Book - Sex and internal secretions (1961) 9

Embryology - 21 Jun 2024        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)

Young WC. Sex and internal secretions. (1961) 3rd Eda. Williams and Wilkins. Baltimore.

Section A Biologic Basis of Sex Cytologic and Genetic Basis of Sex | Role of Hormones in the Differentiation of Sex

Section B The Hypophysis and the Gonadotrophic Hormones in Relation to Reproduction Morphology of the Hypophysis Related to Its Function | Physiology of the Anterior Hypophysis in Relation to Reproduction

The Mammalian Testis | The Accessory Reproductive Glands of Mammals | The Mammalian Ovary | The Mammalian Female Reproductive Cycle and Its Controlling Mechanisms | Action of Estrogen and Progesterone on the Reproductive Tract of Lower Primates | The Mammary Gland and Lactation | Some Problems of the Metabolism and Mechanism of Action of Steroid Sex Hormones | Nutritional Effects on Endocrine Secretions

Section D Biology of Sperm and Ova, Fertilization, Implantation, the Placenta, and Pregnancy Biology of Spermatozoa | Biology of Eggs and Implantation | Histochemistry and Electron Microscopy of the Placenta | Gestation

Section E Physiology of Reproduction in Submammalian Vertebrates Endocrinology of Reproduction in Cold-blooded Vertebrates | Endocrinology of Reproduction in Birds

Section F Hormonal Regulation of Reproductive Behavior The Hormones and Mating Behavior | Gonadal Hormones and Social Behavior in Infrahuman Vertebrates | Gonadal Hormones and Parental Behavior in Birds and Infrahuman Mammals | Sex Hormones and Other Variables in Human Eroticism | The Ontogenesis of Sexual Behavior in Man | Cultural Determinants of Sexual Behavior

Historic Disclaimer - information about historic embryology pages

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)

SECTION C Physiology of the Gonads and Accessory Organs

Action of Estrogen and Progesterone on the Reproductive Tract of Lower Primates

Frederick L. Hisaw, Ph.D.

The Biological Lab0Rat0Rif:S, Harvard Tniversity, Cambridge, Massachusetts

and

Frederick L. Hisaw, Jr., Ph.D.

Department Of Zoology, Oregon State College, Corvallis, Oregon

I. Introduction

Cyclic menstruation is the most characteristic feature of primate reproduction, and distinguishes it from the estrous cycle of lower mammals. This cardinal primate event is heralded by the bloody uterine effluent emanating from the vagina, whereas in estrus the dominant characteristic is a sudden modification in behavior featuring an intense mating drive. However, the internal secretions that regulate the various events in the menstrual and estrous cycles are the same, and this similarity is fundamentally more significant than the key descriptive differences just mentioned. Estrus comes at the peak of the growth phase of the cycle and is associated with ovulation. In contrast, menstruation occurs in the cycle midway between times of ovulation and is not accompanied by an increase in sexual activity. From earliest times menstruation has been recognized as degenerative: the characteristic odor, and the necrotic changes in the lining of the uterus, part of which is cast off at this time, sustain this interpretation. Therefore, menstruation is at the opposite phase of the cycle from estrus. It is such an obvious event that menstrual cycles are dated from the onset of bleeding. Menstruation is not analogous to the proestrous bleeding in the dog or cow nor to the slight bleeding of primates at midpoint between menstrual periods (Hartman, 1929). The study of menstruation was at first almost entirely the province of the clinician and the material for investigation limited to w^omen. Hitschmann and Adler (1907), Meyer (1911), Schroder (1914). Novak and Te Linde (1924), and Bartelmez (1933) are among many of the earlier investigators who contributed descriptions of the cyclic changes in the human endometrium. The physiology of the menstrual cycle and attendant morphologic changes have continued to be an area of active research interest in science and medicine. Among the many more recent contributors are Bartelmez (1937), Latz and Reiner (1942), Haman (1942), Knaus (1950), Mazer and Israel (1951), and Crossen (1953).

The earlier concepts regarding the menstrual cycle were based primarily on the changes occurring in the human endometrium and for convenience of description the cycle was divided into four stages or periods. The first of these was the period of active menstruation, and the length of the cycle was dated from its onset. Most authors agreed that menses began by leaking of blood from superficial vessels to form lakes under the surface epithelium and that there was some sloughing of tissue after the beginning of bleeding. There was considerable disagreement as to the amount of destruction and loss of tissue; estimates of various authors ranged from very little to almost complete denudation of the surface. Bartelmez (1933) emphasized both the wide individual variability of the amount of tissue lost and differences in the stage of development of the endometria at the time of menstruation.

The second period immediately following menstruation began with regeneration of the surface epithelium, which started sometimes before menstrual bleeding had ceased and was completed in a very short time. This lieriod included the 5 to 7 days after cessation of menses, during which the endometrium grew in thickness. Frequent mitoses were recognized, especially in the glands which lengthened but remained straight and tubular.

The third ("interval") period, lasting 6 to 10 days, was characterized by a somewhat thickened endometrium, still with straight glands and showing little evidence of secretory activity. At first this was considered a quiescent period as indicated by the term "interval." However, as will be shown later, such a characterization was not justified from the physiologic viewpoint.

The fourth period, called the premenstrual period, included the 10 days or 2 weeks before menstruation. During this phase the glands continued to increase in size and became distended, coiled, or even sacculated. The glandular cells increased in height, and there was evidence of glycogen mobilization and secretion. Next, the epithelium became "frayed out" along the outer borders, then decreased in height, indicating secretory depletion. Decidual cells appeared in the stroma at this time. The endometrium was much thickened and extremely hyperemic. At the height of this period the endometrium was approximately 5 mm. in thickness, as compared with V2 mm. toward the end of menses. The term premenstrual was usually applied to this phase but today the term progestational would seem preferable.

During and after these descriptions of the changes in the human endometrium, many attempts were made to locate the time of ovulation in the menstrual cycle. When it was found, as will be discussed later, that ovulation occurred approximately midway between two menses, and was preceded by follicular growth and followed by development of a corpus luteum, it became customary to refer to the two halves of the menstrual cycle as the follicular phase and the luteal phase. One advantage of this descriptive terminology was the emphasis it placed on the homology of the two phases of the menstrual cycle in primates with the follicular and luteal phases of the estrous cycles of lower mammals.

A theory to explain menstruation, widely adopted in 1920, was formulated from this morphologic evidence. The essentials were that menstruation occurs because the lining of the uterus, prepared for implantation of the ovum, degenerates if fertilization of the egg does not occur. This required that ovulation and corpus luteum formation precede the i^remenstrual changes in the endometrium. Subsequent research disclosed that menstrual cycles frequently occur in which ovulation does not take place and bleeding results from the breakdown of an "interval" rather than a progestational endometrium.

The discovery of anovulatory cycles not only brought about a revision of ideas regarding an explanation for menstruation but also raised questions as to what constituted a normal menstrual cycle. The length of the cycle and the amount and duration of bleeding are approximately the same regardless of whether or not ovulation has taken place. The gross features of menstruation under these two conditions are indistinguishable one from the other. However, the biologic purpose of the menstrual cycle is reproduction wliicli obviously cannot l)0 fulfilled unless an ovum is made available for fertilization. Therefore, in this sense it seems quite clear that anovulatory cycles should be considered incomplete and abnormal.

The investigation of changes taking place in the uterine endometrium at various periods of the menstrual cycle in women was confronted with many difficulties, the chief one being that of obtaining normal tissue representative of specific times of the cycle. The entire uterus and both ovaries are essential for proper evaluations and it was rarely possible to meet these requirements. The material for such studies came from autopsies and surgery and tissues usually had suffered postmortem changes or the surgical condition was one involving serious pelvic disease. There have been, however, a goodly number of instances in which these difficulties were adequately overcome (Stieve, 1926, 1942, 1943, 1944; Allen, Pratt, Newell and Bland, 1930) and the clinic will continue to make important contributions (Rock and Hertig, 1942; Hertig and Rock, 1944), but quite early the need became obvious for a suitable primate that could be used as an experimental animal for research on the different aspects of the physiology of reproduction.

Since the initial observations by Corner (1923) on the menstrual cycles of captive rhesus monkeys iMacaca mulatta) , more has been learned about the physiology of reproduction of this animal than any other primate. Monkeys of this species thrive under laboratory conditions, which has made it possible to devise accurately controlled experiments on normal healthy animals and obtain reliable information on the menstrual cycle, gestation, fetal development, and the interaction of hormones concerned with regulating reproductive processes.

Other features that make the rhesus monkey such an attractive animal for these purposes are the many morphologic and physiologic attributes that are strikingly like those of the human being. Tiic modal length of their menstrual cycles is 28 days but there is wide variation (Corner, 1923; Hartman, 1932; Zuckcrman, 1937a). From an analysis of 1000 cycles recorded for some 80 females of different ages, Zuckerman (1937a) found an average cycle length of 33.5 ± 0.6 days, and the mode 28 days with an over-all range of 9 to 200 days. Ovulation occtu's api:)roximately midway between two menstrual periods, most between the 11th and 14th days (Hartman, 1932, 1944; van Wagenen, 1945, 1947), and although these animals breed at all seasons of the year many cycles are anovulatory, especially during the hot summer months (Eckstein and Zuckerman, 1956). A method developed by Hartman for detecting the exact time of ovulation by palpation of the ovaries in the unanesthetized animal greatly facilitated the timing of events of the menstrual cycle. This procedure also made it possible to determine the age of corpora lutea with great accuracy (Corner, 1942, 1945) and correlate their develojiment and involution with corresponding changes in the endometrium (Bartelmez, 1951 ) and, in ju-egnancy, with the exact age of developing embryos ( Wislocki and Streeter, 1938; Heuser and Streeter, 1941).

The primary purpose of the present discussion is to review the results of experimental investigations of physiologic processes occurring in the female reproductive tract of lower primates during the menstrual cycle, and particularly those processes that are under hormonal control. The brief introductory presentation of basic observations could be greatly extended and we take up the discussion of endocrine problems knowing that we must return often to the work of these authors and that of others to be cited, as conclusions based on experimental data take on meaning only in terms of normal function.

II. Ovarian Hormones and Growth of the Genital Tract

The changes that are repeated in different parts of the reproductive tract with each menstrual cycle are produced by ovarian hormones, estrogens, and progesterone. The dominant hormone of the follicular phase is estradiol-17/y, which is secreted by the Graafian follicle, and in the tissues is readily transformed in i)art to estrone, an estrogenic metabolite. Progesterone, secreted by the corinis luteum, is jirimarily a hormone of the luteal phase of the cycle. However, small amounts of progesterone may appear ill the blood of monkeys as early as the 7th (lay and attain a concentration of 1 fxg. per ml. of serum at ovulation, whereas a maximal concentration of 10 /xg. per ml. is reached at approximately the middle of the luteal phase (Forbes, Hooker and Pfeiffer, 1950; Bryans, 1951). Also, some estrogen is present during the luteal phase, probably secreted by the corpus luteum (estrogens can be obtained from luteal tissue) or it may be partially derived from developing follicles. It is unlikely that estrogen is ever entirely absent during a normal menstrual cycle or that the presence of progesterone is completely restricted to the luteal phase.

The dependence of the reproductive tract on ovarian hormones is strikingly demonstrated by the profound atrophy that follows surgical removal of the ovaries. A progressive decrease in size of the Fallopian tubes, uterus, cervix, and vagina takes place, and usually, the involution of the uterine endometrium involves tissue loss and bleeding, commonly referred to as post castrational bleeding. Dramatic as these effects may seem, it is equally dramatic to find that these atrophic structures can be restored entirely to their original condition by the administration of ovarian hormones. Therefore, it is seen at the beginning that investigations dealing with the physiology of the female reproductive tract of primates in large measure involve a study of the independent and combined actions of estrogens and progesterone on the activities of the various structures concerned.

Much can be learned about the action of ovarian hormones by observing the changes they produce in the gross appearance of the atrophied reproductive tract of castrated animals. Daily injections of an estrogen in doses equivalent to 1000 I.U. or more for 10 days or a fortnight will restore the uterus, cervix, and vagina to a condition comparal)le with that found in a normal monkey at the close of the follicular phase of a menstrual cycle (Fig. 9.1A). If a similar castrated monkey is given 1 or 2 mg. of progesterone daily for the same length of time, very little if any change in size of the reproductive organs results. However, if first the normal condition is restored by giving estrogen and then is followed by the progesterone treatment, the size of the uterus is maintained but that of the cervix and vagina decreases to an extent approaching that in a castrated animal (Fig. 9.1B). Such experiments show that an estrogen promotes growth of the reproductive tract whereas progesterone is comparatively ineffective when given alone. Yet progesterone can maintain the size of the uterus when administered following an estrogen treatment but it does not prevent involution of the cervix and vagina.

An additional feature of the growth-stimulating action of the ovarian hormones is brought out when estrogen and progesterone are administered concurrently. If, after repair of the reproductive tract of a castrated animal has been accomplished by a series of injections of estrogen, both estrogen (1000 I.U.) and progesterone (1 or 2 mg.) are given daily for 20 days, it will be found that the uterus is larger than when either hormone is given alone for a similar length of time, whereas the cervix and vagina have involuted and are approximately the size found in animals given only progesterone. Thus it can be demonstrated that a synergistic effect on growth of the uterus occurs when the two hormones are given simultane

FiG. 9.1. Reproductive tracts of three adolescent monkeys which were castrated and given estrogen daily for approximately 3 weeks. A shows the condition at the conclusion of the estrogen treatment, B the condition following the injection of progesterone for an additional three weeks, and C the effects of continuing the treatment with both estrogen and progesterone for a like period.

ously whereas in the cervix and vagina the growth -stimulating action of estrogen is inhibited by progesterone (Fig. 9.1C). This presents a most interesting situation in which the combined actions of two hormones on three closely associated structures of the reproductive tract join in a synergistic effort in promoting the growth of the uterus although progesterone prevents estrogen from affecting growth of the cervix and vagina.

These changes in gross morphology in response to the ovarian hormones are reflected in the histology of the responding tissues. Also, the character of the response differs depending upon the physiologic nature of the tissue concerned; therefore, for the sake of clarity each will be discussed separately. The first of these to be considered is the uterus and particularly growth of the endometrium.

It has been reported (Hisaw, 1935, 1950) that growth of the endometrium, as induced by estrogen, is limited. That is, a dosage of estrogen capable of maintaining the endometrium of a castrated animal for an indefinite period without the occurrence of bleeding, stimulates rapid growth for approximately the first 2 weeks. Within this time a maximal thickness of the endometrium is attained which remains constant or may become less during the course of treatment (Fig. 9.2). Engle and Smith (1935) made similar observations. They found that the endometria of castrated monkeys receiving estrogen for 100 days or longer were thinner than endometria of animals on estrogen for a much shorter time. Also, on prolonged treatment, the stroma of the endometrium becomes dense and the lumen small whereas the size of the uterus remains about the same. In fact, they state that in

Fig.,. (1.2. I'll n Ml lour i-a.-^tra(c>d monkeys wliicli were givi^i 10 /ug. estradiol daily for 10 to 78 days. A was given estrogen for 10 days, B for 30 days, C for 60 days, and D for 78 days. Depression in the endometrium of anterior wall of D is the result of a biopsy taken a year previously. (From F. L. Hisaw, in A Si/niposium on Steroid Hormones, University of Wisconsin Press, 1950.)

four experimental animals the only well developed fundus was found in the animal on the shortest treatment, i.e., 60 days.

The mitotic activity in the epithelium of the glands and surface mucosa also indicates a limited effect of estrogen. This can be demonstrated to best advantage in the endometria of castrated monkeys that have been on estrogen for different lengths of time and have received an injection of colchicine 8 hours before their uteri were removed. A comparison of the number of cells in mitosis per square centimeter of surface mucosa at 10, 30, 45, and 60 days is shown in Figure 9.3. From this it can be seen that mitotic activity approaches that in a castrated animal. Although the five points used in drawing the curve are quite inadequate for an accurate analysis of the mitotic response in the epithelial components of the endometrium, they do show that cell division is most rapid soon after the beginning of an estrogen treatment and subseciuently declines.

The loss of responsiveness of the endometrium to estrogen seems related more to the length of treatment than to dosage of hormone. An endometrium of normal thickness can be produced in 2 or 3 weeks at a dosage level of estrogen that will not maintain the growth induced for longer than about 40 days without bleeding (Hisaw, 1935; Engle and Smith, 1935; Zuckerman, 1937b). The response to a low dosage of estrogen that will prevent bleeding during the course of treatment (about 10 /xg. estradiol-17/8 daily) is one of rapid endometrial growth at first, as has been described, followed by a thinning of the endometrium. The refractoriness of the endometrium to estrogen becomes so pronounced after about 100 days of treatment that very few cell divisions are seen in the epithelium of the glands and surface mucosa. The general morphology of the endometrium retains the characteristic appearance of the follicular phase of the menstrual cycle except that the stroma is usually more dense and the cells of the glandular epithelium have large deposits of glycogen between the nucleus and the basement membrane. However, metabolically such endometria are surprisingly inactive. Although they are dependent on the presence of estrogen and may bleed within about

Fig. 9.3. The number of mitoses per square centimeter of surface epithelium of the endometrium in a castrated monkey and in four castrated animals given 10 /xg. estradiol daily for 10, 30, 45, and 60 days, respectively. One-tenth of actual number of mitoses is shown on the ordinate. (From F. L. Hisaw, in A Symposium on Steroid Hormones, University of Wisconsin Press, 1950.)

48 hours if the treatment is stopped, the activity of their oxidative enzymes and the ratio of nucleoproteins (RNA:DNA) are about the same as in the involuted endometria of castrated animals.

The effects that accompany moderate estrogenic stimulation become exaggerated in several respects when large doses of estrogen are given for an extended period. The disparity between the area of myometrium and endometrium becomes greater as the treatment progresses (Fig. 9.4). Kaiser (1947) described the destruction of the spiraled arterioles of the endometrium in monkeys given large doses of estrogen and Hartman, Geschickter and Speert (1941) reported the reduction of the reproductive tract to the size of that of a juvenile animal by the end of 18 months during which injections of large doses of estrogen were supplemented by subcutaneous implantation of estrogen pellets. These observations not only show that the endometrium becomes unresponsive to estrogen when the treatment is prolonged but that large doses produce injurious effects.

Fig. 9.4. .4. ri:,,ii of the uterus of a castrated monkey which had received 1.0 mg. estradiol daily for 35 days. Compare with B which shows the effects of 1/10 this dosage (100 fig. daily) when gi\en for 185 days.

The limited response of the endometrium to estrogen is in some respects surprising in view of its remarkable growth potentialities and regenerative capacity. These qualities were dramatically demonstrated by Hartman (1944) who dissected out as carefully as possible all of the endometrium from the uterus of a monkey and wiped the uterine cavity with a rough swab and yet the undetected endometrial fragments that remained were capable of restoring the entire structure. Also, considering the enormous increase in size of the uterus during gestation, it is even more difficult to account for the rather sharp limitation of growtli under the influence of estrogen.

The increase in tonus of the uterine musculature, a known effect of estrogen, has been considered as possibly exercising a restrictive influence on growth of the endometrium. An attempt has been made to remove this containing influence the muscle may have by making an incision through the anterior wall of the uterus (Hisaw, 1950) . A castrated monkey was given 10 ixg. estradiol daily for 21 days at which time the operation was performed and the treatment continued with 30 ^g. estradiol daily for 40 days. The uterus was laid open by a sagittal incision from fundus to cervix and most of the endometrium was removed from the anterior wall. This caused gaping of the incised uterus and exposure of the endometrium on the posterior wall. The incision was not closed and after hemorrhage was completely controlled the uterus was returned to the abdomen.

Examination of the uterus at the conclusion of the experiment showed no indications that endometrial growth had been enhanced. The muscularis had reunited and only a few small bits of endometrium were found in the incision (Fig. 9.5). It seemed probable that the purpose of the experiment had been defeated by rapid repair of the uterus. Therefore, a similar experiment was done in which the musculature of the incised uterus was held open by suturing a wire loop into the incision. Yet the incision closed and no unusual growth of the endometrium was detected (Fig. 9.6).

Observations under these conditions are necessarily limited to those made on the uterus when it is removed at the conclusion of an experiment and comparisons must be made between uteri of different animals. Obviously, it would be more desirable if the response of an individual endometrium could be followed during the course of treatment. It is possible to meet most of these requirements under conditions afforded by utero-abdominal fistulae, exteriorized uteri, and endometrial implants in the anterior chamber of the eye. In continuing our discussion we first shall present information obtained by such techni(iues that have a bearing on the response of the endometrium to estrogen.

The uteri of these two castrated monkeys were opened from fundus to cer\'ix by an incision through the anterior wall while the animals were receiving estrogen. Part of the endometrium of the anterior wall was removed and the incision in the myometrium was not closed.

Fig. 9.5. Estradiol, 10 fig., was given daily for 7 days; the uterus was opened and the animal continued on 30 /xg. estradiol daily for 40 days. (From F. L. Hisaw, in A Sy7nposiii7n on Steroid Hormones, University of Wisconsin Press, 1950.)

Fig. 9.6. Estradiol, 10 ^g., was given daily for 7 days; the uterus was opened and the treatment continued at a dosage of 30 /ig. estradiol daily for 20 days. (From F. L. Hisaw, in A Symposium on Steroid Hormones, University of Wisconsin Press, 1950.)

The surgical procedure used by Hisaw ( 19501 for preparing utero-abdominal fistulae foi' studies of the exj^erimental induction of endometrial growth by estrogen and progesterone was a modification of that used by van Wagenen and Morse (1940) for observing changes in the endometrium (luring the normal menstrual cycle. This procedure makes frequent inspections possible either by hand lens or dissecting microscope, of most of the upper part of the endometrium on the anterior and posterior walls of the uterus. The elliptic slit formed by the endometrium of the two opposing walls can be located easily, the two sides pressed apart by any small smooth instrument, and the surface of the endometrium examined. Changes in thickness of the endometrium cannot be ascertained without resorting to biopsies but it is free to grow out of the opened uterus if it is so inclined. However, in such preparations the growth produced in the endometrium by daily injections of 10 fjig. estradiol for periods of 2 or 3 weeks is not sufficient to show any tendency whatever to grow out through the fistular opening or obstruct examination of the walls of the uterus. The limited growth observed in these experiments is in agreement with that obtained with estrogen on intact and incised uteri.

The cervix uteri of the rhesus monkey is sufficiently long to make it possible to bring the entire fundus to the exterior through a midal)dominal incision. Advantage of this

was taken in an attempt to exteriorize the uterus and maintain it outside the body for long enough periods to make it possible to study the growth responses of the endometrium (Hisaw, 1950). These preparations did not i)rove satisfactory in all respects but they did contribute a number of interesting observations.

The operational i^rocedure used in these experiments involved dividing the uterus transversely from fundus to cervix so that the anterior wall was deflected downward and the posterior wall upward (Fig. 9.7). The endometrium of the exteriorized uterus is difficult to maintain but with proper care it seems to retain its normal condition for at least the first few days after the uterus is opened. Small localized areas of ischemia can be seen to come and go, probably action of the coiled arteries, and there is a periodic general blanching of the endometrium associated with rhythmic contractions of the muscularis. This, however, does not seem true of the whole endometrium. A zone surrounding the internal os of the cervix tends to retain its blood-red color even during strong contractions of the uterus and the growth reactions of the endometrium in this area are of particular interest.

Fig. 9.7. Exteriorized uteii. The uteri were divided transversely from fundus to cervix. The anterior half is seen deflected to the right and the posterior half to the left. A and B are of the same uterus taken 13 days after exteriorization showing the "blush" and "blanch" reaction of uterine contractions. It can be seen that during blanching the endometrium of the cervix does not become ischemic. C. Uterus 18 days after exteriorization, showing response to estrogen. Ridges formed by growth of the endometrium surrounding the internal OS of the cervix can be seen at the upper edge of the photograph. D, taken 83 days after exteriorization, shows response produced by a series of injections of 10 ^ig. estradiol and 1 mg. progesterone daily. The transverse ridge is formed by the two opposed lips of endometrium derived from the area surrounding the internal os of the cervix. Growth when the two hormones are given is greater than when only estrogen is injected. (From F. L. Hisaw, in A Symposium on Steroid Hormones, University of Wisconsin Press, 1950.)

The endometrium on the exposed anterior and posterior halves of the uterus underwent deterioration despite the best of care that could be given, but that surrounding the internal OS of the cervix survived and retained its capacity to grow, in one animal, for as long as 9 months. When estrogen was given this endometrium grew rapidly and within a few days stood out as large elliptic lips surrounding the internal os (Fig. 9.7). Within 2 to 3 weeks the lips appeared to reach their full size and further growth was slow or absent. When estrogen treatments were discontinued the endometrial lips underwent bleeding within a few days and were entirely lost. At no time were activities observed that could be ascribed to coiled arterioles, nor did ischemia occur during involution previous to bleeding. It seems that the response of this tissue to estrogen is like that found in other experiments but the absence of ischemia preceding bleeding is exceptional. The endometrium on the anterior and posterior walls of uterine fistulae invariably showed ischemia for several hours before active bleeding following the withdrawal of estrogen.

Markee (1940) approached the problem of endometrial growth in monkeys by studying the changes that occur in bits of endometrial tissue transplanted to the anterior chamber of the eye. Such transplants retain in large measure the normal morphology of endometrial tissue and changes in their cyclic growth parallel those going on simultaneously in the uterus. So much so that if the animal has an ovulatory cycle, the ocular implants show conditions characteristic of both the follicular and luteal phases, but if ovulation fails to occur then the luteal phase is omitted. Also, the morphologic events taking place at menstruation can be seen and recorded, since the transplants regress and bleed at each menstrual period.

These ingenious experiments will be referred to often in the course of our discussion but at present the response of endometrial transplants in the eye to estrogen is of primary interest.

Monkeys having ocular transplants were given 200 to 300 R.U. of estrone daily for about 1 to 3 months. The transplants did not grow to a certain size and then remain stationary, but instead periods of rapid growth were interrupted by periods of regression which usually involved a marked decrease in size, and if regression was extensive and rapid, bleeding ensued. It also was found that these episodes of regression in the transplants were usually accompanied by a decrease in the size of the uterus.

Comparisons between the results of these experiments and those we have discussed previously may be misleading since it seems that only 1 of the 5 animals (no. 295) used was castrated. Also, the dosage of estrogen was not sufficient to maintain the endometrium of the uterus for an indefinite period without bleeding and this also was reflected in the transplants. It seems questionable that the growth capacity of endometrial transplants in the eye can be determined unless sufficient estrogen is given to prevent bleeding in the uterus. Therefore, it would seem that these experiments contribute less to an analysis of the effects of estrogen on endometrial growth than they do to an understanding of the events that precede and accompany menstruation.

In summary, it seems clear that the outstanding effect of estrogen on the uterus of the monkey is one of growth (Allen, 1927, 1928). The involuted uterus of a castrated animal can be restored to its normal size in 2 or 3 weeks by daily injections of adequate amounts of estrogen. At this time there is an increase in vascularity, a clear-cut hyperemia as seen in rodents. There also is secretion of luminal fluid (Sturgis, 1942) but this does not distend the uterus as in the mouse and rat. This is accompanied by an increase in tissue fluid, especially in epithelial tissues (surface epithelium and glands) , and in the connective tissue of the stroma. Glycogen may be present at the basal ends of epithelial cells beneath the nuclei (Overholser and Nelson, 1936) but it apparently is not readily released under the action of estrogen alone (Lendrum and Hisaw, 1936; Engle and Smith, 1938) . The glands of the endometrium maintain a straight tubular structure with some branching near the muscle layers. The condition produced experimentally in the monkey's uterus by short term treatments with estrogen is equivalent to that present in the normal animal at midcycle, or even a few days later if ovulation does not occur.

If, however, an estrogen treatment is continued for several months conditions develop in the uterus that are not found during the follicular phase of a normal menstrual cycle. When the daily dose of estrogen is small menstruation occurs at intervals during the treatment (Zuckerman, 1937b) and probably marks periods of endometrial regression as observed by Markee (1940) in eye transplants, but if the dosage is increased by a sufficient amount (about 10 /xg. estradiol- 17/3 daily) injections may be continued for a year or longer without bleeding. Although the size of the uterus remains within the range of normal variation as the injections are continued, the myometrium tends to increase in thickness and the endometrium becomes thinner, a condition not corrected by further increases in dosage or by prolonging the treatment. The cause responsible for the limited response of the endometrium under these conditions is not known but apparently is not a restrictive influence of the myometrium as similar responses are given when the endometrium is exposed by incising the uterus, in abdominal fistulae, and in exteriorized uteri.

III. Effects of Progesterone on the Uterus

It has been mentioned that a menstrual cycle, in which ovulation occurs, can be conveniently divided into a follicular and a luteal phase. The follicular phase extends from menstruation to ovulation and the luteal phase from ovulation to the following menstruation. It has been shown in the previous discussion that the endometrial modifications characteristic of the follicular phase of the cycle can be duplicated in a castrated monkey by the injection of estrogen. Likewise, the progestational condition characteristic of the luteal phase can be developed by giving progesterone. In fact, all the morphologic and physiologic features that are known for anovulatory and ovulatory cycles can be reproduced in castrated monkeys by estrogen and progesterone.

If one designs an experiment to simulate the normal cycle in a castrated monkey then estrogen should be given first to develop the conditions of the follicular phase followed by progesterone for the progestational development of the luteal phase. Experience has shown that this is the most effective procedure for the production of a progestational endometrium. Progesterone, as compared with estrogen, is a weak growth jH'omoter and although it can produce progestational changes in the atrophic endometrium of a castrated monkey when given in large doses, its action is greatly facilitated when preceded by estrogen. The first experiments in which progesterone was used for this purpose were planned on this principle (Hisaw, Meyer and Fevold, 1930; Hisaw, 1935; Engle, Smith and Shelesnyak, 1935).

The first noticeable effect of progesterone is an elongation of the epithelial cells of the surface membrane and necks of the glands. When the treatment is continued, this effect progresses down the gland towards the base. This change is followed closely by a rearrangement of the nuclei which is more pronounced in the glands than in the surface epithelium. The nuclei under the influence of estrogen in doses which reproduce the conditions of the follicular phase of a normal cvcle, are situated niostlv in the basal half of the cells, some of them touching the basement membrane. The nuclei retreat from the basement membrane when progesterone is given leaving a conspicuous clear zone. This zone is produced by intracellular deposits of glycogen. These early changes usually appear before pronounced spiraling and dilation of the glands.

Fig. 9.8. Uterus of a castratefl monkey which was given 2 mg. progesterone daily tor 113 days. The endometrium is thin but bleeding occiu\s when such treatment is stopped. The myometrimn is soft and pliable and ilir l)lood vessels are cnlarsed and have thick wails.

Secretion begins in response to estrogenic stimulation and increases greatly as progestational changes are established. It appears first in the necks of the glands and progresses basalward. The surface epithelium takes a less conspicuous part in secretion and is usually reduced to a thin membrane when injections of progesterone are continued until a fully developed progestational endometrium is established. This progressive action of progesterone is such that it is possible to find all conditions in a single gland from active secretion and fraying in the neck region through primary swelling to an unmodified condition at the base.

When treatment is continued for 25 to 30 days at doses of about 2.0 mg. daily, the glands enter a state that has been called "secretory exhaustion" (Hisaw, 1935). This condition also is seen first in the necks of the glands and progresses toward the base. The glandular epithelium decreases in thickness, and active secretion, as judged by fraying of the cells, is absent. The glands may become narrow and straight and the endometrium may resemble that in castration atrophy. These involutionary changes become even more pronounced if the treatment is continued for several months or a year (Fig. 9.8). The endometrium by this time is extremely thin. The glands are straight, short, and narrow, and the stroma very dense. The myometrium is thick in proportion to the endometrium and the uterine blood vessels are large and have greatly thickened walls. Such uteri tend to be somewhat smaller than normal and are soft and pHable.

Thus, it is seen that when growth is produced in the endonietiiuni of a castrated monkey by giving estrogen and then continued on injections of progesterone, there follows a sequential development of all stages of the luteal phase of a normal menstiual cycle terminating in secretory exhaustion. However, this condition cannot be maintained by continuing the progesterone treatment, and involutionary processes set in and the endometrium is reduced to a thin structure. Yet, such degenerate endometria are dependent upon progesterone and will bleed within about 48 hours if the injections are stopped. It also was found that after discontinuence of progesterone daily injections of 10 /i.g. estradiol may not prevent bleeding.

IV. Synergism between Estrogen and Progesterone

There is considerable evidence that in primates progesterone under normal conditions rarely if ever produces its effects in the absence of estrogen. Large quantities of estrogen are present in human corpora lutea (Allen, Pratt, Newell and Bland, 1930) and during pregnancy the placenta secretes estrogens as well as progesterone (Diczfalusy, 1953) . This apparently is a common feature of primates, as indicated by the excretion of estrogens in the urine of pregnant chimpanzees and rhesus monkeys (Allen, Diddle, Burford and Elder, 1936; Fish, Young and Dorfman, 1941 ; Dorfman and van Wagenen, 1941). Also, correlated with this is the observation that estrogen and progesterone when given concurrently produce a greater effect on the uterus of castrated monkeys than either alone (Hisaw, Greep and Fevold, 1937; Engle, 1937; Hisaw and Greep, 1938; Engle and Smith, 1938) and that an ineffective dose of progesterone is greatly potentiated by estrogen. This synergistic effect of the two hormones on the uterus of monkeys is quite different from their action on the uteri of laboratory rodents and rabbits. In these animals the effects of progesterone can be inhibited quite easily by a surprisingly small dose of estrogen (see chapter 7).

The synergism between estrogen and progesterone in the promotion of endometrial growth can be demonstrated to best advantage under the conditions of some of the physiologic preparations that have been discussed. For instance, it was shown (Fig. 9.5) that growth of the endometrium under the influence of estrogen was not enhanced by relieving muscle tension by a midline incision through the anterior wall of the uterus. Now, if a similar operation is performed on the uterus of a monkey that is receiving 10 /tg. estradiol daily and the treatment continued with the addition of a daily dose of 1 mg. progesterone, there usually follows a rapid growth of endometrial tissue out through the incision until by about 3 weeks a mass is formed which approximates the size of the entire uterus (Fig. 9.9). If this experiment is repeated and the same dosage of progesterone is given without estrogen, there is no outgrowth of the endometrium (Fig. 9.10).

FiG. 9.9. Uterus of a castrated monkey that received 10 fig. estradiol and 1 mg. progesterone daily for 18 days, at which time the uterus was opened from fundus to cervix and most of the endometrium of the anterior wall removed. The incision was not closed and the treatment was continued for an additional 20 days. (From F. L. Hisaw, in A Symposium on Steroid Hormones, University of Wisconsin Press, 1950.)

A similar synergistic action can be seen in utero-abdominal fistulae. We have mentioned that estrogen does not cause excessive growth of the endometrium under these conditions. However, endometria that have reached their maximal response to estrogen will show a resumption of growth if 1 or 2 mg. progesterone are added daily to the treatment. By the 4th or 5th day lobes of blood-red endometrium begin to protrude through the opening of the fistula. Within a few days tongue-like processes of endometrial tissue are thrust out of the opening with each uterine contraction and are entirely or i^artially withdrawn at each relaxation.

Fig. 9.10. Uteru.s of a castrated monkey which was given 1 mg. progesteione (lail>' for 18 days following an estrogen treatment. The uterus was opened as described for Figure 9.9, and the injections of progesterone continued for 20 days. (From F. L. Hisaw, in A Syiyiposium on Steroid Hormones, University of Wisconsin Press, 1950.)

Such outgrowths are difficult to protect from mechanical injury and consequent tissue loss so it is not possible to determine accurately how much endometrium is produced in a given time. In one experiment an animal was kept on 10 fxg. estradiol and 2 mg. progesterone daily for 98 days and it was found that the endometrium continued to grow, but the rate seemed considerably slow^er toward the conclusion of the treatment than at the beginning. How long an endometrium would continue to grow under these conditions was not determined, but it is obvious that much more endometrial tissue was produced by the treatment than is ever found at one time in the uterus of a monkey during a normal menstrual cycle. This takes on added significance when it is compared with the endometrial response in the intact uterus of an animal given the same dosage of estrogen and progesterone for a similar length of time.

The progestational development of the endometrium, when both hormones are given, passes through the same stages as those following the injection of only progesterone; i.e., presecretory swelling of the glandular epithelium, active secretion, and secretory exhaustion. The endometrium, however, is considerably thicker than when a comparable dose of progesterone is given alone, and secretory exhaustion may not be so pronounced by the 30th day (Fig. 9.11). The glandular epithelium in the necks of the glands may be reduced to a thin membrane scarcely thicker than the nuclei whereas some secretion is usually present in the dilated basal parts of the glands. Also dilation of the glands in the basalis is more pronounced following a 30-day estrogen-])rogesterone treatment than when the same amount of progesterone is given separately.

Secretory exhaustion appears to be the initial indication of an involutionary process that ensues when an estrogen-progesterone treatment is continued for a long time (Hisaw, 1950). When a combination of the two hormones, known to be capable of producing a large uterus with a thick, fully develojied, progestational endometrium within al)out 20 days, is given for 100 days, an astonishingly different endometrium results (Fig. 9.12). It is thin, the stroma is dense and the narrow straight glands are reduced to cords of cells in the basal area. The condition is one suggesting inactivity and atrophy.

When such dosages of estrogen and i)rogesterone are given to castrated monkeys for 200 days or a year further changes in the endometrium occur. By 200 days the epithelium of the surface mucosa and glands is lost except for small glandular vestiges along the musciilaris at the base of the endometrium. There are no glands, coiled arteries, or large blood vessels in what one might yet call the functionalis. All that remains is a modified stroma that resembles decidual tissue (Fig. 9.13.4 and B). It is also of interest that these endometria will menstruate if the treatment is discontinued and in most if the injections of progesterone are stopped and estrogen continued, but not if estrogen is stopped and progesterone continued.

Fk;. 9.11, .\ late i)r()ges1ati()iial condition produced in the endometrium of a castrated monkey by giving 10 (ig. estradiol daily for 18 days followed by 10 /xg. estradiol and 2 mg. progesterone daily for 31 davs.

Fig. 9.12. The endometnuin of a castrated monkey that had received 10 /xg. estradiol and 1 mg. progesterone daily for 99 days.

Even though in such experiments the endometrium has been under the influence of both estrogen and progesterone for a year and has undergone extremely abnormal modification, it yet is capable of responding to estrogen in a more or less characteristic way when progesterone is stopped and in

jections of estrogen continued. Apparently within about three weeks the modified endometrium is replaced, under the influence of estrogen, by one that has few glands which tend to be cystic, a mesenchymatous stroma, and no coiled arteries (Fig. 9.14).

Under similar circumstances, if estrogen is stopped and jjrogesterone is continued, the modified endometrium is lost without bleeding and there is almost no repair of the endometrium even after a period of 3 weeks. There seems to be an incompatability between the epithelial outgrowths from the mouths of the glands and the underlying stroma of the denuded surface. Consequently the epithelium crumbles away and epithelization of the raw surface is not accomplished (Fig. 9.15j. How long this condition could continue has not been determined.

Fig. 9.13. The endometrium shown m .4 is (h;i( from a castraled mdiik.N wlml, l,a,l received 10 /xg- estradiol and 2 mg. progesterone daily for 200 days. In B, jiart of ilie endometrium of a snndai animal given the same treatment for 312 days is shown at a higher magnification. The endometrium is almost entirely a modified stroma in which glandular epithelium and coiled arteries are absent. Only vestiges of glands are present in the basal area next to the myometrium.

One of the most interesting aspects of these observations is that these effects were jiroduced by dosages of estrogen and progesterone that are very probably within the range of normal physiology. From this it appears that although growth of the endometrium is greater when the two hormones are given together, due to their synergistic interaction, this does not prevent involutionary changes from setting in when the treatment is continued for a period of weeks or months. In fact, greater damage to the endometrium occurs under the simultaneous action of the two hormones than when either is given alone. Also, increasing the dose intensifies the damaging action of both estrogen and progesterone, so much so that very large doses will almost completely destroy the endometrium.

The myometrium, however, shows a different response to these treatments. Estrogen stimulates myometrial growth, which is

Fig. 9.14. Uterus of a castrated monkey which was given 10 ixg. of estradiol and 2 mg. progesterone daily for 307 days at which time the injections of progesterone were stopped and estrogen continued for 20 days. Bleeding occurred the second day following discontinuance of progesterone. The absence of coiled arteries and the presence of cystic glands and a mesenchymetous stroma characterize the endometrium.

Fig. 9.15. Uterus of a castrated monkey which was given 10 yug. estradiol and 2 mg. progesterone daily for 275 days at which time estrogen was stopped and progesterone was continued for 21 days. A shows the thin endometrium and dense stroma whereas B shows failure of formation of a surface epithelium following the loss of the modified functionalis presumably present at the conclusion of treatment with both hormono.^^ (see Fig. 9.13).

intensified both by cln-onic treatment and high dosage, and seems to be equally effective when it is given alone or in combination with progesterone. Progesterone also promotes growth of the muscularis but seems less effective than estrogen and differs from it by causing pronounced thickening of the walls of the arcuate blood vessels. These vascular changes extend to the coiled arteries of the endometrium, which are also affected by high dosages of estrogen. It seems remarkable that estrogen is capable of preventing the action of progesterone on the myometrial blood vessels and correcting such effects after they are produced and yet at the same time it assists in the destruction of the coiled arteries in the endometrium.

V. Experimentally Produced Implantation Reactions

Progestational endometria of the normal menstrual cycle or those produced in castrated monkeys by progesterone, if mechanically traumatized, will develop endometrial proliferations which seem identical with those found at normal implantation sites of fertilized ova (Figs. 9.16 and 9.171 (Hisaw, 1935; Hisaw, Creep and Fevold, 1937; Wislocki and Streeter, 1938; Rossman, 1940). The proliferated cells originate from the surface and glandular epithelium and grow into the surrounding stroma. The reaction spreads from the point of injury and within a few days may involve the entire inner l)ortion of the endometrium bordering the lumen. The implantation plaques on the 3rd or 4th day present a fairly homogeneous appearance but soon thereafter certain cells attain the proportions of giant cells and many are multinucleated.

The development of the plaques is most rapid during the first week, by the end of which cell division is found only in the basal half of the proliferation and evidence of regression is seen in the superficial portion adjoining the uterine lumen. After 10 days degenerative and phagocytic processes are the dominant features and by 24 days the ut'prus contains few or no ]iroliferation cells. Wislocki and Streeter ( 1938,1 found that implantation plaques during pregnancy and those experimentally induced underwent ajjl^roximately the same development arid subsequent degeneration except for modifications produced by the invading trophoblast. Rossman (1940j made an extensive morphologic study of these epithelial proliferations and concluded that they should be regarded as typical metaplasias \vith an embryotrophic function.

Fig. 9.16. An area of the normal implantation site of a developing ovum. (From Carnegie Institution, No. C467.)

Fig. 9.17. An experimentally induced implantation reaction in a castrated monkey showing condition 6 days after mechanical traumatization of the endometrium.

VI. The Cervix Uteri

The cervix uteri of the rhesus monkey is remarkable for its size and complexity. It forms a large segment that is set off from the fundus by a conspicuous constriction at the level of the internal os (Fig. 9.1). A sagittal section (Fig. 9.18) shows the cervical canal not straight but thrown into several sharp turns by colliculi that extend from its walls into the lumen. The largest of these projects from the midventral wall. The functional advantage of such tortuosity of the cervical canal is not obvious but since the cervix probably serves as a barrier between the bacterial flora of the vagina and the corpus uteri, this may be a useful adaptation.

The physiology of the cervix has received much less attention than has been given the uterus. This is regrettable in view of the consideration it must receive in practical obstetrics and gynecology, as well as the possibility that physiologically the monkey cervix may be homologous with that of the human regardless of morphologic difTerences. Recent observations indicate that this is indeed quite probable.

Fig. 9.18. Sagittal section of the cervix from a normal monkey. The vagina and the external os of the cervix are shown at the left and the entrance to the fundus is at the right.

Fig. 9.19. Sagittal section of the cervix of a pregnant monkey showing conditions present just previous to parturition on the 154th day of gestation. The dominant features are dilation of the cervical canal and reduction of the cervical lips (shown at the left) and the coUiculi. (From Carnegie Institution, No. C713.)

Hamilton (1949) made a detailed study of the changes in the cervix of rhesus monkeys during the menstrual cycle, paying particular attention to alterations that took place in the cells of the surface epithelium of the endocervical canal and the cervical glands. It was found that heights of the cells showed consistent increases and decreases during the cycle. The peaks came on the 3rd, 13th to 15th, and 22nd days, the greatest of these being the 14th day which is approximately the time of ovulation. It also was observed that, following a peak, secretion was associated with the decline.

Attention was called b3^ Hamilton to the rather close correlation between the fluctuations in height of the cervical epithelium in monkeys and the fluctuations observed by Markee and Berg (1944) in the blood estrogens of the human menstrual cycle. It was concluded that, if similar changes in estrogen levels also occur in monkeys, one would be justified in concluding that the increase in cell height in the cervical mucosa was due to the action of estrogen and the sudden periodic drops in blood estrogen caused secretion and consequent regression. However, it is not clear how this could account for the abundant secretion of the cervical glands in the presence of high levels of estrogen during late pregnancy (Fig. 9.19).

Much has been learned regarding the physiology of the primate cervix from experiments on castrated monkeys. The cervical mucosa is very responsive to estrogen and castration atrophy can be repaired and a normal condition maintained by daily injections of small doses. Cervical secretion may become abundant when an estrogen treatment is prolonged and especially if large doses are injected. However, the amount of secretion induced by estrogen never equals that of the last half of pregnancy, and it usually subsides if the injections are continued for several months.

Under conditions of chronic treatments with estrogen metaplastic aberrations invariably appear in the epithelium of the endocervix. This reaction was first reported in monkeys by Overholser and Allen ( 1933, 1935) and has been confirmed by many investigators (Engle and Smith, 1935; Hisaw and Lendrum, 1936; Zuckerman, 1937c (. Similar lesions may be found in the cervix uteri of women (Fluhmann, 1954). They seem especially prone to occur under conditions characterized by excessive production of estrogen, such as hyperplasia of the endometrium (Hellman, Rosenthal, Kistner and Gordon, 1954) and granulosa-cell tumors of the ovary. Various degrees of metaplasia may occur in the cervix during pregnancy both in the mother and newborn but Fluhmann (1954) did not find it as frequently as in nonpregnant women.

This reaction to estrogen as seen in the cervix of castrated monkeys is initiated by growth of small undifferentiated cells below the columnar mucous cells of the secretory epithelium. Fluhmann (1954) suggests that these cells are really undifferentiated cells of the cervical mucosa which have the potentiality of becoming columnar or squamous or simply undergoing multiplication and remaining as indifferent or reserve cells. These cells accumulate, in response to estrogen, to form aggregates of several cells in thickness and, although this may occur in any area of the endocervix, it is generally more pronounced below the base of the glands. As this process proceeds the columnar mucous cells are pushed outward and are finally desquamated thus exposing the underlying metaplastic cells to the lumen of the gland (Fig. 9.20).

Fig. 9.20. Al.iaph

in a castrated monkey that li.-id rci-civcd 1 nig. estriol daily for 48 days.

The cells of these lesions undergo a characteristic differentiation. When first formed they are small, cuboidal, and have spherical nuclei with dense chromatin. As they increase in number those in the center of the cellular mass become larger and acquire an eosinophilic cytoplasm. Such collections, as seen at the base of the cervical glands, may grow in height and form cone-shaped masses with the apexes protruding through the mucous epithelium into the lumen or they may remain as more or less compact structures. This difference in growth seems to have a general relation to the dosage of estrogen. Large doses cause more rapid growth and cone formation with the loss of cells from the apex either singly or in groups, whereas small doses produce slower growth and desquamated cells are seldom seen in the lumen. However, regardless of the rate of growth, the cells at the base of the lesion remain undifferentiated and continue as the principal area of cell proliferation.

Pearl formation is occasionally seen and may be quite common in animals on low dosages of estrogen. Under strong estrogenic stimulation and consequently rapid growth, these structures apparently are desquamated before they are completely formed. However, very early stages are frequently seen and may even be present in small clumps of metaplastic cells, but they are more commonly found in the larger collections at the base of the glands. Their appearance is initiated by swelling and disintegration of one or more adjacent cells that form a center around which epidermidization takes place. Further development does not proceed under the influence of esti'ogen, beyond the formation of a small central cavity.

The most conspicuous difference between the metaplastic growths produced by estrogen and true cancer of the cervix in the monkey (Hisaw and Hisaw, Jr., 1958) is that the former remain noninvasive even when the treatment is continued well over a year. They also involute when the treatiiiciit is discontinued and they do not appeal' when progesterone is given simultaneously with estrogen. When the injections of progesterone are started after metaplastic growths have been formed in response to estrogen, further growth is inhibited and the keratinized cells of the lesion become vacuolated and are lost.

In contrast with the effects of estrogen on the cervix, the modifications that occur as pregnancy advances are remarkable. The cervix becomes a soft thin-walled structure, the glands increase in number, and their lumina become greatly enlarged, pressing the stroma into thin partitions between them, and the amount of mucus secreted is enormous (Fig. 9.19). Attempts at duplicating these changes in castrated animals by hormone therapy have been only partially successful. Estrogen produces a solid thickwalled cervix that tends to be larger than normal, an effect that is especially noticeable in young animals. Progesterone does not promote cervical growth and repair of the glands unless large doses are given and even then there is little if any secretion. The best results were obtained when both estrogen and progesterone were given and especially so when relaxin was added to the treatment (see chapter by Zarrow).

VII. The Vagina

The general features of the vaginal smear of rhesus monkeys have been described by several investigators (Allen, 1927; Hartman, 1932; Westman, 1932) and a detailed study of the cellular components at different times of the menstrual cycle has been made by Lopez Columbo de Allende, Shorr and Hartman (1945). The changes in the vagina of a monkey are in most respects like those found for the human being (Papanicolaou, Traut and Marchetti, 1948; Lopez Columbo de Allende and Orias, 1950). Epithelial growth and desquamation of cornified cells continue at all stages of the cycle but at various rates. The epithelium is thinnest at menstruation and gradually increases in thickness during the follicular phase, reaching a maximum at ovulation. At this time there is a well developed basal area in which numerous mitoses can be seen and from which many papillae or bulbs" extend into the underlying stroma. Above this is an intermediate zone, an interepithelial zone of cornification (so called Dierk's layer), and a heavily cornified outer zone (Fig. 9.21).

Cellular proliferation is less rapid during the luteal phase and apparently cells are desquamated more rapidly than they are replaced. Consequently there is a decrease in the thickness of the epithelium in the luteal phase which may include an almost complete loss of the cornified zone (Davis and Hartman, 1935). The effects are probably due to progesterone because similar changes are seen following the introduction of progesterone into a treatment in which estrogen is being given.

The vaginal epithelium of a castrated monkey is remarkably sensitive to estrogen. A small daily dose of 5 to 10 /xg. estradiol will stimulate growth of an atrophic epithelium of 4 to 8 cells in thickness to one of 60 or even 80 layers thick within 3 weeks. One of the first things that is noticed as the vaginal epithelium thickens is the numerous mitotic figures in the stratum germinativum followed by a marked increase in the number of epithelial papillae along the basement membrane. This condition of rapid growth, cornification, and loss of cells into the vaginal lumen is typical of the follicular phase of the menstrual cycle and can be maintained indefinitely.

Fig. U.21. the vaginal epithelium of a castuUMJ monkej' showing growth antl cornification induced by estrogen.

Progesterone, in contrast with estrogen, does not produce rapid growth of the vaginal epithelium but at the same time it is not without an effect. The vaginal epithelium, weeks or months after castration, has relatively few papillae projecting from its basal border into the underlying stroma. When progesterone is given, this condition is changed but not in a spectacular way. There is very slow growth without cornification. The epithelium remains thin but the papillae become more numerous. These are mostly small epithelial buds which tend to remain solid but may show enlargement of the cells in their centers.

When estrogen and progesterone are given concurrently, the effects of estrogen on the vaginal mucosa are modified. If an estrogen treatment has continued for a sufficient time to produce full cornification and then progesterone is added, the first indication of an inhibition of estrogen is a decrease in mitotic activity. This is followed by a continuation of cornification and loss of cells faster than they are replaced ; consequently, most of the functionalis is lost and the epithelium becomes thinner. There is also a noticeable decrease in the intensity of cornification, which in the monkey is never as pronounced as in rodents, and under these conditions is quite incomplete, each cell retaining a conspicuous nucleus. Partly cornified cells may be present for several weeks when both estrogen and progesterone are given, but eventually they almost entirely disappear and the epithelium attains a condition resembling that of late pregnancy.

Fig. 9.22. ^^•tgi^al epithelium of a pregnant monkey showing condition on the 154th day of gestation. (From Carnegie Institution, No. C713.)

The inhibitory effect of progesterone on the action of estrogen is shown perhaps even better when a castrated monkey having a fully involuted reproductive tract is first given progesterone for a few days and then (>strogen is added to the treatment, or when injections of the two hormones are started at the same time. In such experiments estrogen has little effect on the vaginal mucosa even in doses that would produce marked cornification if given alone. These observations show that a fully cornified vaginal epithelium cannot be produced or maintained by estrogen when an effective dosage of progesterone is included in the treatment (Hisaw, Greep and Fevold, 1937).

Estrogens and progesterone are the dominant hormones of gestation and their simultaneous action is reflected by the changes in the vaginal epithelium. The fully cornified vagina, present at the time of ovulation, is gradually modified as pregnancy progresses into a condition strikingly like that seen in experiments when estrogen and progesterone are given concurrently. In late pregnancy the most striking feature of the thin, uncornified epithelium is the presence of numerous epithelial buds extending deeply into the underlying stroma. They may branch and rebranch and along their course there is conspicuous enlargement of the more centrally situated cells among which cavities ai^pear, enlarge, and join each other (Fig. 9.22). It seems quite probable that this process may be of considerable importance in increasing the diameter of the vagina.

VIII. Sexual Skin

A so-called sexual skin is jiresent in most catarrhine monkeys, is not found in platyriliine monkeys, and among the anthropoids occurs regularly only in the chimpanzee I l^]ckstein and Zuckcrman, 1956). Changes in the sexual skin during the menstrual cycle have been observed most extensively in the monkey (Macaca), the baboon (Papio), and the chimpanzee (Pan). The sexual skin of t!ie baboon and chimpanzee undergo pronounced swelling during the follicular phase of the cycle. A maximal size is attained by the middle of the cycle followed by a rapid regression and loss of edema which at least in the baboon is associated with a marked increase in the output of urine (Gillman, 1937a; Krohn and Zuckerman, 1937). The subsidence of the sexual skin begins approximately at the time of ovulation and remains in the reduced condition throughout the luteal phase, followed by a subsequent initiation of swelling during or soon after menstruation (Zuckerman, 1930, 1937e; Zuckerman and Parkes, 1932; Gillman and Gilbert, 1946; Young and Yerkes, 1943; Nissen and Yerkes, 1943).

A well developed sexual skin is present in the monkey {Macaca mulatta) only during adolescence. With the appearance of the menstrual cycles the sexual skin undergoes a process of maturation into the adult condition in which cyclic changes in edema are absent and the most noticeable feature is a vivid red color. Such coloration is due to vascular engorgement rather than pigment (Collings, 1926) and involves the perineum, the buttocks, and may extend for various distances down the legs and over the symphysis pubis. The development and maturation of the sexual skin have been described in considerable detail by several investigators (Hartman, 1932; Zuckerman, van Wagenen and Gardiner, 1938) .

The sexual skin has been of considerable interest both as to the nature of its responsiveness to ovarian hormones and the manner in which its grossly visible changes during the menstrual cycle parallel events occurring in the reproductive tract. The sudden loss of edema at the conclusion of the follicular phase not only signals ovulation but also raises the question as to whether the loss of tissue fluid is due to a decrease in estrogen or is the direct effect of progesterone. The importance of this becomes obvious when it is considered that a similar process also goes on simultaneously in the endometrium and raises the question again as to the respective roles played by estrogen and progesterone in endometrial growth and menstruation.

That the development and edema of the sexual skin of adolescent rhesus monkeys depend on the ovaries was first demonstrated by Allen ( 1927 ) . Involution and loss of color follow castration, and the normal condition can be restored by the injection of estrogen. Also, when estrogen treatment is continued for several weeks maturation of the sexual skin occurs and a condition characteristic of that in the adult is established (Zuckerman, van Wagenen and Gardiner, 1938). The genital area loses its edema and develops a brilliant red color which is retained as long as estrogen is administered. Once this mature condition is established the response of the sexual skin to subsequent estrogen treatments is limited to a change in color.

Similar experiments have been performed on the chacma baboon, Papio porcarius (Parkes and Zuckerman, 1931; Gillman, 1937b, 1938, 1940a). The large sexual skin of these animals is very responsive to estrogen and development equal to that of the follicular phase of the menstrual cycle can be readily induced by daily injections for about 2 weeks. However, the perineal swelling of the baboon differs from the sexual skin of the genital area of the rhesus monkey in that it does not "mature" under the influence of estrogen.

When large doses of estrogen are given to a rhesus monkey a generalized edema of the skin occurs beyond the genital area. This first appears as deeply indented swellings along the sartorii from groin to knee, and next appears at the base of the tail and spreads gradually upward until it involves the entire dorsal portion of the trunk. At the same time, the skin of the face, scalp, and supraorbital ridges becomes swollen and finally the edema may extend out on the arms and down the legs to the ankles (Bachman, Collip and Selye, 1935; Hartman, Geschickter and Speert, 1941). A daily dose of 500 /xg. or more of estriol or estradiol w^ll produce this condition within 2 to 3 weeks and, when the treatment is continued for an extended period the effect tends to subside.

Progesterone has a strong inhibitory action on the effects produced by estrogen on both the genital and extragenital sexual skin of the monkey. If daily injections of progesterone are added to the treatmeiu after full development of the sexual skin has been induced by estrogen, there is a noticeable loss of edema by the 4th or 5th day followed by rapid involution and reduction of the turgid folds of skin to loose, flabby wrinkles within about 10 days. When estrogen and progesterone are given concurrently to a castrated monkey from the beginning of treatment edema does not appear but the sexual skin regains its normal color. In fact, progesterone alone, like estrogen, can restore the color to the sexual skin of castrated adult monkeys (Hisaw, Greep and Fevold, 1937; Hisaw, 1942).

The interaction of estrogen and i)rogesterone on the sexual skin of rhesus monkeys can best be demonstrated by the reaction of the skin of the sexual area in adolescent animals. The most striking effect and probably the most important is the sequence of events initiated by a single dose of progesterone when given to an animal on continuous estrogen treatment. Under such treatment a full response of the sexual skin is obtained by the end of 20 days. If at this time 1 mg. progesterone is given in a single dose and the estrogen treatment continued uninterruptedly, the first indication of an effect of the luteal hormone is a slight loss of edema and color of the sexual skin on the 4th or 5th day thereafter. The sexual skin is markedly reduced by the 8th day, almost gone by the 9th, and at the end of about a fortnight regains its ability to respond to estrogen as shown by a return of color and swelling. However, the most remarkable eventuation of such treatment is menstruation which usually begins on about the 10th day (Hisaw, 1942).

Involution of the sexual skin and menstruation following a single injection of progesterone also have been produced in the baboon by Gillman (1940a). He found that 5 mg. progesterone, when given on the 8th day of a normal menstrual cycle, would cause an appreciable loss of edema of the swollen perineal sexual skin by the day after injection. This was followed by a progressive involution of the perineum until the 13th day and swelling was re-initiated by the end of the 15th day. Reduction of the sexual skin at this dosage of progesterone was not associated with menstruation. However, when the dose was increased to 20 mg. both deturgescence of the sexual skin and menstruation occurred. These effects produced by progesterone in the presence of endogenous estrogen have much in common with those described above as occurring in castrated monkeys on continuous estrogen treatments.

IX. Menstruation

An experimental ai^proach to the physiology of menstruation dates from the observations of Allen (1927) that uterine bleeding would occur in castrated monkeys following the discontinuance of an estrogen treatment. He suggested that normal menstruation is due to a fluctuation in estrogen secretion and proposed the "estrogen-withdrawal" theory to account for the observed facts. This concept led to an extensive investigation of the effects of estrogens on the endometrium and of conditions that modify their action. It was soon found that in both castrated monkeys and human beings there was a quantitative relationship between the dosage of estrogen given and the maintenance of the endometrium. Bleeding occurred during treatment when the daily dose of estrogen was small, but with larger doses a point was reached at which the injections could be continued for months or even years without bleeding (Werner and Collier, 1933; Zuckerman, 1937b, d).

Estrogen also will inliihit i)ostop('rative bleeding which usually follows total castration, provided the ovaries are removed before or soon after ovulation (Hartman. 1934). With the advent of a corpus luteum and development of a progestational endometrium it becomes progressively more difficult, following castration, to prevent menstruation by injecting estrogen. Similar results are obtained when estrogen is given during a normal menstrual cycle. Small doses may not prevent the onset of menstruation, but if continued, subsequent menstrual periods are delayed (Corner, 1935). Large doses when given during the luteal phase of the cycle do not disturb the normal menstrual rhythm, but may do so if the treatment is started during the follicular phase (Zuckerman, 1935. 1936a).

Progesterone, in contrast with estrogen, will prevent menstruation from an endometrium representative of any stage of the normal cycle. It will delay onset of the next menses even when the treatment is started only a few days before the expected menstruation (Corner, 1935; Corner and Allen, 1936) . Also, the bleeding that invariably follows the discontinuance of a long treatment with estrogen can be inhibited indefinitely by giving progesterone (Hisaw, 1935; Engle, Smith and Shelesnvak, 1935; Zuckerman, 1936b).

An impression held by many of the earlier investigators was that progesterone could not produce its effects on the primate endometrium unless it w^as preceded by the action of estrogen. It is true, of course, that progesterone is a comparatively weak growth promoter and its effects can be demonstrated to best advantage on an endometrium that has been developed by estrogen. However, Hisaw, Greep and Fevold (1937) produced a progestational endometrium in a monkey that had been castrated 242 days previously by giving synthetic progesterone. Also, the endometrium of this animal was found capable of forming a decidual plaque upon traumatization. Soon afterwards Hartman and Speert (1941) observed menstruation following the withdrawal of progesterone in castrated monkeys that had not been given estrogen and more recently similar results have been reported by Eckstein ( 1950) . At the same time it has l)een found that progesterone will induce menstruation in women suffering from amenorrhea and also that uterine bleeding can l)e jirecipitated l)y similar treatment (hiring the follicular j^hase of the cycle (Zondek and Rozin, 1938; Rakoff, 1946).

These observations have been confirmed and extended by Krohn (1951; 1955) who finds that menstrual bleeding can be induced in monkeys wdth secondary amenorrhea by the injection of 5 daily doses of progesterone. Progesterone (5 mg. daily for 5 days) also precipitates uterine bleeding in castrated monkeys at intervals of about 8 days provided the treatment is started innnediately a menstrual bleeding has been induced either by removel of the ovaries or withdrawal of estrogen. The most interesting aspect of these observations is that the number of short 8-day cycles that can be obtained in this way in a castrated animal seems to be related to the size of the initial dose of estrogen used to induce withdrawal bleeding. This also applies to progesterone-withdrawal bleeding, so the effect does not depend upon the particular hormone used to obtain the bleeding. It also is of interest that such conditioning of the endometrium to subsequent responses to the 5-day treatments with progesterone may last for several months on a continuous regime. It is surprising that such a series of responses cannot be initiated unless the first injection of progesterone is given within 6 days following the initial withdrawal bleeding. These observations have much in common with those of Phelps (1947) who also studied the influence of previous treatment on experimental menstruation in monkeys.

There seems to be a quantitative relationship between the dosage of progesterone given in combination with estrogen and the ability of estrogen to prevent bleeding after the injections of progesterone are stopped. It has been mentioned that once a fully developed i^rogestational reaction has been produc(Hl l)y progesterone, it is extremely difficult, if not impossible, to inhibit menstruation by giving estrogen following the withdrawal of progesterone. However, Hisaw and Greep (1938) found that progestational endometria produced ijy small doses of estrogen plus api^roximately 0.5 mg. progesterone daily for 18 to 21 days did not bleed following progesterone withdrawal when continued on 10 to 20 times the original dosage of estrogen. In fact, such endometria were brought back to a condition typical for the action of estrogen and again transformed into a presecretory progestational state without the intervention of bleeding. Similar observations were made previously by Zuckerman (1936a, 1937d).

These experimental results give grounds for some doubt as to the adequacy of the estrogen-withdrawal theory to account fully for menstruation. Not only can progesterone bring about menstruation without the intervention of estrogen but other steroid hormones are capable of pi'oducing similar effects. Desoxycorticosterone in large doses can inhibit estrogen-withdrawal bleeding in castrated monkeys (Zuckerman, 1939, 1951 ) and induce phases of uterine bleeding in rapid succession in normal monkeys (Krohn, 1951). So too can testosterone prevent estrogen-withdrawal bleeding (Hartman, 1937; Engle and Smith, 1939; Duncan, Allen and Hamilton, 1941) and inhibit progesterone-withdrawal bleeding as well (Engle and Smith, 1939). Testosterone also will precipitate bleeding during an estrogen treatment (Hisaw, 1943) and in normal monkeys if given early in the cycle (Krohn, 1951). Just what specific action these compounds have in common that enables them to produce these effects or whether there are different modes of action that lead to the same results is not known, but, before mentioning certain possibilities, it may be helpful to consider information regarding the influence of estrogen-progesterone interactions on menstruation.

Among the most significant observations regarding primary causes of menstruation are a few indications that there may be an intrinsic difference in the ways in which estrogen and progesterone produce their effects on the endometrium. One of the first indications of this was the discovery that a short series of injections of progesterone during treatment with estrogen will precipitate menstruation (Corner, 1937; Zuckerman, 1937d; Hisaw and Greep, 1938). This can be demonstrated by giving a castrated monkey a maintenance dose of estrogen daily for 2 or 3 weeks, then adding a daily injection of progesterone for 5 to 10 days and continuing the estrogen treatment. As a rule bleeding appears within 2 or 3 days after stopping progesterone. The most interesting point brought out by such experiments is that bleeding can occur under these conditions in the presence of an otherwise maintenance dosage of estrogen.

Perhaps the most surprising as well as most important fact brought out by subsequent experiments was the small amount of progesterone required to bring about bleeding under these conditions. It was found that only a single injection of 1 mg. was required for animals on chronic treatment with a maintenance dose of estradiol (1000 LIT.) and some bled when 0.5 mg. progesterone was given (Hisaw, 1942). The sequence of events following the injection of progesterone can be seen to best advantage in an adolescent monkey whose sexual skin also respond" to the estrogen treatment. When the 1 mg. ]irogesterone is given on the 20th day of estrogen treatment the edema of the sexual skin will have attained its maximal development. The first indication of an effect of progesterone is a slight loss of edema and color of the sexual skin which appears on the 4th or 5th day and by the 9th or 10th day the edema is almost gone and the sexual skin is pale. Blood apjiears in the vaginal lavage between the 7th and 10th days, of about 70 per cent of the animals on this dosage. The sexual skin may remain markedly reduced and pale until about the 15th day after which both color and edema rapidly return. These effects can also be seen when 1 mg. progesterone is given for a series of days. However, neither the time of appearance nor loss of edema of the sexual skin is significantly hastened, and if the injections extend over no more than 5 days the time between the first injection and bleeding remains approximately the same.

Similar observations have been made by Gillman and Smyth (1939) on the South African baboon iPapio porcarius). They found that 3 mg. or more of progesterone when given in a single injection during the follicular phase of the cycle would cause the relatively enormous perineal swellings to pass rapidly through deturgescence and reach a flabby resting condition within 5 to 7 days, and after a delay of about 24 hours once again begin to swell. As much as 10 or 15 mg. in a single dose caused perineal deturgescence without bleeding, whereas 20 mg. in a single dose or a total of 15 mg. if divided into 2 or 3 injections and given at 3 or 4 day intervals, produced both deturgescence and bleeding (Gillman, 1940b). The l)aboon diff"ers from the monkey in that larger doses of progesterone are required to produce the effects and the sexual skin does not mature" on repeated treatments and lose its responsiveness; otherwise the basic physiology of the reaction in both animals seems to be the same.

The most important fact l)rought out by these experiments is that the effects of a single injection of progesterone can continue in the presence of estrogen for as long as 10 to 15 days. It is highly imi^robable that progesterone lingers in the body for so long a time (Zarrow, Shoger and Lazo-Wasem, 1954). In general it is considered the most ephemeral of the sex steroids and is probablv inactivated within at least a few hours after it is administered. It seems more lilvely that progesterone modifies the sexual skin in a way that renders it unresponsive to estrogen and that about a fortnight is required to recover the original condition.

This takes on added significance when the possibility is considered that effects similar to those seen in the sexual skin might also be going on simultaneously in the uterine endometrium. An appreciable dehydration of the endometrium occurs just previous to menstruation (van Dyke and Ch'en, 1936) and a loss of interstitial fluid before bleeding has been observed in endometrial implants in the eyes of monkeys and described in detail by Markee (1940) . This was shown by periodic regression in size and compactness of the grafts which resulted in a decrease in area of 25 to more than 75 per cent. Because cyclic changes in endometrial grafts in the eye are correlated with events of the menstrual cycle there is reason to believe that similar reactions were going on in the endometrium of the uterus.

Endometrial regression, as described by Markee, did not always lead to menstruation although it invariably preceded, accompanied, and followed menstrual bleeding. Menstruation occurred only when regression was rapid and extensive. This was seen in the endometrial grafts in the eye during a normal menstrual cycle at the time of involution of a corpus luteum and during an anovulatory cycle soon after the involution of a large follicle. It also begins soon after the last of a series of injections of estrogen or i^'ogesterone. A slow decrease in size of the ocular grafts, without concomitant bleeding, can be induced in castrated monkeys by gradual withdrawal of estrogen, and when estrogen is given in amounts that are inadequate for maintaining the endometrium for an extended period the "break through" bleeding that eventually ensues is preceded by a rapid and extensive endometrial regression. Because this reaction also occurs w^hen menstruation is induced by such an unusual procedure as spinal transection (Markee, Davis and Hinsey, 1936), it probably is a phenomenon that always precedes menstruation.

It seems from these observations that the changes in the endometrium preceding menstruation are initiated by a sudden withdrawal of a stimulus on which the endometrium at the time relies for the maintenance of a particular physiologic condition, and bleeding and tissue loss are incidents that occur during the readjustment necessary for the return to an inactive state. What this involves is only partly known, but an understanding of the initial changes in the endometrium that usher in menstruation most certainly holds the explanation of the real cause. This has been a perennial subject for discussion and many suggestions and theories have been set forth in an extensive literature to account for various aspects of menstruation. Among the more recent general discussions are those by Zuckerman (1949, 1951), Corner (19511, and Zondek ( 1954 ) .

The estrogen- withdrawal or estrogen-deprivation theory proposed by Edgar Allen has received more attention than any other. From what has been mentioned earlier it is clear that this theory can account for uterine bleeding subsequent to the discontinuance of a series of estrogen injections and also perhaps menstruation at the conclusion of an anovulatory cycle. However, it is not so obvious as to how this theory can explain the occurrence of menstruation at the close of the luteal phase of a normal cycle. Estrogen in large doses will not inhibit such bleeding, but it is postponed if progesterone is given. It is equally difficult to see how this theory is helpful in accounting for the fact that a small dose of progesterone will precipitate bleeding in the presence of a maintenance dosage of estrogen. As little as 2 /xg. progesterone will induce bleeding when applied topically to the endometrial lips of an exteriorized uterus (Fig. 9.7) in a monkey that is receiving 10 fig. estradiol daily (Hisaw, 1950).

Uterine bleeding precipitated by administering progesterone during an estrogen treatment has been explained on the grounds that progesterone in some way interferes with the action of estrogen on the endometrium. Therefore, it is assumed that an animal receiving both estrogen and progesterone is in a sense "deprived" of estrogen. That is, when the two hormones are given simultaneously, progesterone itself is capable of maintaining the endometrium without bleeding; but when it is stopped, the suggestion is that the animal is physiologically deprived of estrogen and literally deprived of progesterone (Corner, 1951). Although this view is in descriptive agreement with the observed facts the idea of the inhibitory effect of progesterone does not take into consideration the synergistic interaction of the two hormones on the endometrium.

The physiologic function of progesterone is the conversion of an estrogen-endometrium into a progestational endometrium suitable for receiving and nourishing a developing blastocyst. Such an endometrium is adapted for this specific reproductive function and accordingly its physiologic nature must be quite different from that of the follicular phase of the cycle. Indeed, it is known that these two structures (follicular and luteal phase endometrial are morphologically and biochemically unlike in a number of respects. This gradual transformation, following ovulation, occurs as progesterone becomes the dominant hormone, and consequently, as this proceeds, the endometrium progressively loses competence to respond to estrogen. However, this does not imply that estrogen is without effect in the general economy of the progestational endometrium. It has been shown in a number of ways that the action of progesterone on the primate endometrium is greatly facilitated by the presence of estrogen. In fact, it seems probable that rarely if ever does progesterone perform its function in the absence of estrogen (Hisaw, 1959; chapter by Zarrow).

After consideration of the endometrial specializations brought about by ]irogesterone, it seems rather jwintless to hark back to the follicular phase and inject the past recoi'd of accomjilishments and prerogatives that estrogen had at that time into the explanation of an entirely different hormonal situation. It seems more in keeping with the facts to state outright that menstruation following the involution of a corpus luteum or the discontinuance of progesterone, even though estrogen is present, is due to a decrease or absence of progesterone.

It also has become less certain that menstruation at the conclusion of an anovulatory cycle is really an estrogen-withdrawal bleeding. This is possible, of couisc, but at the same time the exceedingly small amount of progesterone required to induce bleeding in the presence of estrogen makes it difficult to be sure what the situation might be. Even a negative test for progesterone in the blood, by our present methods, does not necessarily indicate the absence of a physiologically effective amount of progesterone. Zarrow, Shoger and Lazo-Wasem (1954) found that in rabbits an intramuscular injection of 40 mg. progesterone was required to produce an appreciable concentration of the hormone in the blood as determined by the Hooker-Forbes method. Yet, 0.2 mg. progesterone daily for 5 days will produce a progestational reaction in the uterus equivalent to that of the 5th day of normal pseudopregnancy. In monkeys 0.5 mg. daily when given with 10 fxg. estradiol is an adequate dosage of progesterone to induce unquestionable progestational changes in the endometrium and much less will cause bleeding. These observations indicate that the minimal effective concentration of progesterone in the blood may be less than is possible to detect by our present methods.

This also seems to hold for the human being. Estimates of secretion and metabolism of progesterone in the human being have been based primarily on the recovery of its excretory product sodium pregnanediol glucuronidate in the urine. It seems obvious that such determinations must be only general approximations because only about 20 per cent of the progesterone secreted or injected can be accounted for by the pregnanediol in the urine. Also, it is generally known that a physiologically effective dosage of progesterone does not necessarily lead to the excretion of pregnanediol (Hamblen, Cuylcr, Powell, Ashley and Baptist, 1939; Seegar, 1940). In other words, the threshold dose of progesterone for endometrial stimulation is l)elow that at which the hormone is excreted as pregnanediol. In fact, it has been suggested by some investigators that there is no quantitative relationship between the l)rogesterone present in the blood and the pregnanediol excreted in the urine (Buxton, 1940; Sommerville and Marrian, 1950; Kaufmann, Westphal and Zander, 1951).

These findings and the wide variation in the amount of prc'gnancdiol excreted during a menstrual cycle (Venning and Browne, 1937) suggest that, even in the absence of ovulation, sufficient progesterone may be present to influence menstruation. There also is the possibility of progestational hormone from some extra-ovarian source, such as the suprarenal cortex. This was suggested by Zuckerman (1937b, 1941 j as a possible explanation for periodic bleeding in monkeys on a constant submaintenance dose of estrogen. This thought becomes more plausible in view of the fact that progesterone is one of the precursors in the metabolic synthesis of androgens, estrogens, and adrenal cortical steroids (Dorfman, 1956). Also, it has been shown that desoxycorticosterone acetate is converted to progesterone in vivo (Zarrow, Hisaw and Bryans, 1950). Therefore, progesterone is not restricted to ovarian luteal function but instead is of rather general occurrence in the body and the likelihood is that small amounts are a constant constituent of the blood.

Also, the amount of progesterone from extra-ovarian sources may fluctuate, as suggested by Zuckerman (1949), and consequently disturb the normal menstrual rhythm and probably cause bleeding in monkeys on a continuous submaintenance dose of estrogen. However, as to the latter, there is an alternative explanation. Castrated monkeys on a continuous treatment of 10 fxg. of estradiol daily do not show "break-through" bleeding, and a synergistic effect on growth of the uterus is seen when 0.5 mg. or more of progesterone daily is introduced into the treatment. However, the simultaneous administration of 0.25 mg. or even 0.125 mg. progesterone daily in similar exj^eriments results in bleeding between a!)out the 10th to 16th day of the combination treatment. Thus, a dosage of progesterone less than that required for synergism or prevention of bleeding when given alone, modifies the endometrium so that it can no longer be maintained by 10 fxg. estradiol daily (Hisaw, Jr., unpublished). When it is considered that the endometrium becomes increasingly dependent on estrogen during a chronic treatment, even after maximal growth is attained (Hisaw, 1942), it seems plausible that the effectiveness of a dosage of estrogen only slightly alcove the threshold for bleeding may be decreased sufficiently by the endogenous progesterone from extra-ovarian sources to precipitate bleeding.

Although it is obvious that the normal menstrual cycle is primarily under the control of the ovarian estrogens and progesterone, it is also equally clear that menstruation is not due to a specific hormonal action. Experimental evidence indicates that any natural or synthetic compound having the capacity for promoting growth or sustaining an existing metabolic state in the endometrium is also capable of inducing withdrawal bleeding. However, this does not imply that all compounds capable of inducing menstruation do so by the same biochemical action; in fact, there is considerable evidence that this is not so (see chapter by Villee). Yet in each instance a series of events is set in motion that leads up to active bleeding.

X. The Mechanism of Menstruation

The immediate cause and mechanism of menstruation has continued to be a topic of special interest for many years and the subject of frequent general discussions. A generalization in keeping with our present knowledge is that no gross morphologic feature of the endometrium is distinctive of menstruation. A menstruating endometrium may be representative of any stage of the follicular or luteal phase of the cycle. The most frequently discussed hypothesis regarding the mechanism of menstruation is that proposed by Markee (1940, 1946) which is based on direct observations of vascular changes in endometrial grafts in the anterior chamber of the eye of monkeys (see p. 564). The changes observed in the endometrium shortly before bleeding are, briefly, as follows. (1) There is extensive and rapid regression of the endometrium due to loss of ground substance from the stroma (Fig. 9.23). (2) The rapid regression brings about a disproportion between the length of the coiled arteries and thickness of the endometrium with the formation of additional coils. (3) The increased coiling of the arteries retards the circulation of blood through them and their branches. This stasis begins 1 to 3 davs before the onset of the flow, and is associated with leukocytosis in the endometrium. (4) The portion of the coiled arteries located adjacent to the muscularis constricts 4 to 24 hours before the onset of the flow. This vasoconstriction persists throughout the menstrual period except when individual coiled arteries relax and blood circulates through them for a few minutes. Markee postulated that the immediate cause of menstruation under these conditions was the injurious effect of anoxemia upon the tissues of the endometrium l)rought about by mechanical compression and constriction of the coiled arteries. Therefore, the coiled arteries and their modifications become the central feature upon which the theory is based.

Fig. 9.23. A diagram indicating correlated changes in ovary and endometrium during an ovulatory cycle of rhesus monkey. Thickness of endometrium, density of stroma, gland form, and three types of arteries are indicated. There is a gradual rise in thickness up to the time of ovulation, and a brief decline followed by development of the luteal or progestational phase with accumulation of secretion in the glands due to relaxation of the myometrium. This is followed by loss of ground substance from the stroma, which is the primary factor in the premenstrual regression of the ischemic phase. This is a prelude to extravasation and shedding of tissue. Incidentally, secretion is extruded and glands collapse. There is further regression throughout the phase of menstruation. More than the basal zone (coarse stipple) survives menstruation. During repair, thickening of the endometrium is associated with increase in ground substance in the stroma and growth in the glands. (From G. W. Bartelmez, 1957, Am. J. Obst. & Gynec, 74, 931-955, 1957, with some modification of description.)

Although this offers an explanation for many of the facts, it falls short in that now it is known that menstruation can occur in the absence of coiled arteries. Kaiser (1947) showed that no spiral arteries are present in the endometrium of three species of South American monkeys known to menstruate. He also found that the coiled vessels of the endometrium could be destroyed almost completely by giving large doses of estrogen and yet bleeding followed estrogen withdrawal.

Several experimental conditions under which the coiled vessels of the endometrium are destroyed have been mentioned in the present discussion and in each instance bleeding invariably followed withdrawal of the supporting stimulus. The extremely atrophic endometrium present at the conclusion of a prolonged treatment with progesterone (Fig. 9.8) will bleed when the injections are stopped, and if estrogen injections are started immediately thereafter the endometrium that develops is normal with the exception of the absence of coiled arteries; even so, it also will bleed when the treatment is stopped. Even a more drastic destruction of endometrial structures occurs when both estrogen and progesterone are given for several months. Not only are the coiled arteries destroyed but also the glands and the luminal epithelium. All that remains is a modified stroma penetrated by a few small blood and lymph vessels and scattered glandular rudiments along the myometrium (Fig. 9.13). Yet, in spite of this, bleeding follows discontinuance of the treatment.

These observations prove conclusively that the spiral arteries of the endometrium do not hold the solution to the menstrual process. However, the descriptive account by Markee of the events that take place in the endometrium during the cycle remains one of the major contributions to our knowledge of the primate endometrium. Phelps (1946) also made a very careful study of the vascular changes in intraocular endometrial transplants in ovariectomized monkeys receiving estrogen and progesterone, and concluded that the primary function of the coiled arteries is concerned with vascularization of the implantation site of a developing embryo.

There also is reason for doul^ting that ischemia is a determining factor in the menstrual process. That constriction of the endometrial vessels does occur is well established, but that tissue destruction and bleeding are consequences of prolonged anoxemia may be questioned. The endometrium around the internal cervical os as seen in incised exteriorized uteri (Fig. 9.7) contains very few coiled arteries and does not take part in the periodic blushing and blanching of the fundus, but instead remains blood-red even during menstruation. Also, certain tongues of endometrium in a uterine fistula may become crowded by their neighbors to an extent of being partly or completely deprived of blood, yet they do not bleed even though their unfavorable situation leads to deterioration within a few days.

Emmel, Worthington and Allen (1941) attempted to induce menstruation in monkeys by operative ischemia. Circulation to the fundus of the uterus was interrupted by means of a tourniquet for periods of 1 to 8V4 hours, and in two instances for 19 hours. This procedure did not precijiitate uterine bleeding nor did it hasten the onset of an expected bleeding following estrogen withdrawal. In fact, when the uterus was deprived of blood for periods longer than 3 hours impairment of the bleeding response to estrogen withdrawal was observed, and 19 hours of ischemia caused atrophy of the uterus without bleeding.

It also has been reported that a toxic substance formed in the endometrium is responsible for menstruation. This menstrual toxin is supposed to be present in the endometrium just previous to and during menstruation, and to be a substance resembling or identical with necrosin, a material found in pleural exudate following an inflammatory reaction (Smith and Smith, 1951). Zondek (1953) reports that menstrual blood, when obtained under relatively sterile conditions, is no more toxic to experimental animals than sterile tissue extracts. He also found that death of animals given injections of menstrual blood was due to bacteremia, an effect that could be prevented by giving antibiotics. Nor was he able to demonstrate a toxic substance in the premenstrual or menstrual endometrium. It might be mentioned in this connection that endometrial tissue destroyed by experimental ischemia in the experiments by Emmel, Worthington and Allen (1941), obviously did not influence menstruation nor did involuting endometrial tissue in uterine fistulae (p. 564). Therefore, the presence of a specific toxin that may induce menstruation has not been conclusively demonstrated.

Regardless of the specific cause of menstruation, the evidence shows that it can occur in the absence of coiled arteries, endometrial glands, or surface mucosa, and is unrelated to the thickness of the endometrium. This statement is based on conditions that have been experimentally induced in the monkey and they strongly indicate that menstruation, whatever the cause, is a stromal phenomenon. This view seems to be in agreement with the observations reported by Bartelmez in his elegant studies of the morphology of the endometrium of both monkeys and the human being. He emphasizes changes taking place in the connective tissue elements of the stroma and points out that much less tissue is lost at menstruation than i.< commonly thought (Bartehnez, 1957). The reduction in thickness is clue primariiy to loss of ground substance from the stroma, and conversely, the outstanding feature of repair is the increase in stromal ground substance (Fig. 9.23). ^Mitoses are rarely seen in the stroma during repair and arc not abundant enough in any phase according to Bartelmez to account for the observed increase in thickness of the endometrium. Our present knowledge indicates that an explanation of menstruation may be found in the metabolic effects induced in the stromal connective tissue of the endometrium by a sudden withdrawal of a supporting hormonal stimulus.

XI. References

Allen, E. 1927. The menstrual CA'cle in the monkey, Macacus rhesus: observations on normal animals, the effects of removal of the ovaries and the effects of injections of ovarian and placental extracts into the .spayed animals. Contr. Embrvol., Carnegie Inst. Washington, 19, 1-44.

Allen. E. 1928. Further experiments with an ovarian hormone in the ovariectomized adult monkey, Macacus rhesus, especially the degenerative phase of the experimental menstrual cycle. Am. J. Anat., 42, 467^87.

Allen, E., Diddle, A. W., Burford, T. H., .and Elder, J. H. 1936. Analyses of urine of the chimpanzee for estrogenic content during various stages of the menstrual cycle. Endocrinology, 20, 546-549.

Allen, E., Pr.\tt, J. P., Xewell, Q. U., .and Bl.and, L. J. 1930. Human tubal ova ; related early corpora lutea and uterine tubes. Contr. Embryol., Carnegie Inst. Washington, 22, 45-76.

B.ACHMAN, C, CoLLip, J. B., .\ND Selye, H. 1935. The effects of prolonged estriol administration upon the sex skin of Macaca mulatta. Proc. Roy. Soc, London, .■^er. B., 117, 16-21.

B.artel.mez, G. W. 1933. Histologic studies on the menstruating mucous membrane of the human uterus. Contr. Embryol.. Carnegie Inst. Washington, 24, 141-186.

B.artel.mez, G. W. 1937. ^Menstruation. Plnsiol. Rev.. 17, 28-72.

B.ARTELMEZ, G. W. 1951. Cyclic changes in the endometrium of the rhesus monkey {Macacus mulaltn). Contr. Embryol., Carnegie Inst. Washington. 34, 101-144.

B.ARTELMEz, G. W. 1957. The pha.scs of the menstrual cycle and their interpretation in terms of the pregnancv cycle. Am. J. Obst. & Gvnec, 74, 931-955.

BRY.AN.S, F. E. 1951. Progesterone of the lilood in the menstrual cycle of the monkey. Endocrinology, 48, 733-740.

BrxTo.N, C. L. 1940. Pregnanediol determinations as an aid in clinical diagnosis. Am. J. Obst. & Gynec, 40, 202-211.

CoLLiNGS, M. R. 1926. A study of the cutaneous reddening and swelling about the genitalia of the monkev, Macacus rhesus. Anat. Rec, 33, 271-287.

Corner, G. W. 1923. Ovulation and menstruation in Macacus rhesus. Contr. Embrj^oL, Carnegie Inst. Washington, 15, 73-101.

Corner, G. W. 1935. Influence of the ovarian hormones, estrin and progestin, upon the menstrual cvcle of the monkev. Am. J. Phj'siol., 113,238-250.

Corner, G. W. 1937. Experimental menstruation. Science, 85, 437-438.

Corner, G. W. 1942. The fate of the corpora lutea and the nature of the corpora aberrantia in the rhesus monkey. Contr. Embryol., Carnegie Inst. Washington, 30, 87-96.

Corner, G. W. 1945. Development, organization and breakdown of the corpus luteum in the rhesus monkey. Contr. Embrj'ol., Carnegie Inst. Washington, 31, 119-146.

Corner, G. W. 1951. Our knowledge of the menstrual cycle, 1910-1950. Lancet, 1, 919-923.

Corner, G. W., and Allen, W. M. 1936. Inhibition of menstruation by crj-stalline progesterone. Proc. Soc. Exper. Biol. & Med.. 34, 723724.

Crossex, R. J. 1953. Diseases of Women. St. Louis: C. V. Mosby Company.

Davls, M. E., and H.art^lan, C. G" 1935. Changes in vaginal epithelium during pregnancy in relation to the vaginal cvcle. J. A. M. A., 104, 279-285.

Diczfall'.sy, E. 1953. Chorionic gonadotrophin and estrogens in the human placenta. Acta endocrinol., Suppl. 12, 87-167.

DoRFMAN, R. I. 1956. Metabolism of androgens, estrogens and corticoids. Am. J. Med., 21, 679687.

DoRFMAN, R. I., AND VAN W.AGENEN, G. 1941. The

sex hormone excretion of adult female and pregnant monkevs. Surg. Gvnec. & Obst., 73, 545-548.

Duncan, P. A., Allen. E., and Ha.milton. J. B. 1941. The action of testosterone proprionate on experimental men.struation in the monkev. Endocrinology. 28, 107-111.

Eckstein, P. 1950. The induction of progesterone withdrawal bleeding in spayed monkeys. J. Endocrinol., 6, 405-411.

EcK.STEiN, P., AND ZucKERMAN, S. 1956. In Marshall's Physiology of Reproducliori. A. S. Parkes, Ed. Vol. 1, ]\ 334. London: Longmans Green & Company.

EmMEL, V. M., WORTHINGTON, R. V., AND AlLEN. E.

1941. Attempts to induce menstruation by operative ischemia in monkey's. Endocrinologv, 29, 330-335.

Engle. E. T. 1937. Problems of experimental menstruation. Cold Spring Harbor Svmposia Quant. Biol., 5, 111-114.

Engle, E. T., .and S.mith, P. E. 1935. Some uterine effects obtained in female monkeys during continued estrin administration, with especial reference to tlio r('i\ ix uteri. Auat. Rec, 6, 471-483.

Encle, E. T., Smith, P. E., and Shelesnyak, M. C. 1935. The role of estrin and progestin in experimental menstruation. Am. J. Obst. & Gynec. 29, 787-797.

Engle, E. T., and Smith, P. E. 1938. The endometrium of the monkey and est rone-progesterone balance. Am. J. Anat., 63, 349-365.

Engle, E. T., and Smith, P. E. 1939. Certain actions of testosterone on the endometrium of the monkey and on uterine bleeding. Endocrinology, 25^ 1-6.

Fish, W. R., Young, W. C, and Dorfman, R. I. 1941. Excretion of estrogenic and androgenic substances by female and male chimpanzees with known mating behavior records. Endocrinology, 28, 585-592.

Fluhmann, C. F. 1954. Comparative studies of squamous metaplasia of the cer\ix uteri and endometrium. Am. J. Obst. ct Gynec, 68, 1447-1462.

Forbes, T. R., Hooker, C. W., and Pfeiffer, C. A. 1950. Plasma progesterone levels and the menstrual cycle of the monkey. Proc. Soc. Exper. Biol. & Med., 73, 177-179.

GiLLMAN, J. 1937a. The cyclical changes in the vaginal smear in the baboon and its relationship to the perineal swelling. South African J. M. Sc, 2, 44-56.

GiLLMAN, J. 1937b. Experimental studies on the menstrual cycle of the baboon (Papio porcarhis). South African J. M. Sc, 2, 156-166.

GiLLMAN, J. 1938. Experimental studies on the menstrual cycle of the baboon (Papio porcarius). South African J. M. Sc, 3, 6&-71.

GiLLMAN, J. 1940a. Experimental studies on the menstrual cycle of the baboon (Papio porcarius). VI. The effect of progesterone upon the first part of the cycle in normal female baboons. Endocrinology, 26, 80-87.

GiLLMAN, J. 1940b. The effect of multiple injections of progesterone on the turgescent perineum of the baboon (Papio porcarius). Endocrinology, 26, 1072-1077.

GiLLMAN, J., AND GILBERT, C. 1946. The reproductive cycle of the chacma baboon (Papio itrsiDiis) with special reference to the problems of menstrual iriegularities as assessed by the behaviour of the sex skin. South African J. M Sc, Biol. Suppl., 11, 1-54.

GiLLMAN, J., .AND Smyth, G. S. 1939. The hormonal content of the human luteal follicle of pregnancy as determined by its effect on the perineum of the baboon. South African J. M. Sc, 4, 3&-45.

Haman, J. O. 1942. The length of the menstrual cvcle. A study of 150 normal women. Am. J. Obst. & Gynec, 43, 870-873.

Hamblen, E. C., Cuyler, W. K., Powell, N. B., Ashley, C, and B.aptist, M. 1939. Some clinical observations upon the metabolism and utilization of crystalline progesterone. Endocrinology, 25, 13-16.

Hamilton, C. E. 1949. Observations on the cervical mucosa of the Rhesus monkey. Contr. EnibryoL, Carnegie Inst. Washington, 33, 81-101.

H.artman, C. G. 1929. Three types of uterine bleeding in the monkey and the homology of menstruation (Abstr.). Anat. Rec, 42, 19.

Hartman, C. G. 1932. Studies in the reproduction of the monkey, Macacus (Pithecus) rhe.S-//.S, with special reference to menstruation and piegnancy. Contr. Embryol., Carnegie Inst. Washington, 23, 1-16.

H.ARTMAN, C. G. 1934. Some attempts to influence the menstrual cvcle in the monkev. Am. J. Obst. & Gynec, 27, 564-570.

Hartman, C. G. 1937. Menstruation inhibiting action of testosterone. Proc Soc. Exper. Biol. & Med., 37, 87-89.

H.ARTMAN, C. G. 1944. Regeneration of the monkey uterus after surgical removal of the endometrium and accidental endometriosis. Western J. Surg. Obst. & Gynec, 52, 87-102.

Hartman, C. G., .and Speert, H. 1941. Action of progesterone on the genital organs of the unjirimed Rhesus monkev. Endocrinology, 29, 639-648.

H.ARTMAN, C. G., GeSCHICKTER, G. F., AND SpEERT, H.

1941. Effects of continuous estrogen administration in verv large doses. Anat. Rec, Suppl. 2, 79, 31.

Hellman, L. M., Rosenthal, A. H., Kistner, R. W., AND Gordon, R. 1954. Some factors influencing the proliferation of the leserve cells in the human cervix. Am. J. Obst. & Gvnec. 67, 899-915.

Hertig, a. T., and Rock, J. 1944. On the development of the early human ovum, with special reference to the trophoblast of the previllous stage ; a description of 7 normal and 5 pathologic human ova. Am. J. Obst. & Gvnec, 47, 149-184.

Heuser, C. H., and Streeter, G. L. 1941. Development of the macaque embryo. Contr. Embryol., Carnegie Inst. Washington, 29, 17-55.

Hisaw, F. L. 1935. The physiology of menstruation in Macacus rhesus monkevs. Am. J. Obst. & Gynec, 29, 638-659.

HiSAW% F. L. 1942. The interaction of the ovarian hormones in experimental menstruation. Endocrinology, 30, 301-308.

HiSAW, F. L. 1943. Androgens and experimental menstruation in the monkey (Macaca viulatta). Endocrinology, 33, 39-47.

HisAW, F. L. 1950. Factors influencing endometrial growth in monkeys (Macaca mulatta). In A Symposium on Steroid Hormones, E. S. Gordon, Ed., pp. 259-276. Madison: University of Wisconsin Press.

Hisaw, F. L. 1959. Endocrine adaptations of the mammalian estrous cycle and gestation. In Columbia University Symposium on Comparative Endocrinology, pp. 533-552.

Hisaw, F. L., Creep, R. O., and Fevold, H. L. 1937. Effects of progesterone on the female genital tract after castration atrophy. Proc Soc. Exper. Biol. & Med., 36, 840-842."

His.\w, F. L.. AND Creep, R. O. 1938. The inhibition of uterine bleeding with estradiol and progesterone and associated endometrial modifications. Endocrinology, 23, 1-14.

His.\w, F. L., .^ND His.wv, F. L., Jr. 1958. Spontaneous carcinoma of the cervix uteii in a monkey (Macaca mulatto). Cancer, 11, 810816.

HiSAW, F. L., .4ND Lendrum, F. C. 1936. Squamous metaplasia in the cervical glands of the monkey following oestrin administration. Endocrinology, 20, 228-229.

HiSAW, F. L., Meyer, R. K., axd Fevold, H. L. 1930. Production of a premenstrual endometrium in castrated monkeys by ovarian hormones. Proc. Soc. Exper. Biol. & Med., 27, 400-403.

HiTSCHMANN, F., AND Adler, L. 1907. Die Lehre von der Endometritus. Ztschr. Geburtsh. u. GynJik., 60, 63-86.

Kaiser, I. H. 1947. Absence of coiled arterioles in the endometrium of menstruating New World monkeys. Anat. Rec, 99, 353-363.

Kaufmann, C, Westphal, U., and Zander, J. 1951. Untersuchungen liber die biologische Bedeutang der Ausscheidungsprodukte des Gelbkcirperhormons. Arch. Gynak., 179, 247-299.

Knaus, H. 1950. Die Physiologie der Zeugung des Menschen. Wien: Wilhelm Maudrich.

Krohn, p. L. 1951. The induction of menstrual bleeding in amenorrhoeic and normal monkeys by progesterone. J. Endocrinol., 7, 310-317.

Krohn, P. L. 1955. The induction of cyclic uterine bleeding in normal and spayed rhesus monkeys by progesterone. J. Endocrinol., 12, 6985.

Krohn, P. L., and Zuckerman, S. 1937. Water metabolism in relation to the menstrual cycle. J. Physiol., 88, 369-387.

Latz, L. J., and Reiner, E. 1942. Further studies on the sterile and fertile periods in women. Am. J. Obst. & Gynec, 43, 74-79.

Lendrum, F. C., .and Hisavv^ F. L. 1936. Cytology of the monkey endometrium under influence of follicidar and corpus luteum hormones. Proc. Soc. Exper. Biol. & Med., 34, 394-396.

Lopez Columbo de Allende, I., and Orias, O. 1950. Cytology of the Human Vagina. New York: Paul B. Hoeber, Inc.

Lopez Colu.mbo de Allende, I., Shorr, E., and Hartman, C. G. 1945. A comparative study of the vaginal smear cycle of the rhesus monkey and the human. Contr. Embryol., Carnegie Inst. Washington, 31, 1-26.

Markee, J. E. 1940. Menstruation in intraocular endometi'ial tr;msplants in the I'hesus monkey. Contr. Embrvol., Carnegie Inst. Washington, 28, 219-308.

Markee, J. E. 1946. Morphologic and endocrine basis foi' menstrual bleeding. In Progress in Gynecology, Meigs and Sturgis, Eds. Vol. II, pp. 37-47. New York: (hune and Stiattdii.

Markee, J. E., and Berg, B. 1944. Cyclic fluctuations in blood estrogen as a possible cause of menstruation. Stanford Med. Bull., 2, 55-60.

Markee, J. E., D.wis, J. H., and Hinsf.y, J. C. 1936. Uterine bleeding in spinal iii()nk(>vs. Anat. Rec, 64, 231-245.

Mazer, C, and Israel, S. L. 1951. Diagnosis and Treatment of Menstrual Disorders and Sterility. New York: Paul B. Hoeber, Inc.

Meyer, R. 1911. Uber Corpus luteum-Bildung beim Menschen. Arch. Gynak., 93, 354-404.

Nissen, H. W., and Yerkes, R. M. 1943. Reproduction in the chimpanzee: report on 49 births. Anat. Rec, 86, 567-578.

Novak, E., and Te Linde, R. W. 1924. Endometrium of menstruating uterus. J. A. M. A., 83, 900-906.

OvERHOLSER, M. D., AND Allen, E. 1933. Ovaiiau hormone and traumatic stimulation of monkey's cervix to a condition resembling early cancer. Proc Soc. Exper. Biol. & Med., 30, 1322-1326.

OvERHOLSER, M. D., AND Allen, E. 1935. Atypical growth induced in cervical epithelium of monkey by prolonged injections of ovarian hormone combined with chronic trauma. Surg. Gynec. & Obst., 60, 129-136.

OvERHOLSER, M.D.,. AND Nelson, W. 0. 1936. Migration of nuclei in uterine epithelium. A monkey following prolonged estrin injections. Proc Soc Exper. Biol. & Med., 34, 839-841.

P.APANicoL.Aou, G. N., Traut, H. F., and M.archetti, A. A. 1948. The Epithelia of Woman/s Reproductive Organs. New York: Commonwealth Fund.

Parkes, a. S., and Zuckerm.an, S. 1931. The menstrual cycle of the Primates. II. Some effects of oestrin on baboons and macaques. J. Anat., 65, 272-276.

Phelps, D. 1946. Endometrial vascular reactions and the mechanism of nidation. Am. J. Anat., 79, 167-197.

Phelps, D. H. 1947. The factor of previous treatment in experimental menstruation. J. Clin. Endocrinol., 7, 611-623.

Rakoff, a. E. 1946. Studies on high dosage progesterone therapy of amenorrhea. Am. J. Obst. & Gynec, 51, 480-491.

Rock, J., and Hertig, A. 1942. Some aspects of early human development. Am. J. Obst. & Gynec, 44, 973-983.

RossM.AN, I. 1940. The decidual reaction in the rhesus monkey {Macaca mulatta). I. The epithelial proliferation. Am. J. Anat., 66, 277365.

Schroder, R. 1914. Uber das Verhalten der Uterusschleimhaut um die Zeit der Menstruation. Monatsschr. Geburtsh. u. Gynak., 39, 3-21.

Seegar, E. G. 1940. The histologic effect of progesterone on hyperplastic endometria. Am. J. Obst. & Gynec, 39, 469-476.

Smith, O. W., and S.mith, G. V. 1951. Endocrinology and related phenomena of the human menstrual cvcle. Recent Progr. Hormone Res., 7, 209-253.

SoMMKKVILLK, I. V ., AND MaRRIAN, G. F. 1950. Urinary excretion of prcgnanediol in human subjects following the administration of progesterone and of pregnane-3a:20a-diol. I3iochem. J., 46, 285-289.

Stieve, H. 1926. Di(> regelmassigen Verliinderungen der Muskulatur und des Bindegewebs in der meuschlichen Gebarmutter in ihier Abhangigkeit von der Follikelreife und der Aiisbildung eines gelben Korpers, nebst Beschreibung eines menschlichen Eies im Zustand der ersten Reifteilung. Ztschr. mikroskop.-anat. Forsch, 6, 351-397.

Stieve, H. 1942. Der Einfluss von Angst und psychischer Erregung auf Bau und Funktion der weiblit'hen Geschlechtsorgane. Zeutralbl. Gynak., 66, 1698-1708.

Stieve, H. 1943. Weitere Tatsachen zur Kliirung der Frage: Wann wird das Ei aus dem Eierstock ausgestossen? Zentralbl. Gvnak., 67, 5877.

Stieve, H. 1944. Paracyclische Ovulationen. Zentralbl. Gynak., 68, 257-272.

Sturgis, S. H. 1942. Method for obtaining uterine fluid from the monkey : effect of pilocarpine, atropine, physiologic salt solution and adrenalin. Endocrinology, 31, 664-672.

VAN Dyke, H. B., and Ch'en, G. 1936. Observations on biochemistry of genital tract of female macacjue particularly during menstrual cycle. Am. J. Anat., 58, 473-499.

VAN Wagenen, G. 1945. Mating in relation to pregnancy in the monkey. Yale J. Biol. & Med., 17, 745-760.

VAN Wagenen, G. 1947. Early mating and pregnancy in the monkev. Endocrinologv, 40, 3743.

VAN W.^genen, G., and Morse, A. H. 1940. Cyclic changes in the exteriorized uterus. Endocrinology, 27, 268-273.

Venning, E. H., and Browne, J. S. L. 1937. Studies on corpus luteum function. I. The vn-inary excretion of sodium pregnandiol glucuronidate in the human men.^trual cvcle. Endocrinologv, 21,711-721.

Werner, A. A., and Collier, W. D. 1933. The effect of theelin injections on the castrated woman. J. A. M. A., 100, 633-640.

Westman, a. 1932. Studien iiber den Sexualzyklus bei Makakus-Rhesus-Affen, nebst einigen Bemerkungen liber den menstruellen Blutungs-mechanismus. Acta obst. et gynec. scandinav., 12, 282-328.

WiSLOCKi, G. B., and Streeter. G. L. 1938. On the placentation of the macaque {Macaca mulatta), from the time of implantation until the formation of the definitive placenta. Contr. Embryol., Carnegie Inst. Washington, 27, 166.

Young, W. C, and Yerkes, R. M. 1943. Factors influencing the reproductive cycle in the chimpanzee; the period of adolescent sterility and related problems. Endocrinology, 33, 121-154.

Z.ARRow, M. X., Hisaw, F. L., and Bryans, F. 1950. Conversion of desoxycosterone acetate to progesterone in vivo. Endocrinology, 46, 403-404.

Zarrow, M. X., Shoger, R. L., and Lazo-Wasem, E. A. 1954. The rate of disappearance of exogenous progesterone from the blood. J. Clin. Endocrinol., 14, 645-652.

Zondek, B. 1953. Does menstrual blood contain a specific toxin? Am. J. Obst. & Gvnec, 65, 1065-1068.

-Zondek, B. 1954. On the mechanism of uterine bleeding. Am. J. Obst. & Gynec, 68, 310-314.

Zondek. B., and Rozin, S. 1938. Production of uterine haemorrhage in the normal cycle and hi amenorrhoea through progesterone. J. Obst. & Gynaec. Brit. Emp., 45, 918-931.

Zuckerman, S. 1930. The menstrual cycle of the Primates. I. General nature and homology. Proc. Zool. Soc, London, 1930, 691-754.

Zuckerman, S. 1935. The menstrual cycles in the Primates. VIII. The estrin-vvithdrawal theory of menstruation. IX. The effect of estrin on the denervated sexual skin. Proc. Rov. Soc, London, ser. B., 118, 13-33.

Zuckerman, S. 1936a. Inhibition and induction of uterine bleeding bv means of estrone. Lancet, 2, 9-13.

Zuckerman, S. 1936b. The interrelation of estrone and progestin in the menstrual cvcle. J. PhvsioL, 86, 31-33.

Zuckerman, S. 1937a. The duration and phases of the menstrual cycle in Primates. Proc. Zool. Soc, London, ser. A., 1937, 315-329.

Zuckerman, S. 1937b. The menstrual cycle of the Primates. X. The oestrone threshold of the uterus of the rhesus monkey. XL The part played by oestrogenic hormone in the menstrual cycle. Proc. Roy. Soc, London, ser. B., 123,441-471.

ZucKER.MAN, S. 1937c Effects of prolonged oestrin-stimulation on the cervix uteri. Lancet, 1, 435-437.

Zuckerman, S. 1937d. Further observations on endocrine interaction in the menstrual cvcle. J. Physiol., 89, 49-51.

Zuckerman, S. 1937e. The duration and phases of the menstrual cycle in Primates. Proc. Zool. Soc London, ser. A., 1937, 315-329.

Zuckerman, S. 1939. The effect of sex hormones, cortin, and vasopressin on water-retention in the reproductive organs of monkeys. J. Endocrinol., 1, 147-155.

Zuckerman, S. 1941. Periodic uterine bleeding in spayed rhesus monkeys injected daily with constant threshold dose of oestrone. J. Endocrinol., 2, 263-267.

Zuckerman, S. 1949. The menstrual cvcle. Lancet, 2, 176.

Zuckerman, S. 1951. The hormonal basis of uterine bleeding. Acta endocrinol., 7, 378-388.

Zuckerman, S., and P.arkes, A. S. 1932. The menstrual cycle of the primates. V. The cycle of the baboon. Proc Zool. Soc. London, 1932, 139-191.

Zuckerman, S., van W.agenen, G., and Gardiner, R. H. 1938. The sexual skin of the rhesus monkey. Proc Zool. Soc, London, ser. A., 108, 385-401.