Table II) of the ovarian fragments with the final weight, this becomes especially clear. It is conceivable that in case III an hypertrophy took place; but in case II there was no hypertrophy at all, and in case I there was a difference of only 10 mg. which seems too little to permit of any positive conclusion. There was a similar difference in case VI of group B where the original calculated weight of the ovarian fragment was 6 mg. and the weight found at the end of the experiment was of 16 mg. Indeed, half of the weight of this fragment might be due to the surrounding tissues (see Fig. 5). As there are normally differences in the weight of the two ovaries in the rabbit, no conclusions can be drawn from these data. It seems most probable that there was no hypertrophy of the ovarian fragment during $412$ months either in group A or in group B, though the ovarian mass was highly reduced in group A, where the original calculated weight of the fragments was only 1/15 to 1/5 of the total ovarian mass. No reason can be advanced, at present, to explain the fact that no pronounced hypertrophy took place in this series although, as we shall see below, the changes characteristic for ovarian fragments after partial castration actually occurred. This question is discussed below in greater detail.

The second striking phenomenon seen in Table I is the diminution in the number of young ova in the ovarian fragments of group A as compared with group B. In case VI of group B there was indeed a very small number of young ova. But this case seems to be rather exceptional; the fragment was very small and surrounded with much connective tissue; and it is possible to suppose that the fragment was being gradually absorbed. In both other cases in group B the number of young ova was from about 5000 to 11000 though the fragments themselves were small. In group A the number of young ova in the ovarian fragments was in three cases 2000 to 3000, though the fragments were twice to six times larger than those in group B. In case III the ovarian fragment though also several times larger than the fragments of group B contained only a very limited number of young ova. In this respect our former data on the behaviour of ovarian fragments are fully corroborated by this new series.

It is also of interest to compare the number of young ova in A and B proportionally to the weight of the fragment. We shall see below that there is a striking difference between groups A and B in follicular development, i.e. the ovarian fragments of groups A and B are qualitatively different. It is therefore necessary to compare the number of young ova proportionally to the original weight of the fragment, assuming that the fragments of both groups were qualitatively identical at the beginning of the experiment. In Table II the number of young ova found in the different fragments at the end of the experiment were calculated proportionally to the original weight of the fragment and the figures are given for 1 mg. of the original mass. The figures for group A range from 5 to 154; the figures for group B range from 423 to 1200. Even if we do not exclude the exceptional case VI (see above), the figures remain highly in favour of the conception that the number of young ova in an ovarian fragment is greatly diminished when the total ovarian mass is itself reduced.

It must not be forgotten that the young ova are not equally distributed in the cortex throughout the whole ovary ; by neglect of this fact serious errors can probably be introduced. The uniform results obtained in all the series of experiments in this and in the former papers make it very improbable, however, that at the beginning of the experiment the fragments of group A were all relatively poor in young ova, whilst the fragments of group B were all relatively rich.

We may now discuss another point of difference between the fragments of group A and group B, in the condition of follicular development. As seen from Table III the maximum follicular diameter in an ovary of group A at the beginning of the experiment was i-6 and the same maximum diameter was found in the ovarian fragments of this group. In group B the maximum diameter of the ovaries was 13 and the maximum diameter in the fragments 1·5. There was evidently no difference in the final stage reached in follicular development. There is, however, another very significant feature. The number of large follicles with a diameter not less than i-o mm. in the ovarian fragments of group A is several times greater than in the fragments of group B. Whereas the upper fragments in group B contained o to 2 similar follicles, the upper fragments in group A contained 5 to 10. It clearly follows from these data that follicular development in an ovarian fragment depends upon presence or absence of the remaining ovarian mass. The dependence of follicular development upon the quantity of ovarian mass present in the body is indicated also by the number of follicles between 0·5 and 1·0 mm. which have been counted in the ovarian fragments in groups A and B. The number of these follicles was also found to be strikingly greater in the first group. The number of follicles in the fragments of group A is no smaller than in the total ovarian mass of group B (see Table III). In case III the fragment contained a strikingly greater number of follicles than the ovaries of B, but if we deduct (in case III) the “blood follicles,” which were especially numerous, then the number of big follicles appears to be normal.

In case I the ovary (removed several months before the fragment) contained large follicles crowded together in such a manner that in this region there was no difference between the ovary and the fragment (Figs. 1 and 2). But it is clear that a normal ovary will reveal this picture if by chance most of the follicles which ripen are of the same region. In other regions the same ovary was quite normal.

From all the data mentioned above it seems clear that the condition of the ovarian fragment in partial castration follows the law of follicular constancy.

In former papers we mentioned the formation of follicular cysts in the ovarian fragment. There is no doubt that this phenomenon is of some importance since it was found both in ovarian fragments of cats (2) and incidentally in those of young rabbits (1), but never in those cases in which the total ovarian mass had not been reduced. Though there were no follicular cysts in the present series of experiments, the same tendency is indicated here also by the deviations in case III. This is the only case in which the total number of young ova was so reduced that the reserve was almost exhausted. This suggests that there is some link between marked diminution of young ova and formation of follicular cysts. It has been suggested formerly that the condition of the ovarian fragment after partial castration depends upon substances X, which control follicular development. Persistence of mature follicles or formation of follicular cysts may then be explained as due to loss of the normal balance between production of X in the body and its use in the ovary. This balance seems to be of importance and it is probable that it might be upset experimentally, as, for example, when an ovary is engrafted into a young castrated male.

The feminised male, as described by Steinach (7) and corroborated by Wang, Richter and Guttmacher(8) and by the writer (9), does not grow as quickly as a normal male. The ovarian graft in the male is characterised, as I have shown (10), by persistence of mature follicles, and I have suggested that a diminished rate of body growth in the young feminised male is caused by the X-substances available in the body being, in these abnormal cases, mostly used by the ovarian graft; the latter is in a state of protracted oestrus so that these X-substances are not available for general body growth.

The problem of follicular persistence and formation of follicular cysts as discussed above, and the problem of their dependence upon some deviation from the normal balance between production and use of X-substances, seems of interest also pathologically. The cystic degeneration of the ovary in all the different forms observed in human and animal pathology is possibly never of purely ovarian origin but rather caused by upset in the balance of production and use of X-substances. It must not be forgotten that this balance depends not only upon the ovary but also upon the body as a whole.

The compensatory hypertrophy of ovarian fragments also forms an interesting problem. For many years it has been generally accepted that a compensatory hypertrophy of the testicle or of the ovary takes place after partial castration in the mammal. But it has been shown by the writer that testicular fragments do not hypertrophy (11), and eventually the writer showed (12) that the greater weight of the testicle after unilateral castration is by no means due to hypertrophy but to accelerated growth of the one remaining testicle; the final normal testicular weight is more quickly attained by the remaining testicle without reaching a weight greater than that of one normal testicle of the adult animal. But contrary to this, the idea of an ovarian compensatory hypertrophy remained unaltered. All the investigators and the writer himself found that the ovary increases in weight after unilateral castration; Carmichael and Marshall(13) showed that the same is true for ovarian fragments, this being corroborated by the writer and his associates. But on analysing the data given in the former two papers on ovarian dynamics and in the present paper it became abundantly clear, that the characteristic feature in the condition of an ovarian fragment, after reducing the total ovarian mass, is not the increase in weight but the relatively intensified follicular development according to the law of follicular constancy; the total number of young ova present (if new ones are not formed) greatly diminishes and a normal number of large mature follicles is formed. Whether follicular cysts are produced and whether there is an increase in weight is of secondary importance and depends evidently upon an interaction of different individual factors. If we compare the ovarian fragments of group A with normal ovaries, we see that they contain the same number of mature follicles as the normal ovaries. The difference is only that in the fragments there are fewer young ova and less interstitial tissue. As the interstitial tissue in the rabbit’s ovary forms a considerable part of the whole ovarian mass, it is obvious that less interstitial tissue will mean also considerably less ovarian mass. This may explain why there was no pronounced increase of weight in this series as compared with fully grown rabbits. It is, in fact, interesting to consider why there was comparatively less interstitial tissue in the ovarian fragment of the fully grown rabbit. The phenomenon is most probably not without significance since the interstitial cells of the ovary (the importance of which has been emphasised by Steinach, by the writer and others) almost certainly play a most important part in the ovarian dynamics. This has been shown recently by Zondek and Aschheim(16) and especially by Parkes, Brambell and Fielding(17). In other cases in which there is a more pronounced cystic degeneration as in experiments with cats (2), the cysts themselves produce a great increase in weight, i.e. an “hypertrophy.”

In former papers I insisted that there is no hypertrophy of the testis after partial castration, an hypertrophy being simulated by accelerated growth (see above). The analysis of the ovarian dynamics after partial castration similarly shows that “compensatory hypertrophy” is not the real nature of the change seen in the ovary after its mass has been reduced. Since the number of primary follicles entering into follicular development remains normal, an ovarian fragment will necessarily increase its volume. Whether the weight of one or two normal ovaries is attained or not seems of no significance for the “integrative” processes going on in the body. The condition of the ovarian fragment after partial castration is not always identical with that of a normal ovary. This seems at first sight contrary to the law of follicular constancy. There is sometimes no “restitutio at integrum,” but probably only a new equilibrium after the normal balance between production of X-substances and their use is eventually disturbed when most of the young ova have disappeared. Evidently under certain quantitative conditions the follicular constancy is to a certain degree upset. How far sexual intercourse and gravidity can affect such process has been already examined in the experiments of Carmichael and Marshall (13), but it deserves further experimental attention.

As already stated, each animal used in the present series of experiments was isolated for several months before the operation and afterwards during the whole experiment. No corpus luteum was found by macroscopical observation of the ovaries removed in October and similarly of those removed in March. No corpora lutea were revealed by microscopical examination, with the possible exception of case VI in which several bodies derived from corpora lutea (? or from large atretic follicles) were found in the ovary. The fact that corpora lutea were lacking is confirmatory of Ancel and Bouin(14) and of Hammond (15).

In several cases blood follicles essentially similar to those described by Hammond were found. They could be recognised by the naked eye in the small ovarian fragment of case III where several such blood follicles were present. All the other large follicles were quite normal and contained normal ova.

The comparative condition of ovarian fragments has been studied in fully grown isolated rabbits, some of which possessed normal ovaries, whilst in others the normal amount of ovarian tissue had been previously reduced by partial castration.

No pronounced increase in weight of ovarian tissue was observed in cases of partial castration though the experiments lasted $412$ months; this is contrary to what has been observed in young rabbits.

The features characteristic of ovarian fragments in partially castrated hosts were quite definite: (1) the total number of primary follicles was greatly reduced as compared with ovarian fragments present in normal females; (2) the number of large follicles (with a diameter of not less than 1 mm.) in an ovarian fragment grown in a partially castrated animal was not less than that in both ovaries of a normal animal.

A diminution in the total number of primary follicles was observed also when calculated per mg. of the original mass of the fragment.

The maximum diameter of the follicles in the ovarian fragments in partially castrated rabbits was not greater than in normal ovaries.

These observations are in accordance with the “Law of follicular constancy.”

The deviations from the law of follicular constancy which were formerly observed in partial castration (persistence of mature follicles and formation of follicular cysts) are probably to be explained by a disturbance of the normal balance between formation and use of “X-substances “necessary for follicular development; this disturbance possibly represents a final stage in the life of an ovarian fragment after extensive use has been made of primary follicles.

The meaning of ovarian hypertrophy is also discussed. It is suggested that the integrative processes which take place in an ovarian fragment after partial castration are not to be characterised by increase of weight but only by those processes which are dictated by the law of follicular constancy.

Recently the work of P. E. Smith and E. T. Engle (The Amer. Journ. of Anat. 40, 159, 1927) and of B. Zondek and S. Aschheim (Arch. f. Gynäkol. 130, 1, 1927) came to my knowledge. There is scarcely any doubt that the anterior pituitary tissue produces the X-substances responsible for the facts summarized in the “law of follicular constancy” and the “law of puberty.” The statements of Smith and Engle on the quantitative dependence of follicular development upon the anterior pituitary tissue and my findings in partially castrated females are in complete agreement.