1
S. A. A S D E L L
Cornell University, Ithaca, New York
TABLE OF CONTENTS
Introduction 2 I. Evolutionary Changes in the Organs of Reproduction 2
A. The Female 2 B. The Male 3 II. Anatomy of the Organs of Reproduction 3
A. The Female 3 B. The Male 8 III. Physiology of the Organs of Reproduction 10
A. Hormones Involved 10 B. The Breeding Season 12
C. Ovulation 14 D. The Reproductive Pattern 16
E. The Union of Spermatozoon and Ovum 17
IV. Implantation and Placentation 19 A. Nourishment of the Blastocyst 19
B. Implantation 20 C. Placentation 21 D. Placental Hormones 23 V. Development, Gestation, and Birth 26
A. Development 26 B. The Duration of Gestation 27
C. The Cause of Birth 28
D . Relaxin 29 VI. Some Factors That Influence the Number of Young Produced. _ 29
A. Twinning 29 B. Egg Wastage and Intra-uterine Deaths 29
C. Postpartum Heats 30
VII. Lactation 31 A. Anatomy of the Mammary Glands 31
B. Endocrinology of Mammary Development 32
C. Milk 32 VIII. Secondary Sexual Characters 33
IX. Information That May Be Obtained from Field Studies 34
A. Data To Be Gathered 34 B. Whales, a Special Instance 35
References 36 1
Reproduction and Development
2 S. A. Asdell Introduction
I n this chapter the main emphasis is upon those aspects of reproduction and embryology t h a t are purely mammalian. More general aspects, if t h e y are mentioned at all, are treated in a more cursory way, and usually only as they are needed to explain the features peculiar to mammals. As is usual with new and successful developments, the mammalian reproductive organs display a wide degree of variability, both in their a n a t o m y and their physiology. T h e evolutionary trends, among t h e m some of considerable interest to the physiologist and endocrinologist, are described in con- siderable detail.
The principal distinction between Mammalia and other classes of the Chordata is the development of m a m m a r y glands for the nourishment of the young for a period immediately after their birth. Simple sweat glands at first, they rapidly evolved into specialized organs, and, since their develop- ment is synchronized with, and dependent upon, reproductive activity they are dealt with in this chapter. The platypus (Ornithorhynchus) and echidnas (Tachyglossus and Zaglossus) nourish their young for a time with milk, but they are oviparous so t h a t internal gestation is not the universal criterion for separating the Mammalia from other forms. The marsupials have developed internal gestation, but t h e young are born while t h e y are still in the embryonic condition. Development continues in t h e marsupium, or pouch, where the young are fed with milk. Among t h e Eutheria internal development proceeds much further and t h e reliance upon milk for early nourishment lessens. This need has almost disappeared in some species, e.g. the guinea pig.
I. Evolutionary Changes in the Organs of Reproduction A. T h e Female
Along with the evolution of internal gestation there has come a series of changes in the urogenital organs, especially in those of the female. T h e pair of simple oviducts found in the platypus is differentiated in marsupials into oviduct, uterus, and vagina. A t the same time the connection with the cloaca ceases and the development of a separate urogenital sinus occurs.
The uterine portion of the tract is specialized so t h a t placental nourishment of the embryo becomes an efficient mechanism. Along with this change t h e cells remaining within the ovisac, or graafian follicle, after the ovum has been discharged are retained for some time and take on the function of a temporary endocrine organ responsible for spacing ovulations, for effective implantation of the fertilized ovum and for m a m m a r y development. I n
Ζ. Reproduction and Development 3 some more advanced forms the placenta, too, has become endowed with endocrine functions so t h a t the fetus and its adnexa are beginning t o control some of t h e aspects of their own reproduction previously under control of the maternal organism. T h e placenta in man, for instance, secretes not only steroid hormones, but also a gonadotropic hormone somewhat similar to one t h a t is produced by the anterior pituitary of the mother.
Among the Eutheria generally, and also in some marsupials, the two mullerian ducts display a tendency to fuse in the midline. I n rabbits this has produced a common vagina with two lateral uteri and oviducts; t h e uteri enter the vagina by two cervices. I n the rat there is some fusion of the circular muscle surrounding the cervices. A further fusion produces the bicornuate uterus with a common body and separate horns each of which is connected with its corresponding oviduct. The size of the body relative to the horns varies, with a gradual reduction of the latter. Even
tually, in the higher primates the uterine horns have disappeared entirely and a single, more or less globular uterus remains, the two oviducts entering at the two anterolateral extremes. I n the lower primates the uterine horns are small.
B. T h e Male
I n the male, modifications in the genitalia are not as varied as they are in females. There has been a tendency for exocrine glands to develop as off
shoots of the wolffian ducts and of the urogenital sinus. B u t if one takes the condition in insectivores as the model for Mammalia, since it is generally regarded as the least specialized order and the one from which all other eutherians arose, one finds t h a t there has been a tendency for suppression of the accessory glands in the carnivores, seals, and whales and for their increased elaboration in rodents. T h e functions of these accessory glands are b u t poorly understood, and it has been shown that, even in species in which they are highly developed, their removal causes little impairment of fertility.
II. Anatomy of the Organs of Reproduction A. T h e Female
1. The Ovary
The ovaries are paired organs situated caudad to the kidneys. Their shape varies, but the ovaries are best described as roughly bean-shaped in most mammals, although they are greatly elongated in rabbits. The sur-
4 S. A. Asdell face in most mammals is irregular, especially during the breeding season when follicles and corpora lutea frequently protrude. This irregularity is most pronounced in species in which m a n y ova are shed simultaneously or in which the corpora lutea persist for a long time.
The right ovary of the monotremes tends to be rudimentary (more so in the platypus t h a n in the echidna), t h u s recalling the avian condition. I n marsupials and eutherians little tendency t o asymmetry is found, though in m a n y species ovulation tends to occur characteristically more frequently from one ovary t h a n the other. I n bats the tendency to asymmetry is greatest, and it is the left ovary, not the right, t h a t tends to be rudimentary.
T h e ovisac, or graafian follicle, is the most important structure in the m a t u r e ovary. I t consists, from without, of the theca externa (a fibro- blastic coating), the theca interna with a rich capillary blood supply and cells t h a t are believed to have an endocrine function, and the granulosa cell layer, which m a y aid in nourishing the ovum and which certainly gives rise to the endocrine cells of the corpus luteum. Surrounded by several layers of granulosa cells is the ovum. I n mature follicles, too, m a y be seen a quantity of a moderately viscous fluid t h a t contains proteins and estrogenic hor- mones. This fluid undoubtedly helps to release the ovum from the follicle when t h e latter ruptures during the heat period.
After the eggs have been released from the follicles these structures collapse and the granulosa cells come into closer contact with the capillary circulation. This follows because the theca interna becomes folded and the division between it and the granulosa layer vanishes. Strands of theca cells, together with blood vessels, invade the granulosa region. The cells of the latter now increase in size, but not in number, and they become laden with lipid droplets which, in some species, are intensely colored with a yellow pigment. In a period of about 4 days after ovulation t h e corpus luteum be- comes fully organized. I t secretes progesterone, a steroid hormone t h a t has an important function in regulating the activity of the anterior pituitary gland, t h u s preventing the maturation of new follicles during the life of the corpus luteum. In addition, this hormone is essential for the preparation of the endometrium so t h a t the developing embryo m a y become implanted.
The role of the theca cells in this transformation into a corpus luteum is obscure and variable. I n some species of bats and marsupials they m a y show little differentiation, but in others they m a y differentiate until they are scarcely distinguishable from the lutein cells of granulosa origin. This is a condition found in the bats Pipistrellus [Nyctalis] noctula and Plecotus auritus (Harrison, 1948), but it is not known whether these theca-lutein cells have an endocrine function. I n the New World monkeys, the howler (Alouatta) and t h e spider monkeys (A teles) t the walls of the corpora lutea soon disappear so t h a t these bodies become indistinguishable from t h e
1. Reproduction and Development 5 interstitial tissue of the ovary, which is glandular in these species (Dempsey, 1939). The elephant shrew (Elephantulus) and the Malagasy tenrecs of the genera Hemicentetes and Setifer [Ericulus] have an interesting method of forming their corpora lutea. T h e contents of the ruptured follicles are everted and spread over the surface of the ovary where they proliferate and acquire a coating derived from the germinal epithelium (Van der Horst and Gillman, 1942; Feremutsch and Straus, 1949).
The interstitial tissue of the ovary is most variable. I n some species, e.g., cattle, it consists almost entirely of fibroblasts and connective tissue elements. I n some species, large polyhedral cells with the appearance of glandular elements are present. These are well developed in the moles and in bats. I n the Old World water shrew Neomys fodiens, this glandular-type tissue hypertrophies greatly during pregnancy and its cells cannot be distinguished from those of the corpora lutea. The ovary becomes almost wholly luteinized (Price, 1953). This variability in the appearance and abundance of the interstitial tissue has not been explained, and evidence for an endocrine function is lacking. The interstitial "gland" cells of t h e pocket gopher (Geomys bursarius) are derived from the theca interna of t h e follicle. This layer hypertrophies greatly just before ovulation (Mossman, 1937a).
2. The Oviduct
T h e oviduct, or fallopian tube, is the convoluted tubule t h a t is derived from the upper portion of the mullerian duct. I t serves as a means by which spermatozoa m a y travel from the uterus toward the ovary and for t h e transport of the ovum into the uterus. At the ovarian end it is extended as a membrane t h a t encapsulates the ovary or as a funnel for the reception of t h e eggs as they are shed. This arrangement effectively prevents t h e m from becoming lost in the body cavity. T h e capsule form is common in t h e Chiroptera, Insectivora and Rodentia of the Family Muridae, b u t it appar- ently is never complete, as most reports of its presence mention the exis- tence of a minute opening to the body cavity.
The oviduct wall consists mainly of unstriated muscle arranged as longitudinal and circular fibers. T h e lumen is lined with ciliated cells t h a t beat toward the uterus and with albumen-secreting cells. The latter are especially a b u n d a n t in the rabbit. The cells of the monotreme oviduct secrete keratin, which is deposited on the surface of the egg as it passes through the organ. I n t h e ovarian (ampullar) part, the oviduct wall is folded into very complicated arborizations, but the structure becomes much simpler as the uterus is approached; the muscle layers are thicker and the number of cilia less toward t h e uterus.
6 S. A. Asdell Entrance to the uterus is gained through the tubo-uterine junction which varies much in structure from species to species. The opossum (Didelphis) oviduct is practically continuous with the uterus. I n the vampire bat Desmodus, the entrance is at right angles, with a thickening of the mus- culature at the junction. The rabbit tubo-uterine junction is surrounded by villi which project into the uterine lumen. The orifice of the chipmunk oviduct is protected by a flap of mucous tissue (Andersen, 1928).
3. The Uterus
Except for the difference in form already described, the structure of t h e uterus is fairly uniform throughout the Mammalia. The wall consists of the myometrium with fairly distinct layers of longitudinal, transverse, and circular unstriated muscle in various proportions and of the endometrium.
The latter is a connective tissue coat, richly vascular, with tubular glands discharging into the lumen. The epithelium consists of a single layer of columnar to cubical cells which are subject to hormonal control. A t their maximum development they are so crowded together t h a t their nuclei occupy different levels in adjoining cells. This gives a pseudostratified appearance to the epithelium. The junction (cervix) with the vagina is usually distinguishable by the formation of a neck with heavy musculature and scanty endometrium. Tubular glands are replaced by surface mucus- secreting cells. When the secretion from these cells is present in quantity, e.g., between heats and during pregnancy, it forms an effective barrier to the entrance of foreign bodies from the vagina. The cervix uteri is capable of great distension so t h a t the young m a y be born through it. I n the sow the transition from uterus to vagina is not nearly as well marked as it is in most mammals.
The ruminants have a number of endometrial processes in the form of rounded or oval bosses t h a t project into the lumen. These "caruncles" are devoid of glands and are specialized regions for the reception of the em- bryonic placental attachments. I n Elephantulus polyplike endometrial growths form during each ovulatory cycle and implantation of the o v u m occurs on these structures. They are destroyed at the conclusion of each cycle if pregnancy does not intervene (Van der Horst and Gillman, 1941).
Some workers see in this a forerunner of the phenomenon of menstruation which is characteristic of the Old World monkeys and m a n .
4- The Vagina
The wall of the vagina contains a fibrous layer interspersed with bundles of unstriated muscle. The epithelium varies according to species, position in the tract, and kind of hormonal stimulation. I n the region of the cervix
I. Reproduction and Development
and for a varying distance posteriorly it consists of a basement layer and one or more of cubical to flattened epithelial or glandular cells. I n rodents and carnivores epithelial cells predominate and they grow in numbers under the stimulus of estrogens so t h a t the epithelium at the time of heat becomes several cells thick, with progressive cornification toward t h e surface. Most of these layers are abruptly sloughed off when estrogen stimulation ceases, so t h a t the nature of the cells in the vaginal smear is a good indication of the stage of growth reached by the follicles in the ovary. I n most other mammals about which information is available, the cellular changes, although they occur to some degree, are not sufficiently clear cut for ac
curate diagnosis. The cells of the upper vagina in ruminants are mostly mucus-secreting, and they take little p a r t in these changes.
The posterior vagina, derived from the urogenital sinus, is coated with a typical squamous epithelium which shows little response to hormones. This region is well developed in ruminants.
Marsupials have two lateral vaginas, but birth takes place through a temporary central vagina t h a t develops at each parturition. This structure becomes much reduced or vanishes entirely during the intervening periods.
However, birth takes place through t h e lateral vaginas in Potorous tri- dactylus, a rat kangaroo (J. Pearson, 1944; Flynn, 1922).
δ. Accessory Glands
Accessory glands in the female are usually less in evidence t h a n they are in the male. I n m a n y bats the prostate is well developed (Matthews, 1941), and it is present also in the green monkey Ceropithecus aethiops (Zuckerman, 1938). Analogs of the bulbo-urethral, or "vestibular," glands are commonly found.
6. The Urogenital Sinus
T h e urogenital sinus displays comparatively little variation. T h e clitoris is perforated by the urethra in some species, e.g., in the Muridae; in others, particularly in bears and other Carnivora, an os clitoridis is present. I n the spotted hyena (Crocuta crocuta) the clitoris is peniform and is pene
trated by the urogenital canal so t h a t t h e vagina actually traverses it. This condition is especially apparent in the immature female. During the breed
ing season the terminal orifice lengthens so t h a t coitus and birth are not impeded. T h e female also possesses a "scrotal" pouch, and it is difficult t o distinguish t h e male from the female by external examination. This has given rise to the folk belief t h a t t h e hyena is hermaphroditic (Matthews, 1939).
8 S. A. Asdell Β. T h e Male
1. The Testis
The primary male sex organs, the testes, do not show much variability, not, at any rate, in the cytology of spermatogenesis. The spermatozoa are produced in the seminiferous tubules and travel through t h e rete testis and vasa efferentia into the epididymis. The blood vessels do not enter the seminiferous tubules; they traverse the spaces between them. Here, also, are found the "interstitial" cells of Leydig. These are the most variable elements of the testis and, since they produce the male hormone testoste
rone, they have attracted considerable attention. Their number varies considerably from species to species; in m a n y seasonal breeders they are subject to cyclical variations in numbers and in size. These changes are well marked in Plecotus [Corynorhinus], the lump-nosed bat. Development of these cells is at its height during the breeding season (Pearson et aLy 1952).
There is good evidence t h a t the process of spermatogenesis is very susceptible to changes in temperature. The function of the scrotum is believed, therefore, to be to maintain spermatogenesis by keeping the testes at a temperature a few degrees below t h a t of the abdomen. B u t t h e testes of elephants, armadillos, tenrecs, and several other m a m m a l s are permanently abdominal. I n m a n y others they are outside the abdomen, but inguinal, not scrotal. Among these are t h e mole, rhinoceros, and most rodents. I n a few, e.g., in bats and ground squirrels they are periodically scrotal. Finally, in most marsupials, ungulates, carnivores, and primates they are permanently scrotal (Eckstein and Zuckerman, 1956a). N o t all species are equally sensitive to temperature—if t h a t is the main explanation of this variability.
The interval between major cell divisions during spermatogenesis is about a week, and about 3 weeks are occupied in growing flagella (Asdell and Salisbury, 1941). Accordingly, the whole process of sperm formation takes about 6 weeks. After they have reached the epididymis, another 2 or 3 weeks are needed for the sperm to tranverse this long, ciliated tubule.
During this time t h e spermatozoa appear t o be undergoing a maturation process. Fertility of sperms drawn from the head of the epididymis is less t h a n t h a t of those obtained from the tail, near the point of junction with the vas deferens (Young and Simeone, 1930). The epididymis also serves as a storehouse for the spermatozoa, and during their stay in t h a t organ they are inert. At the time of ejaculation they pass u p the vas deferens, a tubule with a thick layer of circular, unstriated muscle, into t h e urethra where they meet secretions from the accessory glands and become motile. T h e immotility in the epididymis is believed to be due to the low oxygen supply
1. Reproduction and Development 9 in this organ. On this view the accessory glands activate the spermatozoa principally by adding oxygen to the semen.
2. The Accessory Glands
The accessory glands consist of (1) preputial glands, with their greatest development in rodents, particularly in the mouse; (2) bulbo-urethral or Cowper's glands, greatly developed in insectivores and squirrels, but absent in cetaceans and m a n y carnivores; (3) seminal vesicles, greatly developed in Lagidium, the mountain viscacha (Pearson, 1949), and in the pig, but absent in monotremes, marsupials, cetaceans, and carnivores; and (4) prostate, present in all mammals except monotremes. I n rodents the prostate has reached a high degree of development and m a y be divided into three or four lobes t h a t differ somewhat in histology and in the composition of their secretions. T h e restricted distribution of some of these accessory glands suggests t h a t they are not essential for fertility, a suggestion t h a t has been confirmed by their removal in some species with little or no reduction in fertility. Their functions, if any, are obscure because the spermatozoa are exposed to their secretions, in the bull for instance, for a very short time. I n rodents and bats t h e prostatic secretion m a y serve a useful purpose by forming the vaginal plug which imprisons the spermat- ozoa immediately after copulation. T h e plug gradually disintegrates and allows a steady stream of spermatozoa to traverse the female tract. This plug formation has not been closely investigated, but it seems to be due to an enzymatic conversion of a fibrinogen to fibrin, a process similar to t h a t involved in blood clotting. Another enzyme present in the secretion causes a gradual liquefaction of the clot (Mann, 1954).
8. The Penis
T h e intromittent organ, the penis, is very variable. Slijper (1938) has classified two main types: vascular and fibroelastic, with intermediate forms. T h e vascular type depends for erection upon hydrostatic pressure brought about by engorgement of sinusoid spaces with blood. Animals having this t y p e include t h e Perissodactyla, Primates, Carnivora, Chirop- tera, and Insectivora. These have moderately long or very prolonged coitus.
The fibroelastic type of penis is much less vascular and depends more upon muscular action for intromission. T h e duration of coitus is usually very short. This t y p e is characteristic of ruminants and cetaceans. Slijper describes t h e penis in rodents as belonging to an undifferentiated type with moderately long duration of coitus. A baculum, or os penis, is widely distributed; it is easier to point to t h e groups in which it is absent. These include t h e monotremes, edentates, whales, and ungulates, as well as most marsupials.
10 S, A. Asdell III. Physiology of the Organs of Reproduction
A. Hormones Involved
1. The Anterior Pituitary Gland
The activities of the reproductive tract, in males as well as in females, seem to be dominated by three hormones secreted by the anterior pituitary gland. These are the follicle-stimulating hormone (FSH), the luteinizing hormone (LH), and prolactin or luteotropin. I n the male, F S H is important for the process of spermatogenesis; it is reinforced by testosterone in maintaining this activity. I n the female, F S H is necessary for the growth of the ovarian follicles beyond the beginning a n t r u m stage. Luteinizing hormone, in the male, enables the interstitial cells of Leydig to secrete the male steroid hormone testosterone. I n the female it assists in bringing the follicles to final m a t u r i t y and contributes to their rupture and to t h e luteinization of the granulosa cells t h a t remain in the cavity after the egg has been extruded. I n the rat prolactin seems to be necessary to enable the lutein cells of the corpus luteum to secrete the steroid hormone, pro- gesterone. This is not so in t h e cow. I n this species L H is the hormone most concerned in lutein function. These pituitary hormones are proteins and their target organs are the gonads.
2. The Steroid Hormones
The gonadic hormones, produced as a result of anterior pituitary activity, are steroids. They maintain the accessory sex organs in functional condition.
The principal male hormone, testosterone, maintains the male accessories.
I n the castrate, the prostate, seminal vesicles, and bulbo-urethral glands atrophy and their secretion greatly diminshes in amount. Testosterone also activates the sex center of the brain so t h a t sexual desire m a y be aroused with an appropriate stimulus. This seems to be in the nature of a trigger reaction, depending little upon testosterone once it has been elicited.
Secondary sexual characters in mammals are partly influenced by testos- terone, partly by unidentified hormones of the anterior pituitary, and partly by genetic factors.
The group of female steroid hormones, the estrogens, are believed to be secreted by the theca interna cells, but the placenta is also a rich source of them. These hormones are responsible for the periodic nature of the sexual impulse in most mammals, for they bring about sexual receptivity by their action upon the brain. They cause the myometrium to exhibit spontaneous activity. I n the endometrium they play essentially a growth-inducing role, probably in the main b y increasing the extent of the capillary bed. Growth
1. Reproduction and Development 11 of vaginal epithelium in m a n y species, notably in rodents and carnivores, is induced by t h e action of estrogens.
T h e second ovarian steroid hormone is progesterone, secreted by t h e corpus luteum and, in some species, by the placenta. This hormone sensitizes the endometrium, preparing it to receive t h e developing embryo. I t is not certain to w h a t degree it is essential after implantation has occurred because, in m a n y species, removal of the corpora lutea after implantation does not cause death of t h e embryos or fetuses. B u t in most of these ani- mals the placenta secretes progesterone, probably in sufficient quantity to maintain the pregnancy. Whether this is so in all such instances is not yet clear. Progesterone also promotes t h e development of t h e endometrial glands. I n some of its activities it seems to inhibit estrogen functions.
I n marsupials the duration of uterine gestation is shorter t h a n t h a t of the estrous cycle. I n this group, as well as in insectivores, the corpus luteum seems to exert little influence upon t h e reproductive processes. Apparently the newly evolved endocrine organ has not yet become fully coordinated with t h e other organs. And, as has been indicated, soon it tends to be supplanted by the placenta.
8. The Hypothalamus
T h e anterior pituitary in its t u r n is controlled t o some extent by secre- tions reaching it from the hypothalamus. These travel by way of a special portal blood system through the pituitary stalk (Harris, 1955), b u t the nature of these secretions is still a n open question. T h e y appear to be released by the hypothalamus as a result of a variety of influences, some hormonal and some of nervous origin. Sawyer et al. (1949) have advanced evidence suggesting t h a t both t h e sympathetic and the parasympathetic systems are involved in this chain of events, which is especially concerned in the release of L H by t h e anterior pituitary.
Group and species pecularities in reproductive patterns m a y eventually be explained in terms of t h e balance and timing of anterior pituitary secre- tions and of their actions upon their target organs. The timing and secretion rates are influenced by genetic factors which express themselves through the neurohumoral factors mentioned above and through reciprocal activ- ities of the steroid hormones secreted under pituitary influences. "Feed- b a c k " mechanisms appear to be important in this respect. A t present, this field is in a state of confusion and evidence obtained by a study of conditions in one species is often irreconcilable with t h a t obtained in another. Nalban- dov (1958) has subjected the available information t o critical review and is forced to conclude t h a t few, if any, comprehensive deductions can be drawn from it. T h e neurohumoral link is a means b y which exter- oceptive factors m a y influence reproductive processes.
12 S. A, Asdell
B . T h e Breeding Season 1. Seasonal Breeders
After F S H secretion has set in at about the time of puberty, its secretion is probably more or less continuous in m a n y species. These, characteristic- ally, have breeding, or estrous, cycles all the year round. But, in some, heat and ovulations occur only during restricted periods of the year.
These species, e.g., foxes and martens, which are spring breeders, and m a n y deer of the Temperate Zone, which are fall breeders, seem to be readily influenced by changes in the light gradient to which they are sub- jected (see Volume I I I ) . The available evidence suggests t h a t this controls in some way F S H release. I n seasonal breeders this is intermittent.
Besides this direct influence of climatic factors, a quantitative control of follicular growth is exerted by the amount of food available. T h e number of follicles t h a t ripen and the number of eggs shed depend to some extent upon the food supply, especially in those species t h a t shed m a n y at a time. I t is reasonable to believe t h a t this results from control of F S H output because in the normal female more eggs t h a n usual are shed if additional F S H is injected. Likewise, the atrophic ovary of the starved animal m a y be caused to grow and ripen follicles if it is stimulated in this way.
Caution is needed in relating behavior in t h e field to t h a t deduced from observations in zoos or under conditions resembling those of domestication.
The American bison (Bison bison), for instance, exhibited a seasonal and very intense breeding season in the wild (Warren, 1910). But, in zoos, they breed all the year round (Eckstein and Zuckerman, 1956b). When breeding has been in abeyance because of adverse food conditions, such as those t h a t might be encountered on the prairie during winter, a spring flush of grass will undoubtedly affect all the animals to an equal degree, so t h a t they come in heat almost simultaneously. They are all likely to be mated within a short period, so t h a t birth of the young is also concentrated within a relatively short period. Subsequently, new heat periods are likely to be synchronous, as they will return at a fairly definite stage, either as a result of a decline in lactation or in response to the new season. This t y p e of synchronization is most probable in species in which gestation periods last almost a whole year, with births near the beginning of a favorable season.
Environments with severe winters or with severe drought conditions during p a r t of the year are also conducive t o this type of seasonal breeding.
Under partial domestication food restrictions are not so critical and climatic conditions are usually more favorable t h a n those experienced in the field. The true capabilities of species are revealed under t h e more favorable conditions.
J. Reproduction and Development 13 If spermatogenesis in the male continues all the year, a very long breeding season m a y be suspected. A short period of spermatogenesis suggests a short season in the female. Another example of a different behavior in the wild and in captivity is afforded by the short-tailed wallaby, Setonyx brachyuruSj which has an anestrous period of 3-5 months in the wild, although, in captivity the female is capable of breeding a t any time of year (Sharman, 1955a).
I n species with a fairly short gestation period there m a y seem to be two definite breeding seasons during the year for a similar reason. T h u s Brown and Yeager (1945) report that, in Illinois, gray squirrels (Sciurus carolinen- sis) and fox squirrels (Sciurus niger) have two seasons. T h e first is in J a n u a r y and the second in J u n e and early July. A t the first of these seasons practically all the m a t u r e females are m a t e d and become pregnant. During pregnancy and lactation, heat periods do not recur, b u t as lactation nears its end the females all come in heat again practically simultaneously so t h a t a new, short, breeding season results. T h e number of breeding females in this season is increased by the young t h a t were born in the previous fall.
These were too immature to breed in the J a n u a r y season. T h e males do not have such clear-cut cycles of spermatogenesis and their breeding capacity does not disappear until the second season is over. I t is revived again late in December in readiness for the J a n u a r y season.
2. Duration of Heat
Since estrogens cause the symptoms of sexual desire in the female by their action upon the brain, the animal remains in heat as long as estrogen secretion is sufficient. T h e amount required to induce sexual receptivity varies widely from species to species. There is evidence to suggest t h a t the F S H level, the amount of estrogen required to bring the (ovariectomized) female into heat, the duration of heat in normal females, and the rate of estrogen excretion are directly related, a chain of circumstances t h a t is reasonably logical (Asdell, 1946b). I n most mammals the termination of heat closely follows rupture of the follicle, which presumably cuts off the supply of estrogens as the follicle luteinizes. But, in the cow ovulation occurs a t an interval of about 12 hours after the end of sexual receptivity. I n the hibernating bats of the temperate zones ovulation does not accompany the fall estrous period but, as they emerge from hibernation in the spring, ovulation soon takes place. Whether this spring ovulation is always assoc- iated with an estrous period has not been determined. I n the dog and the fox ovulation is early in the heat period, which lasts about a week.
Although estrogens, alone, are sufficient to induce heat there is evidence from several species t h a t the a m o u n t required to produce this effect is m u c h
14 S. A. Asdell smaller if a minute amount of progesterone is present a t the same time. This is quite probably a normal occurrence since, in those species with cyclic activities, the corpus luteum is waning at the time of heat. There is also evidence t h a t theca interna cells of a glandular type are activated as p a r t of the process of follicular ripening. I t is not by any means certain, though, t h a t these cells are secreting progesterone at this time.
C. Ovulation
Experimental work with rats has suggested the view t h a t F S H action upon the ovary in promoting estrogen secretion is a self-regulating device.
According to this view estrogens secreted in larger amounts as the follicles ripen act as a "feedback" t h a t stops or reduces F S H secretion and promotes L H secretion by the anterior pituitary. This prevents excessive F S H secre- tion and causes follicle rupture at the right moment as a result of L H secretion. This theory, however, does not explain the mechanism in the cow, for in t h a t species estrogens do not induce ovulation, but prevent it, possibly by blocking L H secretion. I n whatever way this latter hormone is caused to leave the pituitary, its expulsion from the cells appears to be quite rapid since the cells t h a t secrete it are quickly denuded of granules (Jubb et al., 1955). When secretion is initiated, ovulation soon follows. I n this process, as well as in t h a t of inducing sexual receptivity, progesterone plays a part, since a small quantity of this hormone injected early in heat advances the time of ovulation in the cow (Hansel and Trimberger, 1952).
Large quantities, on the other hand, given before the heat period suppress ovulation (Ulberg et al, 1951), possibly by interfering with F S H activity.
I n the rat, Everett (1948) has shown t h a t a small quantity of progesterone reinforces estrogen promotion of L H release.
/. Spontaneous and Provoked Ovulators
I n the majority of species whose reproduction has been studied in sufficient detail ovulation is spontaneous, i.e., it occurs without the need for a stimulus originating outside the animal. I n some, however, an external stimulus, usually t h a t of coitus is needed to produce this effect. T h e stimulus is usually central in transmission and it acts through the hypothalamus to provoke L H secretion. If a cat is in heat, mechanical stimulation of the cervix uteri is sufficient to evoke the reaction, according to Greulich (1934);
Gros (1936), however, was unable to elicit it in this way, so t h a t this problem needs to be reinvestigated.
The list of species t h a t are provoked ovulators is steadily increasing as more are investigated. A t present this list includes the European rabbit (Oryctolagus), hare, cat, ferret, mink and other mustelids, shrews, among
1. Reproduction and Development 15 t h e m the common and lesser European and the short-tailed (Blarina brevicauda) shrews, the thirteen-lined ground squirrel (Citellus tri- decemlineatus), the flying fox (Pteropus giganteus), European hedgehogs, Microtus guentheri, and the elephant seal. This enumeration does not suggest t h a t this type of ovulation is confined to any one group of mammals, b u t t h a t it is fairly widespread. T h e house mouse (Mus musculus) some- times exhibits this type of reaction (Allen, 1922), as does also t h e vole Microtus agrestis (Chitty and Austin, 1957). T h e short-tailed shrew requires several copulations, repeated a t brief intervals to elicit it (Pearson, 1944).
T h e evidence suggests t h a t coitus-provoked ovulation is the more prim- itive mechanism.
2. The Cause of Follicular Rupture
M a n y suggestions have been made to account for the rupture of the follicle wall and ejection of the ovum, b u t none of t h e m with sufficient evidence for unreserved acceptance. During the final, preovulatory matur- ation of the follicle a considerable amount of liquor folliculi is secreted and the follicle rapidly increases in size so t h a t it protrudes from the surface of t h e ovary. At t h e point at which rupture will occur the follicle wall becomes thinner and a clear area, the stigma, surrounded by a ring of engorged capillaries, arises. Just before ovulation there is a slight reduction of intra- follicular tension accompanied by buckling of the wall. When the stigma area gives way, part of the liquor folliculi, some granulosa cells, and the o v u m ooze from the cavity. Already, b u t to a varying degree, t h e theca interna with its blood vessels has been brought into more intimate contact with t h e granulosa cells and t h e corpus luteum is beginning to form. T h e luteinization process takes about 4 days for its completion, and, in most species, the corpus luteum displays its maximum activity in the non- pregnant female by t h e eighth d a y after ovulation.
3. The Corpus Luteum
Like the graafian follicle, the corpus luteum is a temporary endocrine organ. The length of its effective life depends upon the species and upon the event of pregnancy. If the eggs are fertilized and implanted it usually persists for the duration of the pregnancy, after which time degeneration and reabsorption are rapid. If pregnancy does not supervene, two possi- bilities exist. One is the development of t h e condition known as pseudo- pregnancy in which uterine and other reactions characteristic of a normal pregnancy are evident. I n t h e dog such a condition lasts nearly as long as a normal pregnancy and its termination is marked by arrangement of litter to form a nest and by milk secretion. I n t h e rabbit pseudopregnancy lasts
16 S. A. Asdell about half the duration of a pregnancy, i.e., 12-14 days instead of 30-32 days. At its conclusion the doe plucks fur from her breast and makes a nest, as is usual at the end of pregnancy. There is some evidence t h a t these terminal reactions depend upon prolactin for their evocation.
In the rat the corpus luteum of ovulation secretes but little progesterone and the maturation of new follicles is, in consequence, not held in check as it is in the presence of functional corpora lutea. If mating is not permitted, follicle maturation, heat, and ovulation recur at 4- or 5-day intervals. B u t if the cervix uteri is mechanically stimulated while the rat is in heat, the corpora lutea of ovulation are activated and pseudopregnancy, lasting for about 12 days, follows. This cervical stimulation is believed to cause prolactin release from the anterior pituitary.
I n those species with heats t h a t occur at fairly brief intervals the corpus luteum of the cycle has a span of life t h a t is relatively invariable. This is usually from 12 to 18 days; the actual length depends upon the species.
Decline of the corpus luteum permits a new wave of follicles to ripen so t h a t a new heat period results. Such a type is called a polyestrous species. H e a t s m a y recur throughout the year or they m a y be confined to a limited period each year.
D . The Reproductive P a t t e r n
Following the discussion in the earlier pages of this chapter it is now possible to summarize cyclical reproductive physiology in t h e series of alternative pathways t h a t follow.
(1) Pr-oestrum. Growth of follicles, thickening of vaginal wall. Increased vascularity of the endometrium.
(2) Estrus. Maturation and rupture of follicles. Period of sexual recep- tivity. Continuation of vaginal and uterine changes, with desquamation in the vaginal wall toward the end of the period. (In provoked ovulators this period m a y terminate in follicular atresia if the appropriate stimulus for ovulation is not received. Proestrum m a y follow.)
(3) Metestrum. Growth of corpus luteum, continued desquamation of vaginal epithelium. Some destruction in uterus (in guinea pig and cow).
(4) Alternatives:
a. Proestrum (in rat and mouse).
b. Diestrum. Corpus luteum mature. Rapid growth and maintenance of uterine glands. This is followed by involution of the uterine mucosa and glands. I n higher primates this involution m a y be precipitate and drastic.
1. Reproduction and Development 17 c. Pseudopregnancy. As in diestrum, b u t usually lasting longer and ending with reactions usually observed at t h e end of pregnancy.
d. Pregnancy. Prolongation and intensification of the diestrous re- actions under t h e influence of t h e placenta, embryo, or fetus.
(5) Alternatives:
a. Proestrum, as above.
b. Anestrum. Quiescence of the ovary and hence of the entire repro- ductive tract, followed eventually by proestrum.
Details of the reproductive patterns in each species, so far as they are known, m a y be found in Eckstein and Zuckerman (1956b) and in Asdell
(1946a).
E . T h e Union of Spermatozoon and Ovum 1. Life Span of the Gametes
Sufficient evidence has accumulated to show t h a t spermatozoa have a limited duration of fertile life within the female tract and t h a t ova, likewise, are capable of being fertilized over b u t a short period. These times amount to about 24 hours for spermatozoa and from 4 to 12 hours for ova. There are exceptions. Hibernating b a t s of temperate regions copulate in the fall and ovulation does not follow until the spring, when the females emerge from hibernation. During the intervening period the spermatozoa remain alive in the uterus with their heads embedded in the endometrium. This enables t h e m to survive. For m a n y years it was not known whether these spermatozoa retained their capacity to fertilize ova since some b a t s were known t o copulate in t h e spring for a second time. T h e question was finally settled by Wimsatt (1942), who kept female little brown b a t s (Myotis lucifugus) in a refrigerator through the winter and did not allow t h e m contact with males in the spring. These bats ovulated at t h e usual time and became pregnant as a result of their fall matings.
The ova of the dog and of the red fox (Vulpes vulpes) are shed from the ovary early in t h e heat period, b u t it is some days before they are capable of being fertilized because they are not sufficiently developed a t the time of follicular rupture. T h e ova of most species can be fertilized as soon as they are released, a condition t h a t depends upon the occurrence of t h e reduction division and extrusion of the first polar body. These events have usually t a k e n place while the eggs are still in the follicle. B u t in the dog and fox 2 days or more elapse after ovulation before this maturation process is completed. T h e spermatozoa, too, are able to survive until it has happened (Pearson and Enders, 1943).
18 S. A. Asdell Another interesting feature of reproduction in the males of several species of European bats has been described by Courrier (1927). This is the lack of association between spermatogenesis and the degree of development of the interstitial cells of Leydig. Consequently the accessory organs, de- pendent for their maintenance upon the testosterone secreted by these cells, do not follow exactly the cycle of spermatogenesis. B o t h spermato- genesis and the accessory organs are active in the fall (September) breeding season, b u t spermatogenesis soon ceases whereas the Leydig cells and the accessory organs continue to be active. Included is the epididymis which continues to be full of spermatozoa through the winter. I n the spring retrogression sets in and the male reproductive tract is completely inactive until the fall breeding season begins.
2. Sperm Transport in the Female Tract
The site of deposition of the semen is probably usually in the upper vagina in the neighborhood of the cervix uteri. B u t the anatomy of the glans penis in some species (e.g., in the ram) strongly suggests t h a t deposition m a y be within the cervix. T h e means by which the spermatozoa reach the upper oviduct, where the eggs are fertilized, has been the subject of much controversy, b u t the experimental evidence is rather scanty. I n some species this time is about 3-4 hours, b u t in others it is much less. V a n D e m a r k and Moeller (1951) have shown t h a t only 4 minutes or less are needed for this passage in the cow. Dead spermatozoa arrive in the upper oviduct in less t h a n 15 minutes. Obviously, such rapid transit is impossible for spermatozoa by means of their own flagellar means of locomotion. The uterine and oviduct musculature materially assists them. I n the oviduct of the rabbit secretory activity on the p a r t of t h a t organ is a factor, since, while the spermatozoa are in transit, the secretions discharge upward into the body cavity, t h u s carrying spermatozoa with them. Segmenting and pendular movements on the p a r t of the circular muscle help by distributing and mixing the fluids t h a t are present (Black and Asdell, 1958).
T h e number of spermatozoa ejaculated and deposited within the female tract is very large, of the order of a billion or more, b u t the number actually reaching t h e upper portion of the oviduct and t h u s having a chance to fertilize an ovum is small, usually fewer t h a n a hundred. Several of these m a y penetrate the zona pellucida, b u t only one enters the ovum proper, which becomes impervious to the entry of those t h a t arrive subsequent to the first (Braden and Austin, 1954). Rothschild (1956) has pointed out t h a t the attraction of spermatozoa to the egg by chemical stimuli emitted by the egg has not been demonstrated in mammals. However, the existence of fertilizins on the egg and antifertilizins in the spermatozoa has been
1. Reproduction and Development 19 demonstrated. Fertilization is almost invariably accomplished in the infundibular region of the oviduct. However, fertilization in the tenrec is usually achieved before the egg leaves the follicle.
Besides t h e maturing process t h a t the spermatozoa undergo in the epididymis, Chang (1951a) has demonstrated the existence of another maturing (or "capacitating") period of a few hours in the oviduct. R a b b i t spermatozoa introduced directly into the oviduct are not immediately so fertile as are those t h a t have been present there for some time. Austin and Bishop (1958) believe t h a t this capacitation consists of a shedding of the acrosome from the sperm head.
3. Ovum Transport in the Female Tract
Ovum transport through the oviduct takes about 3-4 days in most mammals, b u t the time is somewhat longer in the dog, in accord with the delay in maturation. Movement is most probably the result of segmenting and pendular activity on the p a r t of the circular muscle. T h e cilia seem to have little influence upon actual movement of the eggs, b u t they do, to- gether with the complex system of folds in the infundibular region, prevent their escape into the body cavity. Transport through the upper and middle portions of the oviduct takes b u t a few hours, b u t the tubo-uterine junction is an effective barrier t h a t delays the eggs a short distance above it (Black and Asdell, 1958). They remain in t h a t position for 2-3 days and their release to t h e uterus is probably due to relaxation a t the junction as the control exerted by estrogens wanes.
IV. Implantation and Placentation
A. Nourishment of the Blastocyst
The first cell divisions of t h e fertilized eggs are slow, and each succeeding cell is smaller t h a n its predecessor since no nourishment is obtained in the oviduct. B u t when the blastocyst reaches the uterus it is nourished by the uterine "milk" secreted by the endometrial glands. These have developed under the influence of progesterone. T h e corpora lutea have become fully organized a t about the same time as the eggs arrive. Chang (1951b), by ovum transfer experiments in rabbits, has demonstrated the importance of synchronization in the reproductive t r a c t of the female. Ova of a given age have little chance of survival unless they are transferred into a doe with corpora lutea of equivalent age. Ova obtained from the oviduct must be transferred to oviduct, those from uterus to uterus.
20 S. A. Asdell Β. Implantation
As a rule implantation begins a t about the seventh day after fertilization, b u t it is later, about the thirty-fifth day, in cattle. I n litter-bearing m a m mals the implantation sites tend to be evenly spaced throughout the uterus and, if the number of ovulations on either side has been uneven, some embryos m a y travel round the uterine body into the opposite horn.
T h e even spacing is brought about by movement of the embryos caused by contractions of the myometrium. Also, the endometrium in the immediate vicinity of one implantation becomes refractory to another implantation;
t h u s too-near crowding is avoided.
In the dikdik (Rhynchotragus kirkii) and some other antelopes implanta
tion is almost invariably in the right uterine horn. W h e n ovulation is from the left ovary the fertilized egg migrates to the right horn, probably by way of the corpus uteri (Kellas, 1954-1955).
1. Delayed Implantation
M a n y species exhibit the condition known as "delayed implantation."
I n this condition the blastocysts lie free within the uterine lumen, or b u t loosely attached to the wall, for a variable length of time without further development. Eventually implantation occurs and development is renewed.
T h e time taken in actual development is near t h a t which might be expected for a fetus of the species. This situation is common among some of the spring-breeding, fur-bearing carnivores (Mustelidae). I t is also prevalent in seals, b u t probably not in walruses (Scheffer, 1958). I t is also found in bears, armadillos, and the roe deer Capreolus. I n the carnivores and seals this device makes the duration of gestation nearly a year. M a t i n g follows parturition fairly closely. I n the roe deer mating takes place late in July or early in August instead of in October, the mating time for other deer of the Temperate Zone. Implantation does not take place until November, the normal time for the rest of the group, and birth is in M a y or June, as it is in the other species. T h e time of mating has changed, b u t not t h a t of birth or the time needed for development.
Among the Muridae and in some insectivores (Sorex), delayed implanta
tion with a physiological origin is found. H e a t and ovulation follow im
mediately after parturition and pregnancy m a y result. If the female is lactating at the time, implantation m a y be delayed for several days to a fortnight, with a corresponding prolongation of gestation. This type of delayed implantation is also encountered in t h e wallabies Setonyx and Protemnodon if pouch young are present (Sharman, 1955b). I t has been suggested b y Brambell (1937) t h a t the delay in implantation is due not t o the diversion of nutriment to the m a m m a r y glands, but to some secretion
1. Reproduction and Development 21 into the lumen of the uterus t h a t inhibits growth of blastocysts. B u t others believe t h a t a failure of the corpus luteum to secrete enough progesterone might delay implantation.
The cause of naturally delayed implantation is as obscure as is t h a t of the delay produced by lactation. I n martens (Martes) implantation has been hastened by exposing the females to increasing amounts of light, t h u s suggesting a pituitary involvement, but a t t e m p t s to hasten it by hormone (prolactin) injections have been disappointing. The nine-banded armadillo (Dasypus novemcinctus) provides evidence t h a t favors Brambell's inhibition hypothesis. I n this species, Buchanan et al. (1956) have found t h a t ovari- ectomy in the preimplantation stage shortens the period during which the blastocyst lies free. This result, however, is anomalous in t h a t the presence of corpora lutea is generally regarded as essential for implantation. How does the armadillo avoid this difficulty?
A recent symposium on the subject has been published (Enders, 1963).
C. Placentation
Placentation is by no means unique to the placental m a m m a l s (Meta- theria and E u t h e r i a ) ; it is also found in m a n y other classes of vertebrates.
W h a t seems to be new in m a m m a l s is the association of placentation with a functional corpus luteum. Even this structure in a simpler form is present in reptiles and elasmobranch fishes, b u t it is not necessarily associated with viviparity, nor is it, so far as is known, functional in these groups (Bragden, 1951). Moreover, it is well developed in the egg-laying mammals, the platy- pus and the echidna, but in these species it m a y have a function in inducing m a m m a r y growth.
1. Placental Types
The varieties of placentation are very many, and they have defied a t t e m p t s to fit t h e m into the framework of the accepted mammalian classification. I n evolutionary terms, modifications have been repeated within different families, especially in the larger ones. They are not, as a rule, confined to any one family. Thorough discussions of implantation by Boyd and Hamilton (1952) and of placentation by Amoroso (1952) m a y be consulted for details of these processes in m a n y groups and species.
I n ungulates, carnivores, lagomorphs, some primates, some edentates, and some of the vespertilionid bats the developing embryo lies in the uterine cavity and fills the lumen. This is described as central implantation.
In m a n y rodents the embryo takes up a position within a crypt of the uterine lumen. T h e crypt becomes obliterated as placentation develops. I n the guinea pig, some bats, some insectivores, and the chimpanzee and m a n
22 S. A. Asdell the embryo penetrates the endometrial epithelium and is buried within the endometrium. This is t h e interstitial t y p e of implantation (Boyd and Hamilton, 1952).
I n some species of rodents t h e embryonic disk is always on t h e side of attachment of t h e uterine ligaments (mesometrial). I n some b a t s and some insectivores it is always on the opposite side (antimesometrial); in other bats it is lateral.
T h e shape of t h e placenta has also provided a basis for classification.
Discoidal placentas are common among rodents, bats, insectivores, and primates; cotyledonary placentas are found in the r u m i n a n t s ; zonary ones in carnivores; and diffuse ones among the Perissodactyla and pigs. Amoroso
(1952) lists exceptions to these generalizations. T h e American mole Scalopus, the camel (Camelus), musk deer (Moschus), and lemurs all have diffuse placentas, whereas the aardvark (Orycteropus) has a zonary one. I n carnivores gradations are found from the zonary placentas of the dog and cat to the discoidal one of the brown bear.
The marsupial placenta is a somewhat simpler one t h a n t h a t of the Eutheria. The yolk sac is much larger in marsupials, and the allantois is small. Nourishment is absorbed through t h e wall of t h e yolk sac and a true vascular chorion with a trophoblast is not developed. Only in the koala
(Phascolarctos), the wombats (Phascolomidae), and t h e bandicoot (Pera- meles) is a vascular chorioallantois developed similar to t h a t in Eutheria.
I n t h e Eutheria a recent a t t e m p t at classification emphasizes t h e number of tissue layers through which exchange between the mother's circulation and t h a t of the fetus has to pass. This classification was suggested by Grosser (1927), and it has been modified by Mossman (1937b). I t has attracted a good deal of attention because of the suggestion t h a t the fewer the layers the more efficient should transfer become. This view seems to rest upon the assumption t h a t most of this transfer is by diffusion, not by chemical means. T h e work of Barron (1955) on oxygen transfer and Flexner and Gellhorn (1942) on sodium transfer supports the view t h a t the t y p e of placenta as denned by Grosser is a potent factor in t h e efficiency with which these substances pass into the fetus. B u t the theory has been severely crit- icized, mainly because of t h e difficulty of obtaining precise measurements t h a t can be related to tissue thickness (Flexner, 1955).
2. The Grosser Classification
T h e first t y p e of placenta according t o Grosser is t h e epitheliochorial one. I n this type six layers separate the maternal and fetal circulations.
These layers are fetal capillary endothelium, connective tissue, trophoblast, uterine epithelium, endometrial connective tissue, and capillary epithelium.
1. Reproduction and Development 23 This is the type found in the pigs, horses, whales, lemuroids, and in the mole, Scalopus aquaticus. T h e placenta is usually diffuse, and at parturition the maternal tissues are shed not at all or only to a slight extent.
T h e second type, t h e syndesmochorial, has lost the uterine epithelium with the result t h a t only five layers now intervene between the two circula- tions. This t y p e is found in the ruminants. The shape of t h e placenta is usually cotyledonary or multiplex. Some maternal tissue is lost at parturi- tion.
Type three, t h e endotheliochorial one, has lost, in addition to the uterine epithelium, t h e endometrial connective tissue. I t is general among the carnivores, seals, sloths, moles, shrews, tree shrews, and fruit-eating bats (Megachiroptera). T h e shape m a y be either zonary or discoid. Much maternal tissue is lost at parturition except in the European mole (Talpa), in which t h e maternal placenta and t h e fetal trophoblast both remain behind to be reabsorbed by leucocytosis.
I n t h e hemochorial type, t h e endothelium of the maternal capillaries has disappeared and t h e trophoblast is in direct contact with the maternal blood. This type, with three layers of tissue between the circulations, is found in most insectivores, some rodents, most bats (Microchiroptera), anteaters, armadillos, and primates. T h e placenta m a y be discoid or zonary in shape and it is deciduate.
T h e last type, the existence of which is disputed by some workers, is t h e hemoendothelial type. I n this, looped fetal capillaries are bathed by the maternal blood. I t has been described as present in the placentas of t h e higher rodents, or Hystricomorpha, and in the rabbits (Lagomorpha).
I n t h e more "advanced" types of placentas, i.e., in those with the fewer layers, the lower types m a y be found along with t h e higher ones. Develop- ment is gradual, and the ultimate type produced m a y be present only late in gestation.
I t is hardly necessary to restate t h e general belief t h a t the separation of the maternal and fetal blood circulation means t h a t only substances with a low molecular weight are able to cross the placenta. B u t the untoward effects on t h e h u m a n fetus of maternal infection with German measles during pregnancy, as well as those caused by the presence of certain blood- group combinations are well known. T h e degree of separation m a y not be as great as one is led to believe on anatomical grounds.
D . Placental Hormones
At the beginning of this chapter reference was made t o the fact t h a t t h e placenta is taking on the role of an endocrine organ. Evidence for this statement stems from several lines of investigation. Hormones, gonado-
24 S. A. Asdell tropes in particular, are found in the urine or blood of certain species during pregnancy and they have been traced in some instances to the placenta, from which they have been isolated. Hormone precursors have been detected in the placenta by histochemical techniques. Ablation of the ovaries or of the pituitary gland is not always followed by abortion. For obvious reasons not much is known of this aspect of reproduction in wild animals.
1. Steroids
Excretion of estrogens in the urine during pregnancy is widespread. I n m a n and the horse it is known to continue in the absence of the ovaries (Hart and Cole, 1934). I n the pig they are excreted early in gestation, then they disappear for a time but m a y again be found in the urine late in gestation. They are absent from the placenta during the period in which they cannot be detected in the urine (Faiermark, 1935).
Evidence for secretion of progesterone by the placenta rests partly on its detection in t h a t organ and partly on the results of ovariectomy. I n several species this operation is always followed by abortion. I n some, t h e embryos survive if the operation is performed after implantation has taken place.
I n those species in which the ovaries m a y be removed without terminating the pregnancy, the placenta is believed to supply the progesterone necessary for its continuation. In man, the rhesus monkey, mare, guinea pig, and sometimes in the rat, cow, and cat the ovaries m a y be removed without causing abortion. I n the mouse and rabbit abortion has always followed the operation. Logically, all t h a t is necessary is to excise the corpora lutea, leaving the rest of the ovary intact. B u t in the guinea pig the writer has found t h a t excision of corpora lutea is much more likely to be followed by abortion t h a n is complete ovariectomy. When some ovarian tissue is left new graafian follicles mature after the corpora lutea have been removed.
The result is estrogen secretion which m a y activate the myometrium. In horses and short-tailed shrews (Pearson, 1944), the corpora lutea retrogress well before the end of pregnancy. Probably the placenta in these species secretes enough progesterone to continue the pregnancy.
2. Gonadotropes
Evidence for the production of gonadotropic hormones by the placenta rests partly upon the effects of hypophysectomy during pregnancy and partly upon their detection in blood, urine, or the placenta. Hypophysec- t o m y has not been performed in pregnancy without causing abortion in t h e rabbit, cat, and dog. In the rat and guinea pig the operation has been performed in the second half of pregnancy without causing abortion. In
J. Reproduction and Development 25 the two latter species it m a y be inferred t h a t the placenta secretes enough prolactin to maintain the corpora lutea, though the role of the placenta in secreting progesterone m a y not be so dependent upon this hormone as are the corpora lutea. However, it is by no means certain t h a t prolactin is essential for progesterone secretion.
The blood serum of the pregnant mare contains a gonadotropic substance resembling F S H . This m a y be detected by the use of biological methods and it is present from the 50th to the 150th day after conception. I t has been traced to its source in the endometrial cups (Clegg et al., 1954). T h e appearance of this hormone coincides with the a b r u p t degeneration of t h e corpus luteum of pregnancy and the growth of a number of new follicles, some of which ovulate (Cole et al., 1931). New corpora lutea are produced from both the ruptured and the unruptured follicles. These regress at about the 150th day, after which time only their vestiges continue to term. At the same time the fetal gonads are stimulated so t h a t they greatly enlarge, owing t o t h e hypertrophy of t h e interstitial cells (Cole et al., 1933). This overgrowth is soon lost after the foals are born. The gonadotropin is found only in the blood plasma; it does not cross the kidneys into the urine. I t s ultimate fate is unknown.
A gonadotropin similar to t h a t found in the horse is present in the blood plasma of the pregnant ass, where it m a y be detected from the 47th to the 117th day (Ajello, 1950). But, according to Bielanski et al. (1955), the blood serum of a mare t h a t is carrying a mule does not give the gonadotropin reaction.
Perry (1954) records t h a t in the African elephant Loxodonta africana the corpora lutea, of which there are several present from the beginning of gestation, degenerate and a new set is formed, some of t h e m as a result of ovulations. These persist until parturition. During the latter half of preg- nancy the fetal gonad displays a large development of interstitial tissue similar to t h a t which is found in the foal.
Accessory corpora lutea also develop in the mountain viscacha (Pearson, 1949) and in the porcupine, Erethizon (Mossman and Judas, 1949), but in the latter they are apparently formed by luteinization of atretic follicles.
A gonadotropic hormone, chorionic in origin, is present in the blood plasma of the pregnant woman. I t passes through the kidneys and is found in the urine. This substance, with biological properties similar to those of L H , appears a t about t h e time of t h e first missed menstrual period, and its excretion rapidly rises to a peak at about 60-90 days of pregnancy.
Afterward its concentration rapidly declines, but it m a y be detected throughout gestation (Evans and Simpson, 1950). T h e detection of this hormone in pregnancy urine is the basis for the various pregnancy tests.
26 S. A. Asdell Chorionic gonadotropin has also been found in the urine of other higher primates during pregnancy. I n t h e chimpanzee Elder and B r u h n (1939) detected it from the second through the fourth month. The urine of the pregnant rhesus monkey gives a positive response for only a few d a y s : days 19 to 25 (Hamlett, 1937).
Wilkinson and de Fremery (1940) have reported t h a t the urine of the pregnant giraffe contains a gonadotropic substance, b u t its nature is still undetermined. I t would seem t h a t the excretion of gonadotropic substances is more widespread t h a n is suggested by t h e limited material from primates.
Probably m a n y other species would merit a test.
V. Development, Gestation, and Birth A. Development
Most of the characteristic developmental patterns t h a t distinguish m a m - mals from other chordates, and especially from the reptiles and birds, stem from two important differences. One is the minuteness of the egg with its small supply of yolk. The other is the elaboration of the placenta as a source of exchange between mother and embryo or fetus. Obviously, t h e monotremes occupy an intermediate position in this respect because their eggs, with diameters of about 3 m m . are larger t h a n those of marsupials and Eutheria, and no placenta is developed. Conditions resemble those of the Reptilia more closely t h a n they do in any other of t h e Mammalia.
Egg cleavage in the platypus is discoidal (meroblastic), as it is in reptiles and birds. The result is a limited embryonic portion poised upon a large yolk sac. The allantois is large, but it does not seem to have a nutritive function.
I n marsupials the eggs are smaller, about 0.15-0.25 m m . in diameter.
Cleavage is not quite equal (holoblastic), but yolk is extruded from the cell at the first cleavage. Even the small quantity present is apparently too much for the needs of the embryo. Gastrulation is of the avian and reptile type, but it is modified by the absence or reduction of t h e a m o u n t of uncleaved yolk. Hence a large blastocyst cavity develops both in marsupials and in Eutheria. T h e yolk sac is small, b u t it is important as a means by which nutrients m a y be absorbed from the maternal circulation. The allantois, too, is small. Only in a few marsupials, referred to earlier in this chapter, is there a large allantois with obvious functional significance.
The eutherian egg is small, the diameter ranging from 0.06 m m . in the mouse to 0.15 m m . in the sheep. The size of the egg bears little or no re- lationship t o t h e size of the adult of the species. As a vertebrate cell, t h e egg is large and the first few divisions after fertilization are made without
