Book - The Hormones in Human Reproduction (1942) 6

From Embryology
Embryology - 23 Oct 2020    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)

Corner GW. The Hormones in Human Reproduction. (1942) Princeton University Press.

   Hormones in Human Reproduction (1942): 1 Higher Animals | 2 Human Egg and Organs | 3 Ovary as Timepiece | 4 Hormone of Preparation and Maturity | 5 Hormone for Gestation | 6 Menstrual Cycle | 7 Endocrine Arithmetic | 8 Hormones in Pregnancy | 9 Male Hormone | Appendices
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)

Chapter VI. The Menstrual Cycle

"In this Enquiry indeed, which we are now attempting, no less useful than agreeable, the Wits of almost every Age have toiVd: hut as there is hardly any Argument, on which Physicians have wrote more; so is there no one, in which they have given less satisfaction to their Readers. . . . I shall not appear to have employed my Time ill, in endeavoring to set the Nature of the Menses in a clearer Light, than I find it hitherto done by Authors. In which Performance the Reader will find nothing abstruse, nothing far removed from common sense: inasmuch as it has been my only Care to find out the Truth, as much as possibly I could." - John Freind, Emmenologia, translated into English by Thomas Dale, 1729.


Surely the process of menstruation is one of the strangest things in all Nature. An important organ - the uterus - serving an indispensable function, is overtaken at regular intervals by a destructive change in the structure of its lining, part of which undergoes dissolution with hemorrhage, and must be reorganized in every monthly cycle. The loss of blood from organic tissues, everywhere else in the animal kingdom a sign of injury, even of danger, is in this one organ the evidence of healthy function. To make the puzzle greater, menstruation is by no means general in the animal kingdom, or even among the mammals. It occurs, indeed, only in the human race, in the anthropoid apes (having been observed in chimpanzees and in the gibbons), in the baboons, and in the Old World monkeys ; in short, in a closely related group of primates, one little portion only of the great class of Mammalia. No other animals, in forest, plain, or sea, hiding in dens or grazing the fields, undergo in the course of their cycles any such phase of hemorrhage. It is a paradox indeed that this curious phenomenon of periodic breakdown, seemingly an imperfection, a physiological flaw, is characteristic solely of the females of those very animals we are pleased to think the highest of earth's creatures (Appendix II, note 9).


The periodicity of menstruation. In human females, menstruation recurs at intervals of about 4 weeks. There is a common impression that the cycles are normally quite regular, but any woman who will keep an accurate calendar of her cycles will find a surprising variability.

A recent statistical analysis of thousands of records[1] shows in fact that the commonest average cycle length (the "mode" as statisticians say) in adult European and American women is 28 days, but it is also very common for a woman to average cycles of 25, 26, 27, 29, and 30 days. The individual woman, moreover, often varies several days, in any one cycle, from her own average. To state this in exact terms, so as to make clear just how much variation a woman may consider normal, is rather difficult, for it is a matter of statistics and such things are hard to translate into everyday language. The clearest statement is that of Professor Arey, already quoted, whose words I paraphrase as follows: let a woman keep a record of her cycles for several years, so that she has enough observations to strike an average. Say, for example, that her personal average is 28 days. Arey's figures show that with this information she cannot hope to predict the onset of any given period with accuracy closer than 2.5 days plus or minus, i.e. she may expect it any day between the 25th and 30th after onset of the last period, and even then one-third of all her cycles will depart still more widely from the average length, say another day or two, and sometimes more. The statement that she averages a 28-day cycle has the same kind of meaning as the average price of eggs during the year, which may be very different from the price at any one time, or a baseball player's average of home runs per game. All it means is that the length of her successive cycles will be within a few days, more or less, of 28 days, seldom however coming out at that precise length. In fact a woman whose cycles were perfectly regular to the day, during many months or years, would be a medical curiosity. No such case has ever been reported.


When we come to consider, a little later, the intricate interplay of hormones that goes to produce the cycles, we shall not be surprised that the timing is not perfectly regular. Nor will we be surprised to learn that in young girls, during the first few months or years of menstrual function, while the endocrine mechanisms are becoming adjusted, the cycle length is extremel}'^ variable. Arey compiled the records of 100 girls during about two years after the onset of menstruation, and calculated the average cycle length of each during this epoch of their lives. One-third of these girls actually never had a cycle that corresponded exactly to the day with their own personal averages ; in other words, every single cycle varied from the arithmetical norm. During the latter part of adolescence there is considerably greater regularity.


The first menstruation most commonly takes place sometime between the ages of 12 and 14 inclusive. The average age at the time of onset, in the white race at least, is 13% years, but onset at any age from 11 to 16 may be regarded as normal. Delay beyond 16 is a matter for medical investigation.


The normal duration of the menstrual flow may be from one day to one week ; the modal duration is 5 days.


Among the infrahuman primates there is only one, the Rhesus monkey, which has been studied in numbers large enough for positive statement. In this species, observed in captivity in the United States and England, the mode, i.e. the most frequent cycle length, is 28 days, as in women. Individual animals have average cycle lengths of 25 to 31 days, and single cycles vary from 14 days up. Rather scanty observations on chimpanzees, baboons, and a few species of monkeys mostly show averages a little longer than 28 days. This statement might or might not hold good if the statistics were more extensive (Appendix II, note 10).


The poetic suggestion quoted at the head of Chapter III, that the reproductive cycles of living things are part of the rhythms of the universe, must not be taken too literally. Menstruation is not regulated by the moon. It happens that the lunar cycle has the same length to the day as the modal human cycle, but we have seen that the human cycle frequently deviates from the mode, and if, for example, the start of the period coincides with the new moon or any other given lunar phase, the odds are it wiU be off cycle by at least a day or two next month and perhaps completely out of phase next season. I once had 4 females of the Java monkey (Macaca irus) in a cage adjacent to a large group of the closely related Rhesus monkeys. While the latter were running cycles of 28 days' modal length, their Javanese cousins, living under the same moon, were exhibiting a modal cycle length of 35 days. As mentioned in Chapter III, the cycles of other mammals may vary from 5 days to a year in length, and if we consider the birds and insects we finds cycles of one day to 17 years. If the heavenly bodies are to control these rhythms, the cycle of the 17-year locust calls for a hitherto unknown comet !


The idea of a relation between human menstruation and the moon is, however, ancient and widespread. It was, no doubt, suggested by the obvious and inescapable relation between the moon and the tides of the sea. If the moon can control the ebb and flow of great waters, why not also the tides of human life . Perhaps the popular mind has also been influenced by the fact that outbursts of insanity in women sometimes accompany the menstrual cycle ; this seems again to link menstruation with the moon, which has long been considered a cause of lunacy. Then there are, of course, certain special cases in nature in which the life of an animal is directly influenced by the moon or the tides (e.g. the palolo. Chapter III) . These cases may have helped to foster the notion we are discussing. As lately as 1898 the eminent Swedish physicist Svante Arrhenius thought he had proved the connection of lunar and menstrual cycles by mathematical evidence, but this has been completely disproved, notably by the English physicians Gunn, Jenkin, and Gunn (1937). It is indeed difficult to conceive of any direct participation of the moon in the reproductive cycles of the land-dwelling primates, for if it were really efl*ective we should expect menstruation to occur at the same phase of the moon in all females of a given species, a state of aff"airs that would have made the social organization of mankind unthinkably different from what it is.


Nature of the Menstrual Cycle

Events of the cycle in non-menstruating animals. To get a clear understanding of the process of menstruation, it is necessary to understand first what takes place during the cycle in animals which do not menstruate. This has already been discussed in part and illustrated in previous chapters, and is summarized in the diagram herewith (Fig. 19) which represents the typical or generalized cycle of mammals. If we wish to talk about any one species, we shall have to introduce modifications into this scheme, but as it stands it can be used as a basis for understanding them all. In the upper portion, which shows events in the ovary, we see (beginning at the left) the growth and ripening of the follicle. The moment of rupture of the follicle and discharge of the egg gives a convenient point of division, which we may consider as the start of a new cycle. Looking at the third part of the diagram, that indicating sex activity, we see that ovulation occurs during estrus, an arrangement which is adapted to secure fertilization of the egg. Next, the follicle is converted into a corpus luteum. This in turn runs its course, secreting progesterone for about two weeks (in typical species) and then, if the Ggg is not fertilized, the corpus luteum suddenly begins to degenerate and ceases to secrete its hormone. Thereafter, a new crop of follicles begins to develop. In some animals the new cycle follows at once (e.g. the guinea pig, which has a cycle of only 15 or 16 days) ; in others several months may elapse, during which the ovaries are relatively dormant (as in dogs and cats) or a whole year, as in many wild animals.


Digression about the cycle in general. We come now to the fundamental question of the female reproductive cycle, namely what causes the alternations of structure and function in the ovary. When the cycle was first discussed, in Chapter III, we could deal with it only as an observed phenomenon of natural history, but we are now in a position to consider the problem in the light of our knowledge of the hormones. To resume this subject where we left off on page 75, there is scarcely any doubt at present that the cycle is somehow produced by interplay of hormones from the ovary and the pituitary gland.


If we look at the pituitary gland or hypophysis (Plate XIX and Fig. 20) we find that this gland of internal secretion is composed of two major parts, the anterior and the posterior lobes. It is the anterior lobe which produces hormones (probably two in number) having the power of stimulating the ovary to produce estrogenic hormones and of promoting the growth of ovarian follicles. They also affect the male organism, causing the testes to grow and produce sperm cells. Because of these actions the hormones we are discussing are called gonadotrophic, a name which signifies "producing growth of the sex glands." From the brilliant work of P. E. Smith, Bennett Allen, H. M. Evans, Zondek and Aschheim, and many others between 1915 and the present time, we have learned (as mentioned in Chapter III) that removal of the anterior lobe of the pituitary stops growth of the ovaries and puts an end to the cycles of the animal. By implanting bits of anterior pituitary, or better by injecting extracts of the gland into immature animals, the ovaries are caused to grow and the cycle to begin. The ovary is thus absolutely dependent upon this action of the pituitary. Removal of the anterior lobe produces all the effects of castration, for without it the sex glands, ovary and testis, deteriorate to inactivity. On the other hand, there is a good deal of evidence that the estrogenic hormone of the ovary represses the production of the pituitary gonadotrophic hormones. After removal of the ovaries, the pituitary gland is found to contain more gonadotrophic potency than before; after injection of estrogenic hormones it contains less (Appendix II, note 11).


When these facts became known, a fairly clear explanation of the reproductive cycle suggested itself almost simultaneously, about 1931-1932, to a number of investigators, among them first perhaps Brouha and Simonnet in Paris, then to Leonard, Hisaw and Meyer in Wisconsin, and Moore and Price[2] in Chicago. This hypothesis suggests that the cycle is Hke a clockwork in which the pituitary is the driving force and the regulatory escapement is the reciprocal action of ovarian and pituitary hormones (Fig. 21). The pituitary makes the follicles grow, ripens the follicles and eggs, and causes the production of estrogenic hormone. The rising tide of estrogenic hormone thus checks the production of pituitary hormone, which begins to fall off as estrus occurs. The estrogenic hormone is used up, and as it reaches a low ebb, the pituitary, now freed from the repressive action of the ovary, again begins to secrete its gonadotrophic hormone. Up goes the pituitary and then up goes the ovary again, thus getting another cycle under way. This scheme, however, cannot fully explain the cycle. As Lamport has shown, a push-pull action of the two hormones would naturally tend, not to effective cyclic fluctuations of estrogen, but to ever smaller changes approaching equilibrium. We must therefore postulate some other event which occurs from time to time to break the balance of the two hormones. Under the push-pull hypothesis it is in fact easier to understand the long diestrous phase of cycles like those of animals that have an annual cycle, when (as we may suppose) the estrogenic and gonadotropic hormones are balancing each other, than to explain what happens to set the see-saw swinging again, or to bring on cycles, in some animals, every few days or weeks. At present we can only make vague conjectures about the possible role of other hormones, e.g., progesterone or another pituitary hormone.


Fig. 20. Above, the pituitary gland, showing the anterior lobe and the posterior lobe with its stalk by which the gland is connected to the brain. Enlarged approximately 5 times. Below, median section showing position of pituitary gland in its bony cavity at the base of the skull; compare with the X-ray photograph, Plate XIX.


Plate XIX. X-ray photograph of the human skull, to show the location of the pituitary gland. The arrow points to the little hollow (sella turcica) in the bone below the brain, in which the gland lies. About 2/5 natural size. Courtesy of the Eastman Kodak Company, Rochester, N.Y.


Plate XX. The human infant at birth, with the placenta and membranes. From the anatomical plates of Julius Casserius, published by Adrianus Spigelius in 1626.


Fig. 21. Diagram illustrating the alternation or "push-pull" hypothesis of the ovarian cycle discussed in the text.


Events of the cycle. To resume the main theme of our discourse, during all these changes in the ovary the uterus is, of course, constantly under the influence of the ovarian hormones. Even when there is a long anestrous interval between one ovulation and the next, the ovary produces enough estrogen to protect the uterus from atrophy. As the follicles enlarge and ripen, there is a period of growth and development of the lining of the uterus (endometrium). When the corpus luteum is formed and begins to produce progesterone, the uterine lining is rapidly brought into the progestational condition, as described in Chapter V and illustrated in Plate XVII. About one week is required to complete these changes. The favorable environment thus prepared for the embryos (pages 107-111) is maintained about one week longer, making two weeks in all between the beginning and the end of the active life of the corpus luteum. This is the progestational phase of the cycle. If the animal mated while in estrus, the embryos arrive in the uterus about the 4th day (later in some species) and begin to attach themselves sometime between the 7th and the 13th day, according to the species. It will be seen that the corpus luteum functions long enough to give time for implantation of the embryos. If this occurs, some sort of signal, probably via the pituitary gland, causes the corpus luteum to survive and maintain the uterus in a state favorable to early pregnancy.


Retrogression of the uterine changes. We are considering here, however, a cycle in which the eggs are not fertilized. In such a case they are transported through the oviduct to the uterus, where about 8 or 9 days after they first left the ovary they go to pieces and disappear. The corpus luteum holds on until the 14th or 15th day, then degenerates and ceases to deliver progesterone to the blood stream. The endometrium is thus deprived of its hormonal support. The changes induced by progesterone disappear in the course of a few days. The blood flow through the uterus diminishes, the lining becomes thinner, the cells of its surface epithelium and glands diminish in number and height, and the glands resume the simpler form that characterizes the interval and follicular phase of the cycle. Generally speaking, the steps of this reversion are gradual ; it is spread over several days, and gives no outward sign to let us know it is in progress.


In our diagram (Fig. 19) the whole sequence of changes in the lining of the uterus is illustrated by the middle portion, which is a conventionalized representation of the glands in their successive phases.


The cycle in menstruating animals. The cycle of the menstruating animals and the human species is fundamentally similar to that of other animals. Two important difl*erences, however, exist. In the first place there is not a sharply defined phase of sexual receptivity like the estrus of other mammals. Although cyclic fluctuations of sex activity occur in some of the apes and monkeys, this is by no means as well defined as in most other animals, and in the human female sex desire is obviously much more influenced by all sorts of moods, social situations, domestic ups-and-downs, and the like, than by any tendency to cyclic alternation. Mating may occur on any day of the cycle. There is no outward sign, like the estrous behavior of lower mammals, to indicate the time of ripening of the ovarian follicle and its egg.


It is interesting to speculate about the effect of this suppression of estrous rhythm upon human life and the progress of the race. Certainly our customs would be very different from what they are if the sexual compulsions of women were like those of animals with strongly marked estrous periods. In these creatures the sex response, intense and irresistible in the female during estrus, is wholly absent at other times ; in the human species it is moderated but diffused over a larger proportion of the time. In their various aspects and sublimations, from downright sex desire to affection and vague romantic yearnings, the impulses of sex color in some degree our entire adult lives, teach us to love nature and art, and call us to sacrifice and devotion. In this respect above all mankind differs from the beast.


Regardless, however, of this all-important difference of behavior, the cycle of the ovary proceeds in the human species as in the others (Fig. 22). The follicle ripens and ruptures, the egg passes to the uterus, the corpus luteum forms and takes up its endocrine function. The lining of the uterus undergoes a profound progestational change. The epithelial cells of its glands multiply so greatly that the glands have to become sinuous and pleated, in order to be accommodated in the available space. The glands fill up with fluid secretion and therefore become dilated. The result is a very characteristic appearance, when seen in sections of the uterus. This progestational or "premenstrual" state is well shown in Plate XXI, C.


Inspection of the diagram (Fig. 22) will show that the premenstrual phase is at its height during the second week after discharge of the egg from the ovary, just as in other mammals. If there is a mating, and the egg is fertilized and becomes an embryo, it will reach the uterus when the endometrium is fully under the influence of the corpus luteum and ready to take care of the new arrival.[3] This is clearly illustrated in Plate XII, C, a photograph of one of the earliest known human embryos, obtained by Dr. Arthur T. Hertig of Boston, and preserved in Baltimore at the Department of Embryology of the Carnegie Institution of Washington. The uterus in which this 11 -day embryo has attached itself is in the typical progestational phase, as shown by the form of the glands.


The menstrual breakdown. About the 14th or 15th day after ovulation, the corpus luteum begins to degenerate, as in other animals. The uterus, thus deprived of support by progesterone, undergoes a violent reaction. In its innermost layer the circulation of blood is disturbed, the surface epithelial cells, the glands and the connective tissue are damaged, and the tissues break down. Blood from small ruptured vessels fills the cavity of the uterus and trickles toward the vaginal canal. A section of the endometrium at this time shows a remarkable picture ; the surface layer has sloughed away, and the stumps of the glands jut into the central mass of blood and cellular debris. In the course of a few days a process of repair sets in, the lost surface cells are replaced, the glands restored and the debris cleared up.


The various stages of the uterine cycle are shown in Plates XXI and XXII, which present a series of specimens from the Rhesus monkey.

The sequence of these events is summarized in the diagram, Fig. 22. From this it will be seen that ovulation takes place about the middle of the interval between the two menstrual periods. It is customary to count the days of the primate

Plate XXI. Three stages of the cycle of the uterus of the Rhesus monkey.

A, 16th day of cycle, just after ovulation; interval stage. B, 23d day of cycle. Effect of corpus luteum hormone appears in the glands; early premenstrual stage. C, 27th day of cycle. Menstruation due one day later. Full progestational (premenstrual) stage. All magnified 10 times. A, Corner collection (no. 2)}

B, courtesy of C. G. Hartman (H. 326); C, courtesy of G. W. Bartelmea (B. 123).


cycle from the first day of menstruation. Ovulation most commonly takes place about the 12th to the 16th day, although in individual cases it may be earlier or later than this. The corpus luteum is active for about 13 or 14 days, and therefore its degeneration brings on menstruation again 25 to 30 days after onset of the last period.

The "safe period.** Let us digress again for a moment, to discuss, in passing, an interesting and important deduction that follows from the schedule of the human cycle, as shown in the diagram, Fig. 22. There is evidence from many species of animals that the eggs can be fertilized only while in the oviduct, during the first two or three days after their discharge from the ovary. We know also that the sperm cells cannot survive more than a few days in the female reproductive tract. It follows that the only part of the human cycle during which fertilization of the egg can occur is the few days following ovulation. Since, however, there is no way of ascertaining the date of ovulation and it may vary by several days, we shall for the sake of caution estimate the presumably fertile period as a few days longer each way, say from the 8th to the 20th day of the cycle, counting from the first day of the menstrual period. All the rest of the cycle, i.e. from about the 20th day to the 8th of the next cycle, will be a period of sterility, during which mating will not result in pregnancy. This is the theoretical basis of the so-called "safe period" method of birth control. If all women had regular cycles and things never happened out of turn, it would no doubt be a fully effective method, but irregularity of cycles and unpredictable variations make it much less than certain.*


Plate XXII. The uterus of the Rhesus monkey during menstruation. A first day of flow; at bl., small collections of blood in the lining of the uterus. B, third day; note loss of surface tissues of lining and disappearance of progestational pattern of glands. C, anovulatory menstruation, first day. Note, in comparison with A, that there is no progestational change of the glands. Magnified 10 times. Ay B, Corner collection (nos. 39, 22) ; C, courtesy of G. W. Bartelmez (B. 128).


Role of the blood vessels of the uterus in menstruation. Within the past few years a good deal has been learned about what actually happens to produce the menstrual breakdown. Much of this advance we owe to G. W. Bartelmez of the University of Chicago, and to his former associate, J. E. Markee, now of Duke University. To make it clear we must first understand the arterial blood circulation of the lining of the uterus. The endometrium is fed by arteries which come up into it from the underlying muscle (Fig. 23). These have branches of two kinds. Those of one kind are very pecuhar, for they are wound into coils, making their extremely tortuous way toward the surface, where they break up into tiny capillary vessels (not shown in the diagram) that supply the inner one-third of the endometrium. The other kind of branching is that of the straight arteries, which run a short course directly to supply the basal two-thirds of the endometrium.


Fig. 23. Diagram of the arteries of the uterus, from the description of Daron. Enlarged about 20 times.

Carl G. Hartman, Time of Ovulation in Women. Baltimore, 1936.


By studying under his microscope a series of uteri of women and monkeys, collected at successive stages of the cycle, Bartelmez showed that the fundamental step in the menstrual breakdown is a shut-off of the coiled arteries. With such material, however, it is possible to see only interrupted stages of the process ; the sequence cannot be seen in full. Markee has therefore made use of a remarkably clever means of watching menstruation in progress.[4] Since we cannot see into the uterus, he undertook to put that organ (or rather, small pieces of its lining) into a situation where it can be watched. He grafted bits of endometrium into the anterior chamber of the same animal's eye, thus applying a method already used by a few investigators for other purposes. The small grafts are placed just behind the clear cornea, and get their blood supply through vessels which grow into them from the iris. The operation of grafting, which is done under complete anesthesia, is relatively simple, though, of course, it requires deft hands. The animal suffers no discomfort from the graft and is inconvenienced only by the fact that while under observation she has to sit in a tight wooden box, something like a pillory (but more comfortable), while the investigator studies her eye through a microscope. He is, by the way, at least as uncomfortable as the monkey, because the task of watching the winking, roving eye of the animal, changing the focus and moving the microscope and light whenever necessary, is enough to exhaust the patience even of a scientist.


The grafts survive and grow. They respond to estrogenic hormone, injected under the skin, by swelling and growing just as if they were still part of the uterus. If the ovaries are removed, the grafts undergo castrate atrophy. Most astonishing of all, when menstruation occurs in the uterus, it occurs at the same time in the eye-graft, runs the same course, and ceases at the same time. The menstrual hemorrhage which occurs in the eye, stains and clouds the aqueous humor for a few days but soon clears away.


Markee was able to watch the process through the microscope, using low to moderate magnification, from 12 to 150 times. What he saw has helped us greatly to understand the nature of the menstrual breakdown, although (as we shall see) there is much still to be learned. Markee tells us that the first sign of impending menstruation in the eye-graft is blanching of the tissues due to shutting off of the blood flow by contraction of the coiled arteries. This does not happen in all the arteries of the graft at one time, but in individual arteries, so that blanched patches appear here and there in the graft, until all the tissues ultimately experience the blanching. After a few hours this phase wears off. Through the relaxed arteries the blood flows again with renewed force, but the tissues of the endometrium and especially the capillary blood vessels have sufi^ered from the lack of blood supply. Here and there the small vessels give way and burst, causing tiny spurts of blood into the tissues. The little pools of blood thus produced coalesce and drain into the anterior chamber of the eye. In the uterus itself, similar hemorrhages are of course discharged into the cavity of the uterus. After a few days this strange series of events is over, and the damage is promptly repaired.


With Markee's direct observations to guide us, the study of prepared specimens of the uterus is much clearer. Observations by the two methods agree perfectly, but without observations of the eye-grafts we should probably not have learned the importance of the periodic shutting off of the spiral arteries.


Coiled arteries of the type thus shown to be fundamentally involved in menstruation in the monkey are also present in the human uterus, but have never been found in non-menstruating animals. Menstruation, then, is primarily an affair of the coiled arteries, which control the blood supply of the inner layer of the endometrium and by their closure cause breakdown, tissue damage, and hemorrhage (Appendix II, note 12).


In view of the violent disruption that characterizes the retrogressive phase of the cycle in women and in the other menstruating primates, it is a matter of great theoretical interest to know whether this stage in the non-menstruating animals is actually as free from tissue breakdown as I have rather summarily indicated. In other words, is menstruation a totally peculiar affair, sharply different from what goes on in mammals generally, or is it merely an exaggeration of a degenerative process that is present but not extensive in lower animals.? This question is being investigated, but the answer cannot be given now. We need to know most of all what goes on in the uterus at the end of the corpus luteum phase in the New World monkeys (the capuchins, spider monkeys, and howler monkeys), which in spite of their close evolutionary relationship to the other primates do not menstruate externally. Here, if anywhere, we may expect to find transitional conditions that may help explain the wherefore of menstruation. The evidence is not yet in, but I may say that there are hints, apparent to the expert microscopist, that even in the rabbit and other non-menstruating mammals the retrogressive phase has an element of acute damage in it. These signs are, however, slight indeed and the statement holds true that in almost all mammals, when the corpus luteum has done its work, and the uterus is released from its phase of progestational proliferation, it settles gently and inconspicuously back to the state it was in before the follicles matured.


Theories about the menstrual cycle. There is no need to discuss outmoded theories of the cycle here, except to exclude one or two ancient fallacies that still crop up occasionally. For example, some people still consider that menstruation is equivalent to estrus. This is a common notion among farmers. Because menstruation is the most prominent event in the human cycle, and estrus the most conspicuous phenomenon in the cycle of the barnyard animals, they are wrongly considered to be fundamentally alike. It would follow from this that in humans the egg is shed from the ovary at the time of menstruation, a notion which is absolutely incorrect, as will be seen from our previous discussion. Menstruation is the last stage, not the first, of the corpus luteum phase of the cycle.


Another false view, which prevailed widely among European gynecologists from 1880 to 1910, asserted that there is no chronological relation whatever between ovulation and menstruation. The egg may be shed at any stage of the cycle. This conclusion was drawn by surgeons who knew very little about other species, and who moreover usually saw at the operating table not normal pelvic organs, but those of patients with gynecological ailments, often subject to disturbances of the cycle.


When it began to be understood clearly that the ovary is an organ of internal secretion, a group of first-class German gynecologists, including especially Robert Schroeder, Robert Meyer, and Ludwig Fraenkel (the latter two now in exile) developed a theory which had been vaguely outlined a generation earlier, that the corpus luteum is in some way associated with the menstrual cycle. Gradually their views, clarified by intensive observation of human material, arranged themselves into a theory of the cycle which was very plausible and which has turned out to be partly correct.


This states that menstruation is simply the downfall of the premenstrual (progestational) endometrium, and that it is caused by the degeneration of the corpus luteum. It will be seen that this theory fits all that has been said about the primate cycle thus far, and that it is compatible with our diagram. Fig. 22. According to this theory, the endometrium cannot menstruate unless it is first built up to the "premenstrual" state. Professor Meyer put this into an aphorism which was much quoted by the gynecologists "Ohne Ovulation keine Menstruation" - without ovulation there can be no menstruation.


This is a beautiful, clear hypothesis, and it is half true. It is also, unfortunately, half false. The fallacy is subtle but fundamental, and leads us headlong into a mass of unsolved problems.


Anovulatory cycles. The failure of the German theory of the cycle is a matter of especial interest to me, for it was my lot to obtain (to my great perplexity) the first undeniable evidence against it. The story is best told as it happened. In 1921, after several years of work on the cycle of the domestic pig, I felt prepared to begin a study of a menstruating animal and for this purpose I chose the Rhesus monkey. Practically nothing was known on the subject. There had been two investigations. Walter Heape, a distinguished English biologist, had gone to India more than twenty years before to study reproduction in Rhesus monkeys and langurs, but illness had forced him to return to Cambridge, where he followed and described the cycles of a few animals he had taken home with him. M. A. Van Herwerden had studied material of a wholly different kind. Hubrecht, the great embryologist of Utrecht, had collected a great many reproductive tracts (uteri with ovaries) from several species of monkeys. These had been obtained largely by Dutch colonial officers in the East Indies. Miss Van Herwerden examined these specimens, which were unaccompanied by life histories or records of menstrual cycles, because the animals had been shot in the jungle by hunters. As regards the relation of menstruation to ovulation in the cycle of the monkey, the results of Heape and Van Herwerden were obscure and puzzling. Heape in his few cases observed no clear relation. Van Herwerden actually found that in some of the Hubrecht specimens the uterus was menstruating but there was no corpus luteum at all in the ovaries. In other menstruating animals a corpus luteum was present. This variability could perhaps be reconciled with the older theories of the human cycle, but not with the Meyer-Schroeder-Fraenkel theory. The absence of life histories, however, cast uncertainty upon the significance of Van Herwerden's observations. A hunter's specimen lets us see only one instant in the life of the animal ; who could tell the significance of these puzzling cases so completely removed from the context of life?


Meanwhile the German interpretation seemed plausible indeed. It could be matched without difficulty to all the recently gained knowledge of the cycles of mammals. Stockard and Papanicolaou's studies of the guinea pig (1917), those of Long and Evans on the rat (1921), which I had been privileged to watch for four years, and my own on the domestic pig had all emphasized the occurrence of regular cycles of ovulation followed by the progestational phase of the uterus. I supposed that application of the same methods to a menstruating mammal, namely the Rhesus monkey, would reveal a strictly parallel sequence, with menstruation as its last stage. If I kept my animals in good condition, observed their cycles with perfect vigilance, and autopsied them at carefully chosen stages in their cycles, I should obtain a series confirming the German theory. I thought that 25 monkeys and three years' work would suffice to establish the normal cycle, after which we could go on to all sorts of experimental studies in confidence that we could elucidate the normal physiology and the disorders of the human menstrual cycle.


Imagine my confusion when the very first monkey we killed disagreed completely with all we had expected. Rhesus monkey No. 1 was in my colony more than a year. She had 12 menstrual cycles in 12 months, the last 5 of which were respectively of 27, 29, 25, 24, 27 days, averaging 26.4 days. In the hope of recovering a young corpus luteum and of finding an egg in the oviduct or uterus, she was killed 17 days after the onset of the last previous menstrual period and 9 days before the expected onset of the next. To our astonishment, neither ovary contained any sign of recent or impending ovulation. There was no large follicle, no recent corpus luteum, nor any older corpus luteum from the last two or three cycles. In short, this animal was undergoing cycles of menstruation without ovulation and therefore without corpora lutea.


Monkey No. 2, on the other hand, fulfilled our original expectations. She, too, had a series of regular cycles. She was killed 14 days after the onset of the last period and 12 days before the onset of the next expected period. The left ovary contained a recently ruptured follicle and the egg was in the oviduct. This, by the way, was the first egg of any primate ever recovered from the oviduct. The case fits the diagram perfectly.


To make a long story as short as possible, it turned out that Rhesus monkeys do not ovulate in every menstrual cycle.[5] When they do ovulate, the corpus luteum of course is formed and causes progestational (premenstrual) changes in the uterus. When the corpus luteum degenerates, typical menstruation occurs, by breakdown of the premenstrual endometrium. When the animal does not ovulate, then naturally there is no corpus luteum and therefore no premenstrual change in the uterus. Menstruation occurs anyway, and the breakdown takes place in an endometrium which is still in the unaltered state. The corpus luteum is necessary for the premenstrual state, but not necessary for the breakdown.


This analysis of the situation has been confirmed by Markee through watching menstruation in endometrium grafted in the eye. Markee tells us that when he applies the microscope to the grafts he sees only one difference between ovulatory and anovulatory menstruation, namely the occurrence of the progestational phase in the former and its absence in the latter. The shutting off of the blood supply, the subsequent reflux of blood through the coiled arteries, the rupture of the small vessels and the hemorrhage are the same in both instances. With all this evidence it can hardly be doubted that anovulatory bleeding is also menstruation.


It is not possible to distinguish between ovulatory and anovulatory menstruation by ordinary observation of the living animals. The cycles are of similar length, the bleeding is similar in appearance and duration. Recent studies by Ines de Allende and Ephraim Shorr suggest that it may be possible in the future to detect anovulatory cycles by studying the vaginal cells.


My description of menstrual cycles without ovulation was at first rather generally mistrusted, but it has been confirmed by everyone who has studied the Rhesus monkey.[6] We know that anovulatory cycles are likely to occur in young animals in the first months after the establishment of menstruation, and in fully mature females in the early fall and late spring, that is to say at the beginning and end of the active breeding season of the winter months. Rhesus monkeys do not menstruate regularly in summer. The anovulatory cycles tend to occur, therefore, when the reproductive tract is preparing for its highest activity or receding from it. My papers describing the monkey cycle set off an active debate among the gynecologists as to whether anovulatory cycles occur in women. After much discussion and a great deal of careful observation, it is generally agreed that anovulatory cycles do occur, though with much less frequency than in monkeys. They seem to be most frequent in young girls and in women approaching the menopause.


There is no place for menstruation without ovulation, in the theoretical scheme which I have called the German theory. Therefore the savants who had formulated that theory simply declared that anovulatory menstruation is not menstruation at all. The rest of us, however, have gone on trying to find an explanation that fits all the facts. In this search the new knowledge of the ovarian hormones has begun to help us.


The Hormones and Menstruation

Experimental uterine bleeding. A simple experiment, made in 1927 by Edgar Allen, opened up the whole problem of the relation of the ovarian hormones to menstruation. Allen found that removal of both ovaries from a mature Rhesus monkey will usually cause within a few days a single period of menstruation-like bleeding. Why the medical profession had failed to discover this fact from human surgical patients is difficult to understand. It has long been known that removal of the ovaries abolishes the menstrual cycles, but the doctors had missed observing the fact that one period of hemorrhage often follows the operation before the cycles cease permanently. They seldom remove the ovaries except in the presence of disease, when the cycles are already altered, or for tumors which themselves produce bleeding, or as part of a larger operative procedure which may cause surgical hemorrhage from the uterus. Thus bleeding due purely to removal of the ovaries escaped notice until Edgar Allen discovered it in monkeys.


As a matter of fact, an experiment like Allen's had once been done on humans, on a large scale, and with the best intentions in the world. Robert Battey, a surgeon of Augusta, Georgia, in 1872 conceived the idea that neuroses and insanity in women are often concerned with the ovaries and may be treated by removal of these organs. He was probably led into the notion by observation of cyclic mental disturbance, paralleling the menses, insanity following child-bearing, and other conditions in which sex and the reproductive functions were of course concerned, though in a far more complex way than he could have imagined. Battey's radical proposal to remove the normal ovaries was put forward just at the time when the surgeons had gained command of the operation of ovariotomy (as they often ungrammatically called it). Antiseptic surgery. Lister's gift to the world, was now in general use, and the great American ovariotomists Ephraim McDowell, the Atlees, and their followers in Britain and Europe had worked out the operative technique. The operation was therefore relatively safe, and no doubt the patient's mental condition was often improved or at least subdued by the surgical intervention, with its anesthesia and opiates, by the rest in bed, and the nursing and general attention. At any rate "Battey's operation" was taken up widely by a profession thoroughly baffled by mental disease. Thousands of women were subjected to this drastic operation, not only in the United States, but in England, Germany and the rest of Europe, until in good time it became obvious that the psychiatric results did not justify it and that insanity with cyclic or sexual symptoms cannot be pinned directly to the ovaries. The late Dr. Edward Mulligan of Rochester, New York, told me of an incident in the last years of Battey's operation. Dr. Mulligan when a young surgeon studied for a time, about 1883, at Bellevue Hospital in New York City with the pathologist William H. Welch, himself a young man on his way to the Johns Hopkins and the leadership of the American medical profession. One morning Welch showed his pupils a tray containing a number of normal ovaries, removed that morning in the operating rooms, and took the occasion to denounce the practice of Battey's operation in words so vigorous that Dr. Mulligan still remembered them more than forty years afterward.


The point of all this is that removal of the normal human ovaries was very often followed within a few days by a period of bleeding from the uterus lasting several days. This appears in many of the case reports in medical journals from 1872 to 1885. The doctors did not always report all the postoperative details, but when they did they generally noted the hemorrhage, but never with comprehension. Thus an important observation was missed because the observers' minds were unprepared.


We must digress for a moment to mention that under the strict corpus luteum hypothesis of menstruation (which I have for brevity called the German theory) removal of the corpus luteum may be expected to bring on menstruation. This had been perceived and demonstrated at the operating table before 1927, when Allen announced, on the basis of his experiments on monkeys, the broader fact that removal of both ovaries, with or without a corpus luteum, has the same effect. It may help keep things clear, if we point out something the reader has probably thought out already, namely that removal of a corpus luteum produces bleeding from a premenstrual endometrium, whereas removal of the ovaries without a corpus luteum produces bleeding from an unaltered endometrium, as in anovulatory menstruation.


Estrin-deprivation bleeding. Edgar Allen reasoned that the effects of removal of the ovaries of his monkeys were really due to removal of the estrogenic hormone, which had recently been discovered, thanks so largely to his own investigations. He therefore took a castrated female monkey and gave her a course of injections of estrogenic hormone. When he discontinued the treatment, bleeding ensued. In other experiments he removed the ovaries and immediately began daily doses of estrogenic hormone. As long as the hormone was given, there was no bleeding; that is to say, the hormone was able to substitute for the ovary. When it was discontinued, the bleeding occurred.

The following diagram represents graphically the experiments just described.

Removal of ovaries; estrin deprivaiion:


Fig. 24. Illustrating the experiments of Edgar Allen, 1927, 1928. In this and the following 3 graphs the black bars indicate uterine bleeding. These diagrams are from an article by the author in the American Journal of Obstetrics and Gynecology, by courtesy of the C. V. Mosby Company.


On this basis Allen formulated the estrin-deprivation hypothesis of menstruation, which suggests that natural menstruation, like the experimental bleeding, is due to a cyclic reduction of the amount of the estrogenic hormone available in the body.


Subsequent experiments done with carefully graded doses of the hormone, including especially those of Zuckerman, of Oxford, have shown that not only total deprivation, but also mere lowering of estrogen dosage below a certain level will produce the bleeding. The word "deprivation," as used in this connection, is therefore to be taken in a relative sense.


Fig. 25. In the lower figure the natural level of estrogen is shown fluctuating cyclically, not as a proved fact, but to show how the actual results of injecting the ovarian hormone bear on the estrin-deprivation theory.


The correctness of the estrin-deprivation h3^othesis can be tested in a very simple way. We need only choose a monkey that is menstruating regularly, and keep up her estrogen level by injecting the hormone, for say ten days before the expected menstrual period. This should prevent menstruation. I tried this in a sufficient number of monkeys, giving them doses of estrogen I thought similar to their own natural supply from their ovaries. Later in the experiments these doses were increased several fold. The hormone did not stop the next menstrual period. If the treatment was continued into later cycles, menstruation was often delayed, perhaps because of roundabout action through the pituitary gland. Zondek has found that in women very large doses of estrogenic hormone disturb the menstrual cycle, owing to inhibition of the gonadotrophic mechanism of the pituitary. These upsets produced with long-continued or very large doses are however not altogether pertinent. The theory calls for relatively small fluctuations, within the body's normal range of hormone production. On this hypothesis, however, the very first period should be prevented, and this did not occur. If my dosage of hormone was really physiological, as we say (that is, something like the amount the animal herself would produce) then the cstrin-deprivation hypothesis in its simple original form is not adequate to explain the observed facts.


Progesterone and menstruation. On the other hand, administration of the corpus luteum hormone even in small doses prevents menstruation in experimental animals, abolishing the very first menstrual period after injections are begun (provided they are started a few days before the expected onset). Both Hisaw and I have found that experimental estrin-deprivation bleeding, produced by removal of the ovaries or by discontinuance of a course of treatment with estrogenic hormone, is prevented by small doses of progesterone. Smith, Engle and Shelesnyak at Columbia University arranged an exceedingly vigorous condition of estrin deprivation. Their monkeys were first given a course of gonadotrophic hormone from the pituitary gland (see p. 141) to stimulate the output of estrogenic hormone from the ovaries. They were also given generous doses of estrogens for good measure. These hormones were suddenly discontinued and at the same time the ovaries were removed. In the face of all these reasons for deprivation bleeding, modest doses of progestin (crude progesterone) completely prevented hemorrhage.


On the other hand, progesterone deprivation, like estrin deprivation, invariably causes menstruation-like bleeding. This was very clearly apparent in a series of my experiments in which progesterone was given to normally menstruating monkeys. During the injections, menstruation ceased. When the hormone was purposely stopped at a time which would have been midway between two periods (had the latter been occurring on their original schedule) progesterone-deprivation bleeding then occurred within a few days. The monkey next menstruated spontaneously about 4 weeks after the experimental period. This tells us that the monkey's timepiece mechanism accepts progesterone-deprivation bleeding as if it were actual menstruation, and takes a fresh start from the induced period.


Progesterone preventa eetrin-deprivaLLion bleeding:


Fio. 26. The upper figure illustrates the experiments of Smith and Engle, 1932, and Engle, Smith and Shelesnyak, 1935. The lower figure represents the results of Hisaw, 1935, and Corner, 1938. Estrin-deprivation bleeding is postponed by progesterone; discontinuance of progesterone is then followed by bleeding.


The diagram (Fig. 26) illustrates these facts.

At this point Dr. Markee may be called again as a witness. He tells us that when bleeding is produced in one of his eye-grafts, by withdrawal of either the estrogenic hormone or progesterone, he observes the same sequence of blanching and hemorrhage that occurs in spontaneous menstruation. If the monkey is given progesterone, the graft will bleed from a "premenstrual" state; if given estrogenic hormone, it bleeds from the interval state.


With this information in hand it is possible to plan an experiment in imitation of the normal ovulatory cycle. This is represented in the upper half of the following diagram, Fig. 27. The underlying idea occurred to workers in the Oxford, Harvard, and Rochester (N.Y.) laboratories at practically the same time, and gave consistent results when tried. The dosage cited here is that of my own version of the experiment. A castrate female monkey is given a daily dose of estrogenic hormone, 125 international units, sufficient to build up the endometrium to normal thickness and structure. After 10 days, a daily dose of progesterone is added (just as would have happened had the animal developed a corpus luteum of her own). Ten days later, at the 20th day of the experiment, the progesterone is discontinued, but the daily injection of estrogenic hormone is continued. In spite of the estrogen, we find that bleeding invariably occurs in a few days. Indeed, the dose of estrogen may be greatly increased, say to 500 international units, beginning on the day on which the progesterone is discontinued ; but menstruation-like bleeding still occurs. Seven hundred units or more may be necessary to prevent it, although such doses as 500 or 700 international units are of course much more than necessary to maintain the uterus when not working against progesterone deprivation.


Progesterone during a. courae of estrin:

Fig. 27. The upper figure illustrates the production of bleeding after discontinuance of a course of progesterone, in spite of continued and more intensive estrogen treatment. The dosage shown is that of the author's experiments (Corner, 1937, 1938) ; similar results were obtained by Zuckerman, 1937, and by Hisaw and Greep, 1938.



These facts enable us to construct a relatively simple hjrpothesis of the menstrual cycle which is really a modified form of the estrin-deprivation hypothesis (Fig. 27, lower part). We start by assuming that progesterone in some way or other has the property of suppressing the menstruationpreventing power of estrogen, while itself holding off menstruation. In the normal cycle the animal does not bleed in the first half of the cycle (follicular phase), because the ovaries are furnishing estrogen. She will not bleed during the second half of the cycle (corpus luteum phase) because the corpus luteum is furnishing progesterone. By our assumption, however, the corpus luteum is suppressing the protective effect of the estrogen; therefore when the corpus luteum undergoes retrogression, the animal finds itself deprived of the action of both estrogen and progesterone, and the endometrium breaks down. Ovulatory menstruation is thus a special case of estrin-deprivation bleeding.


This explanation of the normal cycle is, of course, simply a hypothesis which has been formulated to explain our observations. Whether things happen this way in the normal monkey or in woman remains to be proved. It is at least not contradicted by any of the known facts, and it has the merit of simplicity, because it calls for a cyclic variation in only one hormone, namely progesterone. Its one unproved assumption, that progesterone somehow cuts down the action of estrogen on the uterus, is supported by various other evidences of a sort of antagonism between the two hormones in some of their other activities. It has been suggested that progesterone has the property of speeding the elimination of estrogens from the body. If this proves correct it is all we need to complete our hypothesis.


This scheme does not explain anovulatory menstruation, for in cycles without ovulation there is, of course, no coming and going of the corpus luteum. Anovulatory menstruation is therefore probably due to estrin-deprivation alone. We can imitate it perfectly in castrate animals by simply giving a course of estrogen injections interrupted or sharply reduced at suitable intervals. What is actually happening in the female organism remains to be worked out. We are not yet sure that there is an actual up-and-down of estrogen in the body sufficient to produce deprivation bleeding. The daily assay of estrogens in the blood is very expensive and at present not reliable enough for our purpose. Zuckerman has shown that when a castrate monkey is kept on a relatively small but constant daily dose of estrogen, there is a tendency to occasional uterine bleeding which may become fairly regular. We may conjecture, on this basis, that perhaps some sort of give-and-take relation exists between estrogen and some other hormone, just as between estrogen and progesterone when the corpus luteum is present, so that even without the corpus luteum a periodic state of estrin deprivation occurs. Can it even be that the adrenal gland produces something that can suppress the estrogens (we know that a number of steroidal compounds resembling progesterone are extractable from that gland) ? The play of hormones in anovulatory menstruation is anybody's guess, and those of us who have worked on it can assure our colleagues, on the basis of much vain conjecture and many futile experiments of our own, that the problem is not an easy one. Some little fact is lurking just beyond our grasp.


Since the first draft of this chapter was written, Hisaw has reported from the Harvard zoological laboratory some experiments which show that very small doses of progesterone given and then discontinued (1 milligram a day, for one to five days) will set off menstruation-like bleeding in castrate monkeys which are receiving large daily quantities of estrogenic hormone. He suggests therefore that anovulatory menstruation may be due to progesterone deprivation; even if there is no corpus luteum, he says, there may be a little progesterone produced in Graafian follicles (there is, in fact some collateral evidence for this latter part of the conjecture) and this may be enough to cause menstruation when such a secretion of progesterone ceases. It is a plausible conjecture, and one which calls for no new factor outside the ovary ; but it will be difficult to prove.


The immediate cause of the menstrual process. From the foregoing sections it will be perfectly clear that the breakdown and hemorrhage of menstruation are consequent to the deprivation of estrogenic hormone or progesterone. It is also very probable, from the studies of Markee, that these effects are initiated by constriction of the peculiar coiled arteries of the endometrium, which produces damage to the tissues and ultimate degeneration. But how can it be that a temporary deprivation of one of these two particular hormones can shut off the arteries in one particular tissue.


This question is now the key problem in the theory of menstruation, and it remains unsolved. The attack upon it is in the stage of skirmishing, in which all the possible explanations are being put forward for discussion and trial ; but up to the present no one of them has found support by experiment. The reader may, however, be interested in the mental processes of a group of puzzled investigators, and therefore I list the conjectures for what they are worth. To merit consideration at all, any explanation must fit the facts that (a) hormone withdrawal causes uterine bleeding; (b) this does not take place at once, but only after 3 to 8 days; (c) the bleeding, once hormone deprivation is well under way, cannot be postponed by renewed injections of estrogen or progesterone; (d) it takes place in grafted bits of endometrium (in the eye or elsewhere) which have no connection with the nervous system.


Since all of these conjectures involve the arteries, it may be helpful to recall the fact that the walls of an artery contain numerous cells of involuntary muscle, laid on in circular fashion around the inner tube (endothelium) that conducts the blood. When these muscle fibers contract, they squeeze down upon the blood stream like a man's fingers abput a rubber bulb. It is thus that the blood pressure is raised by a dose of adrenin or by a strong emotional state, both of which cause the arterial muscle cells to contract. Such muscular cells exist in the coiled arteries of the uterus as in all other arteries (Appendix II, note 13).


Hypothesis 1. It may be that the coiled arteries are peculiarly and directly dependent upon the ovarian hormones, in some such way (for example) as the ovary is dependent upon the pituitary. This means that withdrawal of the ovarian hormones would let down the condition of the coiled arteries, causing them to contract. This hypothesis is the simplest, calling for no other hormones or special substances, but it is exceedingly difficult to try out, for the only way of proving that the ovarian hormones are the sole factors involved is to exclude all other possible factors, but when we cut off the estrogenic hormone, how can we know we are not thereby putting some other factor into action? If somebody could show us how to keep a coiled artery alive and working outside the body where we could deal with it alone, we could soon test this hypothesis, but the infant art of tissue culture has by no means reached the point of keeping an artery alive all by itself, and moreover these arteries are so tiny that they would have to be handled under the microscope — a truly difficult project!


Hypothesis 2. This admits the possibility, mentioned above, that withdrawal of estrogen permits something else to go into action. We know that even in the non-menstruating animals, withdrawal of the ovarian hormone causes a certain amount of deterioration of some of the cells of the surface epithelium and of the glands of the uterus. Under the microscope we see fragmentation of the nuclei and the accumulation of protoplasmic debris in the cell bodies. Is it possible that some chemical substance produced in the course of cellular breakdown (as histamine, for instance, is produced in burned tissues) diffuses through the endometrium to the arteries and causes them to contract? This hypothesis has interested me very much and I have made many experiments to test it, but always with negative results.


Hypothesis 3. Another conjecture, a variation upon the foregoing, is that the uterine coiled arteries are sensitive, when not protected by the ovarian hormones, to some substance that is normally present in the blood stream. We must suppose that withdrawal of the hormone allows this substance to act upon the arteries. One of the possible constrictor substances would be pituitrin, the secretion of the posterior lobe of the pituitary gland, a hormone which is highly potent in promoting contraction of smooth muscle. Carl G. Hartman tried this with negative results, and moreover P. E. Smith obtained bleeding by estrin deprivation in monkeys from which he had removed the whole pituitary gland. Adrenin has been thought of, but Edgar Allen succeeded in producing estrin-deprivation bleeding in monkeys from which the adrenal glands had been removed. This experiment is not quite conclusive, for monkeys may possibly have other sources of adrenin beside the adrenal gland. Up to the present, at least, this hypothesis has yielded no valuable clues.


Hypothesis 4. George Van S. Smith and 0. W. Smith have suggested that the bleeding of menstruation is caused by conversion of estrogenic hormone into a non-estrogenic byproduct which is toxic to the endometrium. This hypothesis could perhaps explain ovulatory cycles, but it cannot be fitted very well to simple estrin-deprivation bleeding; in any case it will be acceptable only when somebody comes forward with chemical derivatives of the estrogenic hormones that are especially toxic to the endometrium (Appendix II, note 14).


These are the most plausible current guesses about the immediate cause of uterine bleeding after hormone deprivation. None of them has been proved or even rendered likely, by experiment. It is indeed vexatious that we cannot clear up this important problem.


The modern concept of the cycle. By way of summary, let us now set forth a concise description of the primate cycle as revealed by recent research. What follows will, I think, be accepted by most of the American investigators, and also the British, although a die-hard English gynecologist a few years ago dubbed it "the American theory."


We begin by suggesting that there is a basic tendency to cyclical function of the ovaries, and that this is produced by a sort of reciprocal "push-pull" reaction between the pituitary and the ovarian hormones, as explained in full earlier in this chapter. In most cycles of fully mature human females, the pituitary gonadotropic hormones cause the ripening of a Graafian follicle, the discharge of its egg, and the formation of a corpus luteum. This in turn sets up the progestational or "premenstrual" state of the endometrium. When the corpus luteum degenerates, menstruation ensues because of the withdrawal of progesterone. In some cycles, however, especially in young girls and in women approaching the menopause, a follicle does not ripen in the ovary, but after about the same interval as in an ovulatory cycle, namely 4 weeks, some process not yet understood leads to reduced action of estrogenic hormone, and bleeding ensues which we call anovulatory menstruation.


This view of the cycle requires, of course, much more investigation before we shall be able to understand the whole process; I have already pointed out the larger gaps in our knowledge. It certainly represents a great advance toward the truth; and what is indeed important, it gives us a far clearer and more hopeful viewpoint about the disorders of menstruation than do older concepts of the cycle. Since menstrual bleeding is caused by fluctuation in levels of the ovarian hormones, it follows that noncyclic, pathological bleeding, such as occurs in excessive and irregular periods may also be caused by abnormalities in amount proportion, or kind of these hormones. We must consider that there are not two sharply distinct kinds of functional bleeding, one being normal menstruation and the other abnormal hemorrhage. On the contrary, the modern hypothesis tells us that we must expect a series of types of hemorrhage ranging from normal menstruation through every grade of disturbance to the most severe disorder of the cycle. Gynecologists are already beginning to study and treat these distressing and difficult conditions in the light of this concept, and we may well hope these same hormones that control the normal cycle will help us to control its aberrations and at last to banish the specter of uterine hemorrhagic disease.


The Unknown Significance of Menstruation

In all this discussion of the nature and the course of menstruation, we have had nothing to say about the significance and possible usefulness of the periodic breakdown and hemorrhage. The human mind has an intractable desire that natural phenomena shall be useful. We are not comfortable in the presence of useless or undirected activity. Menstruation in particular ought to have a practical reason for its occurrence, for otherwise it seems a totally wasteful, destructive, and vexatious business. Up to the present, however, no one has been able to demonstrate such a meaning. There are in fact only two guesses that are even worth talking about.


The late Walter Heape of Cambridge, England, one of the pioneers in the study of the reproductive cycle, proposed in 1900 that menstruation is the same thing as a period of bleeding that occurs in female dogs when they are going into heat, i.e. in the week preceding ovulation. A somewhat similar proestrous bleeding occurs in cows, particularly in heifers. Such bleeding is easily explained, for it is clearly due to engorgement of the blood vessels produced by a strong action of the estrogenic hormone. Under the microscope it does not resemble menstruation ; the blood oozes from superficial blood vessels and there is little or no breakdown of the tissues. When Heape wrote, nothing was known of the time of ovulation in the primate cycle, nor of the premenstrual endometrium and its dependence upon the corpus luteum. In the primates, on his theory, ovulation would be expected to occur during menstruation, or immediately after the flow, just as in the bitch and cow ovulation occurs about the end of the proestrous bleeding. Since we know now that ovulation takes place a week to ten days after the cessation of menstruation, we can reconcile Heape's theory with the facts only by supposing that in the menstruating primates there is first proestrous bleeding, then a delay unknown in the other animals, and finally ovulation. This theory, and one or two ingenious variations upon the same theme by later English writers, F. H. A. Marshall and Zuckerman, all seem too complex to be probable, and moreover they suffer from a very serious objection. Unlike the menstruation-like bleeding, which can be produced in monkeys and women by withdrawing estrogenic hormone, the proestrous bleeding of dogs is produced by building up the level of estrogenic hormone, as was shown by R. K. Meyer and Seiichi Saiki in our Rochester laboratory in 1931.


A few years ago, Carl G. Hartman discovered that in Rhesus monkeys there is almost always a slight bleeding from the uterus about the time of ovulation. This does not show externally and is discernible only by applying the microscope to washings of the vagina. Sections of the uterus made at this time reveal that a few red blood cells are escaping from superficial capillary blood vessels of the endometrium, which are engorged by the action of the estrogenic hormone. In the laboratories we call this slight bleeding "Hartman's sign" and take it as evidence that there is a ripe follicle in the ovary. This is the actual equivalent of Heape's proestrous bleeding. It has since been found to occur in women, though probably not as regularly as in monkeys. These observations prove clearly that menstruation is something else than proestrous bleeding.


Hartman has proposed another explanation for menstruation. He points to the fact that in many mammals, at the time of implantation of the embryo, the uterine secretion contains red blood cells. He tells us also that in many viviparous animals lower than the mammals, for example certain salamanders and fish, in which the embryo depends upon the maternal tissues for nourishment, bleeding in one form or another usually occurs into the brood chamber. Hartman compares menstruation to bleeding of this kind and conjectures that it is simply a means of getting the vitally useful blood pigment, hemoglobin, into the region where the early embryo is to reside. We have to suppose that if in any given cycle an egg is fertilized, the premenstrual changes go to the very verge of menstruation, letting a little blood out of the vessels into the tissues to enrich the implantation site for the embryo. Since, however, a "missed period" is the first sign of beginning pregnancy, we must also suppose that attachment of the embryo then stops the process before external hemorrhage occurs, and even before there is any significant breakdown of the endometrium. If there is no embryo, the breakdown goes all the way. This is a very interesting conjecture, for it assigns a necessary and worthy function to the strange flux of menstruation. Careful study, however, of the uterine lining of the monkey just before and during implantation of the embryo, and of the very few human specimens during early implantation that are as yet available, does not support this hypothesis, for they do not show beginning hemorrhage in the endometrium. We know that many other mammals succeed in implanting their embryos without any such provision of free blood or hemoglobin in the endometrium; and we know also that sometimes in women and often in Rhesus monkeys menstruation occurs in anovulatory cycles, when there can be no embryos to profit by it. This hints that perhaps the process of menstruation evolved without reference to the embryo.


Menstruation, then, is still a paradox and a puzzle - a normal function that displays itself by destruction of tissues ; a phenomenon seemingly useless and even retrogressive, that exists only in the higher animals ; an unexplained turmoil in the otherwise serenely coordinated process of uterine function.

  1. Leslie B. Arey, "The degree of normal menstrual irregularity." American Journal of Obstetrics and Gynecology, vol. 37, pp. 12-29, 1939.
  2. For a discussion of this theory, see Carl R. Moore and Dorothy Price, "Gonad hormone function." American Journal of Anatomy, vol. 60, pp. 13-72, 1932, and Harold Lamport, "Periodic changes in blood estrogen." Endocrinology, vol. 27, pp. 673-680, 1942.
  3. We do not actually know the time of arrival of the human embryo in the uterus, nor the precise time of its implantation, since no normal human embryos younger than about 7 1/2 days have as yet been seen. Information from other animals, together with what is known of the human embryo at the 8th day, makes it highly probable that the human embryo becomes attached about the 7th day after ovulation.
  4. J. E. Markee, "Menstruation in intraocular endometrial transplants in the Rhesus monkey." Carnegie Institution of Washington, Publication No. 518 {Contributions to Embryology, vol. 28), pp. 219-308, 1940.
  5. George W. Corner, "Ovulation and menstruation in Macacus rhesus." Carnegie Institution of Washington, Publication No. S32 {Contributions to Embryology, vol. 15), pp. 75-101, 1923.
  6. Carl G. Hartman, "Studies in the reproduction of the monkey, Macacus (Pithecus) rhesus." Carnegie Institution of Washington, Publication No. 433 {Contributions to Embryology, vol. 23), pp. 1-161, 1932.


   Hormones in Human Reproduction (1942): 1 Higher Animals | 2 Human Egg and Organs | 3 Ovary as Timepiece | 4 Hormone of Preparation and Maturity | 5 Hormone for Gestation | 6 Menstrual Cycle | 7 Endocrine Arithmetic | 8 Hormones in Pregnancy | 9 Male Hormone | Appendices
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)

Cite this page: Hill, M.A. (2020, October 23) Embryology Book - The Hormones in Human Reproduction (1942) 6. Retrieved from https://embryology.med.unsw.edu.au/embryology/index.php/Book_-_The_Hormones_in_Human_Reproduction_(1942)_6

What Links Here?
© Dr Mark Hill 2020, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G