Book - Sex and internal secretions (1961) 15

Embryology - 16 Apr 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 D Biology of Sperm and Ova, Fertilization, Implantation, the Placenta, and Pregnancy

Histochemistry and Electron Microscopy of the Placenta

George B. Wislocki, M.D., Sc.D. (Hon.) Late James Stillman Professor Of Comparative Anatomy And Hersey Professor Of Anatomy

and

Helen A. Padykula, Ph.D. Assistant Professor Of Anatomy, Harvard Medical School, Boston, Massachusetts

Between 1951 and 1955 George B. Wislocki worked enthusiastically on the original manuscript of this review. He viewed it as one of his most important contributions. A long delay in its pul3lication, since his death in October 1956, has necessitated considerable revision to permit current publication. Helen A. Padykula, who worked in close association with Dr. Wislocki, agreed to make this revision. Although the original manuscript has been altered in many ways, a genuine attempt has been made to preserve the dynamic approach to the study of placentation which characterized Dr. Wislocki.

I. Introduction

Histochemistry has elucidated the complicated architectural relationships of the mammalian placenta to an important degree. The localization of chemical constituents in tissue sections has provided a framework of evidence and suggestions for the function of various parts of the maternalfetal placental complex. Evidence for the sites of hormonal synthesis is principally of a histochemical nature. Furthermore, as has been true in the study of many tissues and organs, histochemical methods have also permitted further definition and differentiation of placental cell types. This is well illustrated in the use of histochemical methods to study the cellular intermingling which occurs at the maternal-fetal junction in the human placenta (Wislocki, 1951).

Descriptions of the ultrastructure of the placenta appeared soon after the introduction of methods for preparing sections thin enough to allow study with the electron microscope (Boyd and Hughes, 1954; Wislocki and Dempsey, 1955a; Wislocki and Dempsey, 195513; Dempsey, Wislocki and Amoroso, 1955; Dempsey and Wislocki, 1956 ) . There was immediate interest in defining the fine structure of membranes of great physiologic exchange (Low, 1953, pulmonary alveolar lining; Pease, 1955, nephron; Yamada, 1955, glomerulus; Palay and Karlin, 1959a, b, intestinal epithelium). Some comparisons of ultrastructure are already possible between the placental membranes and those mediating exchange in the kidney, intestine, and lung. Observations on ultrastructure and histochemistry are closely related, and this correlation will be delineated here for the placenta.

In this review, emphasis will be placed on the human and rodent placentas, although there will also be a consideration of the comparative histochemistry and histophysiology of the placental barrier. Amoroso's chapter in Marshall's Physiology of Reproduction gives an excellent account of comparative placentation, and thus no attempt will be made in the j^resent review to cover that subject, except for a brief introductory description of placental histology in man and the rhesus monkey. Huggett and Hammond ( 1952) , in a chapter in Marshall's book, drew, to a slight degree, on cytologic and histochemical data in discussing various aspects of fetal nutrition, placental metabolism, and the placental barrier. The endocrine functions of the placenta and the transport activities of the placental barrier will be considered only in so far as studies of the morjihology have contributed to an understanding of them.

Placental morphology is relatively difficult to understand, because of the complex structural relationships between the developing embryo and the uterus. These comprise a succession of stages of placental development involving implantation or nidation, followed by the gradual formation of the so-called definitive placenta. Through its brief life history, the mammalian placenta displays a pattern of differentiation which is manifested by structural, physiologic, and biochemical changes. Studies which define these changes emphasize that the placenta is different at various points in gestation. Histochemical methods have aided in tagging these changes in structure and function. It should be emphasized that, despite the kaleidoscopic changes which take place while the fetal membranes are variously differentiating into the so-called definitive placenta, the structures involved are functionally adequate at all times to meet the metabolic demands of the developing embryo and fetus.

Plate 15.1

Fig. 15.1. A normal human gestation sac containing an embryo approximately 1 month of age. The chorion laeve has been dissected away to repeal the embryo, amnion, exocoelom, yolk sac and chorionic membrane. Numerous delicate chorionic ^■illi extend outward from the chorion, constituting the fetal placenta. X 3'/2. (Carnegie Institution of Wavshington.)

Fig. 15.2. A section through a 16-day-old human ovum. Observe the embryonic shield (h) with the amniotic cavity above it and the yolk sac below (i). The dark stained chorion (ch) encloses the large exocoelomic cavity (;). Secondary villi (b) containing cores of mesoderm are in process of development. The villi are separated from one another by the intervillous space (/) which contains maternal blood. Peripheral to the secondary villi is a zone composed largely of cytotrophoblast comprising cell columns (d) and the trophoblastic shell (c). The latter is poorly demarcated from the surrounding decidua basalis (a) and decidua capsularis (e). X 30. (Rock and Hertig, 1948.)

In various groups of manmials further complications arise from the variety of modes of placentation, involving several kinds of nidation and subsequent intercrescence of the fetal and maternal membranes to constitute the placenta. The walls of the yolk sac and the allantois fuse in different ways with the chorion of mammals so that either a choriovitelline or chorio-allantoic placenta, or both, may differentiate. The former occurs in many orders of mammals (but not in man and monkey), usually preceding the latter in time of appearance, and disappearing after the allantoic placenta has become established. However, in some

groups {e.g., rodents) the two types of placentas continue to function concurrently throughout gestation. Until recently, the significance of the yolk sac placenta has been largely overlooked. The elegant physiologic experiments of Brambell and his associates (1948, 1950, 1951, 1956, 1957) have high-lighted this membrane by establishing it as the exclusive mediator of antibody transport from mother to fetus in rodents and lagomorphs.

As a consequence of the many varieties of placentas and numerous stages of differentiation of the various types, much research has been devoted to comparative placentation, seeking to elucidate the phylogenetic relationships, the topographic structure, and functional roles of the fetal membranes and placenta as they manifest themselves in different groups of mammals (Mossman, 1937; Amoroso, 1952). Despite extensive investigations of these questions, there are still large areas of uncertainty and lack of agreement in reference to the phylogeny of the placenta and of the structural homologies and functional significance of its various parts. As a result of the complexities of placental structure and the attendant preoccupation with comparative placental topography, the placenta is less well known histologically, cytologically, and histochemically than most organs of the body.

Plate 15.11

Fig. 15.3. A section through the placenta of a rhesus monkey on the 29th day of gestation. Secondary chorionic viUi are visible, each comprising a core of vascularized mesoderm surrounded by a darkly stained mantle of cytotrophoblast and syncytium, bordering the intervillous space (i) in which maternal blood circulates. The tips of the definitive villi extend downward as columns of cellular trophoblast (primary villi). The distal ends of these cell columns (c) unite on the periphery of the growing placenta to form the trophoblastic shell (.s). The latter merges indistinctly with the underlying decidually-transformed endometrium (d); uterine gland (g). Iron hematoxylin stain. X 6'/2.

Fig. 15.4. Localization of alkaline phosphatase in an anchoring villus from a human placenta of 8 weeks. The enz.yme occurs only in the outer margin of the syncytium. X 235. (Dempsey and Wislocki, 1945.)

Fig. 15.5. Indophenol oxidase reaction in a human chorionic villus of the 6th week of pregnancy. The oxidase activity is confined to the syncytial trophoblast. X 100. (Dempsey and Wislocki, 1944.)

Fig. 15.6. The birefringent lipids in the syncytium of a monkey's placenta. A fresh spread of the placenta was photographed between crossed prisms. X 80. (Dcmpsev and Wislocki. 1944.)

Fig. 15.7. Localization of basophilic substance in a villus from a human placenta of 13 weeks. An outer, eosinophilic zone (e) and a deeper, basophilic region (b) may be seen. Zenker formol fixative. Eosin methylene blue stain. X 1440. (Demp-sey and Wislocki, 1945.)

Fig. 15.8. A chorionic villus of a normal himian placenta of 10 weeks' gestation stained for iron by the TurnbuU blue method. The iron is concentrated in the outermost part of the stroma just beneath the trophoblast. Some particles staining for iron extend into the interior of the villus. X 250.

In view of the above considerations, a brief review of the topography and histologic structure of the placentas of man and rhesus monkey will be given early in this chapter. Later a histologic description of yolk sac placentation, as it occurs in the rat, will be presented. These morphologic descriptions will provide a background against which the various features of the localization of hormones, enzymes, and other compounds in these placentas will be more readily understood.

II. Histochemical Methods Utilized in the Study of the Placenta

The placenta is a most reactive organ histochemically; some portion of it reacts positively in most histochemical tests. This is not surprising in view of its manifold functions which surpass those of any other organ, with the possible exception of the adult liver. The cytologic methods which have been applied to placentas, such as those for the demonstration of mitochondria, Golgi apparatus, brush borders, reticular fibers, and nuclear morphology, are standard and need not be specifically cited here. On the other hand, histochemical methods have been more recently developed, and are also changing constantly in the light of new advances. To aid the reader in assessing the localizations illustrated in this review, a brief outline of some of the histochemical methods, as they were used in this laboratory, will be presented. The reader is referred to Pearse's Textbook of Histochemistry (1960) for fuller accounts of methods.

A. Basophilia And Acidophilia

Nucleoproteins and acid mucopolysaccharides are conspicuously basophilic substances. Desoxyribonucleoprotein is responsible for most of the basophilia of cell nuclei, whereas much cytoplasmic basophilia, especially in cells which are rapidly growing or synthesizing large amounts of protein, is attributable to ribonucleoprotein. The latter is also present in the nucleoli. The strong affinity of nucleoproteins and acid mucopolysaccharides for cationic or basic dyes is dependent on the presence in them of phosphate and sulfate groups which bear strong negative charge under many

routine conditions of staining. Under controlled conditions of staining, in which the pH of the staining solution is varied, it is possible to distinguish basophilia caused by nucleic acids and acid mucopolysaccharides from that of proteins. Because the isoelectric point of proteins differs, some stain readily with basic dyes whereas others react with acid dyes at a given pH. A discussion of the factors influencing basophilia and acidophilia of tissues is given in papers by Singer and Morrison (1948) and Singer (1952) . The controlled use of acid and basic dyes on the human and rat placenta is illustrated in the papers by Singer and Wislocki (1948) and Wislocki, Weiss, Burgos and Ellis (1957 ».

Many cell types of the growing, differentiating ])lacenta are rich in cytoplasmic ribonucleoprotein (Dempsey and Wislocki, 1946). It can be identified by com})aring histologic sections stained by methylene blue with control sections treated with ribonuclease before staining (Brachet, 1953). Deso.xyribonucleoprotein can be distinguished from ril)onucleoprotein by the Feulgen method.

The acid mucopolysaccharides which contain sulfuric acid, such as heparin, chondroitin sulfuric acid esters, and certain mucoid substances, are strikingly basophilic. This class of mucopolysaccharides can be identified in tissue sections which have been stained with basic dyes in a series of solutions of descending pH, inasmuch as the sulfate-containing moiety continues to stain at pH as low as 1 and 2, whereas the weaker acid groupings of polysaccharides, such as hyaluronic acid, cease to stain at pH 4.5 (Dempsey, Bunting, Singer and Wislocki, 1947).

In addition to exhibiting basophilia, acid mucopolysaccharides possess the property of shifting the absorption spectrum of certain thiazine dyes, thereby inducing metachromatic staining. Thus, toluidin blue and thionin, which color most basophilic substances blue, will stain acid mucopolysaccharides and, under some conditions, nucleoproteins red (Wislocki, Bunting and Dempsey, 1947). Michaelis (1947) attril)uted the metachromasia of dye molecules to their polymerization. An excellent analy

HISTOCHEMISTRY OF PLACENTA

889

sis of the phenomenon of metachromasia was made recently by Bergeron and Singer (1958).

B. CARBOHYDRATES

Alodcni liit^toc'hemistry of carbohydrates revolves principally around the periodic acid-Schiff (PAS) reaction developed by McManus (1945) and Hotchkiss (1948). This procedure involves the oxidation by periodic acid of 1,2-glycol linkages which are common in sugars. The resulting aldehydic groups are colored by the Schiff's reagent. The i)eriodic acid-Schiff reaction also stains other carbohydrate containing substances including glycoproteins, mucoproteins, and glycolipoproteins (Leblond, 1950). Glycogen can be differentiated histochemically from these substances by the use of saliva or malt diastase on control sections. In earlier studies of the placenta, glycogen storage was assessed by Best's carmine stain as well as l)y means of alkaline silver nitrate (Mitchell and Wislocki, 1944) . Also in much of the earlier work a digested control section was not used.

Sudan dyes, which dissolve readily in triglycerides, are most commonly used to demonstrate these lipids. Sudan black B has sul)erseded Sudan III and IV by virtue of its moi'e favorable color and also because it reveals many lipids in addition to triglycerides. Mitochondria are stained by Sudan black B presumably through their phospholipid content. There is some evidence that staining by Sudan black occurs also through chemical binding (Pearse, 1960). Osmium tetroxide is also employed for demonstrating lipid, although it is less specific than Sudan staining. This energetic agent, which is the principal fixative-stain for electron microscopy, will oxidize unsaturated compounds turning them grey or black. For the demonstration of phospholidips, Baker I 1 946 ) introduced an acid hematein test.

A group of histochemical reactions is characteristic of the lipid droplets of the xferoid-producing organs. Besides staining intensely with Sudan black B, these lipid droplets, in frozen sections of formalin-fixed material, react positively in tests for carbonyl groups, such as the Schiff test and

hydrazine methods ( phenylhydrazine method of Bennett (1940) and the naphthoic acid hydrazide method of Ashbel and Seligman (1949)). They are fluorescent when viewed with ultraviolet light and contain birefrigent crystals (Dempsey and AVislocki, 1944; Wislocki and Dempsey, 1946a, b; Kockenschaub, 1952). The droplets also give colored products when treated with mineral acids, as in the LiebermannBurchardt reaction or the Schultz reaction for cholesterol. All of the above reactions are prevented following extraction of the sections with acetone at room temperature.

In the ])ast it was proposed that all or some of these reactions might be given by ketosteroids, i.e., by hormones and their immediate precursors (Dempsey and Wislocki 1946). Seligman and Ashbel (1951) adduced evidence for the ketonic nature of some of the carbonyls and for the specificity of the naphthoic acid-hydrazide reaction in the steroid hormone-producing glands. However, it was gradually realized that none of the above reactions is specific for ketosteroids, and a reevaluation of their significance was undertaken by Deane and Seligman (1953). More recently Karnovsky and Deane (1954, 1955) have shown by both chemical and histochemical means that the carbonyl grouj^s in adrenal cortical lipids seem to be exclusively aldehydes produced by the auto-oxidation of unsaturated fatty acids during fixation. The reaction is inhibited by the addition of various anti-oxidants and chelating agents to the fixative. Moreover, hydrazine reactivity occurs only when there is reactivity to the Schiff reagent, and the appearance of aldehydes is correlated with the disap])earance of double bonds in the lipids.

On the other hand, the experimental observations of Kai'iiovsky and Deane proved rather conclusively that the percentage of unsaturated fatty acids which are the source of the aldehyde groups is high in functionally active adrenals and declines in inactive glands, thus confirming the earlier assertions (Bennett, 1940; Wislocki and Bennett, 1943; Dempsey and Wislocki, 1946) that the intensity of the carbonyl reaction correlates well with secretory activity.

890

SPERM, OVA, AND PREGNANCY

Furthermore, despite the fact that none of the various reactions is specific for ketosteroids, their occurrence in lipid droplets of the adrenal cortex, gonads, and placenta, and their variations under different physiologic conditions of these organs suggest that the methods are empirically useful in identifying the probable sites of ketosteroid hormone formation. Application of the above procedures for the identification of sites of steroid hormone production will be found in the following papers: for the adrenal cortex (Bennett, 1940; Deane and Greep, 1946) , ovary (Dempsey and Bassett, 1943; Deane, 1952), testis (Pollock, 1942), and placenta (Wislocki and Bennett, 1943; Dempsey and Wislocki, 1944; Wislocki. 1952).

D. ENZYMES

Numerous histochemical methods have been devised for preserving enzymes so that they will retain activity toward either naturally occurring or synthetic substrates. In most of these procedures an insoluble primary reaction product is precipitated in situ in frozen sections, and it is then visualized by a color reaction which reveals the location of the enzyme. The methods of Gomori (1941a, b) for acid and alkaline phosphatases, utilizing a variety of substrates, were applied to the study of the placentas of various animals (Dempsey and Wislocki,

1947). Observations by Padykula (1958) on fluctuations of phosphatase activity of the rat placenta included changes in adenosine triphosphatase, as well as acid and alkaline phosphatases.

The presence of tlie cytochrome oxidasecytochrome c system has been demonstrated by the indophenol blue oxidase reaction by treating fresh spreads of placental villi with the nadi reagents (paraphenylene-diamine and a-naphthol (Dempsey and Wislocki, 1944). A series of oxidation-reduction indicator dyes was applied similarly to determine the oxidation-reduction potential of various placental elements.

Succinic dehydrogenase activity of the human placenta, as illustrated in this review, was localized by the tetrazolium method developed by Seligman and Rutenburg (1951) and modified by Padykula (1952). The localization of this enzyme in the rodent placenta was reported by Telkka andLehto (1954), Reale and Pipino (1957), and Padykula (1958). A comparative study of the distribution of this oxidative enzyme in various placental types was made by Reale and Pipino (1959).

Esterase activity was localized in the human placenta in this review according to the procedures of Nachlas and Seligman (1949) and Barrnett and Seligman (1952). In the rat placenta, Padykula ( 1958) demonstrated this enzymatic activity by the

Plate 15. Ill

Fig. 15.9. The trophoblast of a secondary chorionic villus of a 30-day human placenta stained with Mallory's connective tissue stain. Observe the syncytium which contains small nuclei and possesses a brush-border on its free surface facing the intervillous space. Notice the large chromophobic Langhans cells which rest upon a deeply stained basement membrane contiguous to a fetal capillary which contains several nucleated erythrocytes. X 1600. (Wislocki and Bennett, 1943.)

Fig. 15.10. The localization of basophilic substances in a human villus of 13 weeks. Zenker formol fixative. Methylene blue staining. Basophilia is intense in the inner layer of the syncytial cytoplasm whereas the marginal zone and brush border are unstained. The cytoplasm of the Langhans cells is also unstained. The basophilic staining of the sj^ncytial cytoplasm, but not that of the nuclei, is prevented by prior exposure of sections to ribonuclease, a result which indicates that the cytoplasmic basophilia is due to the presence of ribonucleoprotein. X 10 ocular; X 60 objective. (Dempsey and Wislocki, 1945.)

Fig. 15.11. Localization of basophilic substances in a cytotrophoblastic cell island from a human placenta of 13 weeks. The cytotrophoblastic cells contain strongly basophilic material in their cytoplasm, arranged in a fashion reminiscent of Nissl substance in neurons. In addition, the intercellular matrix is weakly basophilic. Zenker formol fixative. Methylene blue staining. (Dempsey and Wislocki, 1945.)

Fig. 15.12. The same region as in Figure 15.11, from a contiguous section, stained following treatment with ribonuclease. The basophilia of nuclei and the intercellular matrix is unchanged, whereas the cytoplasmic basophilia has been abolished. (Dempsey and Wislocki, 1945.)

Pearse procedure. Gomori (1945) devised a method for lipase which has been applied to placentas of several different groups.

E. Iron and Sulfhydryl and Amino Groups of Proteins

Microincineration of tissue sections (Scott, 1933) is a well known means of demonstrating minerals in cells. It was used for the identification of placental iron by Dempsey and Wislocki (1944). Placental iron is also demonstrable in tissue sections by various methods invoh'ing the Prussian blue reaction.

Protein-linked sulfhydryl groups have been demonstrated in the human placenta in the present study by the method of Barrnett and Seligman (1952).

Amino groups of proteins were demonstrated in the granular cells of the metrial gland of the rat by the method of Weiss, Tsou and Seligman (1954).

III. The Placentas of Man and Rhesus Monkey

A. Topography and General Histology

The comparative placentation of primates has been extensively studied by Wislocki (1929), Hill (1932), and Stieve (1944). The placentation of the rhesus monkey {Macaca mulatta) has been described in detail by Wislocki and Streeter (1938).

The human placenta in its earliest stages has been the subject of recent observations by Hertig and Rock (1941, 1945), Rock and Hertig (1948), Hertig, Rock and Adams (1956). In subsequent stages it has been investigated by Grosser (1925a), Spanner (1935a, b, 1940), Stieve (1940, 1941), Stieve and von der Heide (1941), Wislocki and Bennett (1943), and Hamilton and Boyd (1951, 1960).

The chorion, or outer membrane surrounding the developing implanted egg, unites with the vascular, allantoic body stalk in man, apes, and monkeys to give i-ise to the definitive placenta which is hemochorial and villous in form. The essential cells or parenchyma of the chorion and placenta are collectively called the trophoblast, a term introduced by the Dutch embryologist Hubrecht, signifying "nutritive layer."

At the site of implantation of the developing blastocyst in the uterine endometrium, cords of trophoblast grow out from the initially smooth surface of the chorion. These tongues which penetrate and erode the endometrium are called the primary chorionic villi. On approximately the 15th day after fertilization of human and rhesus monkey eggs, mesoderm begins to appear in the proximal, attached portions of the villi and differentiates progressively toward their growing distal ends. The differentiation of mesoderm in this manner converts the primary chorionic villi into secondary villi. With the development of embryonic blood vessels in them, the secondary villi become converted into tertiary or definitive placental villi. The secondary and tertiary villi are covered by a mantle of trophoblast differentiated into an inner layer of large, vesicular trophoblastic cells or trophoblasts designated as Langhans cells after their discoverer, and an outer layer of syncytially transformed cells referred to as the syncytial trophoblast or syncytium. The latter, as the name implies, consists of a continuous mass of cytoplasm devoid of cell boundaries and containing small, darkly stained nuclei (Fig. 15.9).

The distal ends of the chorionic or placental villi continue to grow for a considerable period in the form of columns of cytotrophoblast preceding the differentiation of mesoderm in them. These are designated as the trophoblastic cell columns. The distal tips of these columns unite on the periphery of the growing placenta to constitute the trophoblastic shell. These structures will be referred to collectively as the peripheral cytotrophoblast, and the cells comprising them as peripheral trophoblasts, in contradistinction to the Langhans cells associated with the secondary and tertiary villi. As will be shown, the peripheral trophoblasts have numerous cytologic and histochemical properties which distinguish them from the Langhans cells. The beginning of many of these features can be seen in a human blastocyst on the 16th day following fertilization (Fig. 15.2) and their further development is illustrated in a placenta of a rhesus monkey of the 29th dav of gestation (Fig. 15.3).

Placental villi grow initially everywhere over the circumference of the human chorion (Fig. 15.1). However, on one side, where the thick, well vascularized endometrium favors their growth, the villi become long, branched, and profuse, forming the chorion frondosum which eventually gives rise to the definitive, discoidal placenta.

As the trophoblast invades the endometrium giving rise to the placental villi, dilated endometrial blood vessels are eroded and tapped, their contents providing nourishment for the proliferating blastocyst. The maternal blood lacunae become increasingly confluent, forming a more or less continuous intervillous space between the placental villi. This space acquires afferent and efferent connections with the arteries and veins of the uterine wall, as a consequence of which maternal blood begins to circulate in the intervillous space and to bathe the surface of the villi. In the mesodermal stroma of the placental villi, newly formed blood vessels become connected with the blood vessels and developing heart within the growing embryo, so that toward the end of the first month, fetal blood begins to circulate in the capillaries of the villi. The secondary and tertiary villi continue to lengthen and branch. A number of large stem or anchoring villi extend across the intervillous space to attach to the trophoblastic shell or basal plate which is apposed to the uterine decidua.

As the blastocyst grows, the endometrium surrounding it undergoes significant changes. The zone in which the trophoblastic shell and the endometrium meet has been variously called the "junctional," "composite," or "penetration" zone. The endometrium subjacent to the junctional zone comprises an outer cellular portion, the stratum compactum and a deeper glandular part, the stratum spongiosum, the latter characterized by the presence of conspicuous, actively secreting glands. The spongy layer extends down to the basal zone, the latter forming a narrow strip which is contiguous to the myometrium and contains the fundic ends of the uterine glands. The stromal cells of the compact zone and, later, of the glandular zone, become transformed into large polygonal elements termed decidual cells. This decidual transformation occurs throughout the entire extent of the endometrium. That portion of the decidua directly beneath the implanted blastocyst constitutes the maternal part of the placenta and is called the decidua basalis. It is attacked and resorbed to a large degree by the growing placenta. Its remnants, at the time of birth, are either expelled or undergo resorption. The endometrium renews itself in the puerperium from the basal zone and the deep residual portion of the stratum spongiosum.

Plate 15. IV

All of the figures (excepting Fig. 15.13) on this plate are drawings of frozen sections of material fixed in 10 per cent buffered formalin.

Fig. 15.13. The trophoblast of a human chorionic villus of 30 days. Stained with osmic acid by Champy's fixation, a procedure which differentiates lipids. The syncytium contains numerous fat droplets, whereas the Langhans cells contain none. In the stroma beneath the trophoblast a typical, vacuolated Hofbauer cell is visible. These fat droplets are similarly revealed after staining with sudan dyes. X 1600. (Wislocki and Bennett, 1943.)

Fig. 15.14. Human placental villus at full term stained with sudan black B showing minute lipid droplets in the syncytium. Some villi contain more lipid particles than this one. X 7 ocular; X 60 objective. Compare Figures 15.13 and 15.14 wdth Figures 15.18 and 15.19, the latter stained by the Ashbel-Seligman carbonyl method.

Fig. 15.15. Placenta of a pig (fetal crown to rump length, 120 mm.), showing the chorion in apposition to the endometrium. Sudan black B. The uterine epithelium contains numerous black, sudanophilic droplets whereas none are visible in the faintly gray staining trophoblastic cells of the chorion. Compare with Figure 15.22, showing the Ashbel-Seligman carbonyl reaction which is identical in its distribution with the sudanophilia. X 10 ocular; X 20 objective.

Fig. 15.16. The labyrinth of the chorio-allantoic placenta of a rat on the 18th day of gestation, showing numerous sudanophilic lipid droplets in the cytotrophoblasts. Compare with Figure 15.21, showing the Ashbel-Seligman carbonyl reaction.

Fig. 15.17. The labyrinth of the chorio-allantoic placenta of a mouse on the 18th day of gestation stained by the Ashbel-Seligman carbonyl reaction. The placental labyrinths of mouse and rat are identical in respect to their sudanophilia and carbonyl reactions. X 10 ocular; X 60 objective.

The trophoblast forms the parenchyma of the placenta and the major element of the placental barrier. It mediates the metabolic exchange between mother and fetus. It provides for the nutrition of the embryo, at first, by the local destruction and absorption of the uterine decidua, and later, by transmission of metabolites through the syncytial trophoblast from the maternal to the fetal blood streams. It serves also as an avenue for the excretion of various fetal waste products. The human trophoblast is also an important endocrine organ which produces steroid hormones, chorionic gonadotrophin and other hormones.

B. Chorionic Villi

1. Trophoblasts: the Langhans Cells

These conspicuous cells which possess large nuclei constitute a germinal bed in which mitoses are frequently seen and from which in the early part of gestation the syncytium is evidently derived. The Langhans cells gradually diminish in number, but some of them survive until the end of gestation (Wislocki and Bennett, 1943; Wislocki and Dempsey, 19551. Electron micrographs show that the Langhans cells are closely apposed on their outer surfaces to the syncytium and on their basal surfaces to the basement membrane of the stroma of the chorionic villi.

The cytoplasm of the Langhans cells is characteristically chromophobic (Wislocki, Dempsey and Fawcett, 1948 j, exhibiting only faint cytoplasmic basophilia (Fig. 15.10), little affinity for acid dyes (Fig. 15.9), and no metachromasia. This lack of basophilia correlates with their meager endoplasmic reticulum, i.e., ergastoplasm (Wislocki and Dempsey, 1955) . In the first 4 to 6 weeks of gestation their cytoplasm contains a considerable amount of glycogen, stainable by the PAS method. This glycogen subsequently disappears. Except for slight staining adjacent to the nuclear membrane (Fig. 15.54) , the cytoplasm is negative with the PAS reagents following the removal of glycogen. There are no lipid droplets.

There is a moderate number of rod-shaped and granular mitochondria (Wislocki and Bennett, 1943) . In electron micrographs, the mitochondria of the Langhans cells are relatively few in number but larger than those in the syncytium. A moderate degree of succinic dehydrogenase activity is demonstrable in these cells (Figs. 15.37 and 15.38). The Golgi apparatus is situated on the side of the nucleus toward the syncytium (Baker, Hook and Severinghaus, 1944).

The cytoplasm is faintly stained in the reaction for protein-linked sulfhydryl groups (Figs. 15.33 and 15.34). The acid and alkaline phosphatase activities are of a low order.

2. Trophoblastic Syncytium in the First and Second Trimesters of Pregnancy

Free surface. The syncytium constitutes a broad layer of cytoplasm without cell boundaries and possesses small, irregularly shaped, darkly staining nuclei which are rather uniformly dispersed in its inner zone (Fig. 15.9). It possesses an outer surface, facing the intervillous space, which is extremely variable in structure, ranging from a foamy, vacuolated border possessing delicate streamers and fronds, through various intermediate appearances, to regions where it bears a well defined brush border (Wislocki and Bennett, 1943). In the earliest stages the former appearances predominate, but as gestation advances, the brush border increases in amount. It was suggested by Wislocki and Bennett that these variable surface appearances in fixed material indicate that the living syncytial cytoplasm is pleomorphic and plastic. The parts that are foamy and vacuolated and possess streamers are probably constantly moving and flowing, a physiologic activity that would promote the alisorption of fluid and metabolites from the intervillous space by the process of pinocytosis (Wislocki and Bennett, 1943). In confirmation of this, observations of explanted bits of placenta growing in tissue cultures show that the cytoplasm of both the syncytial and cellular trophoblast moves quite actively, giving rise to a variety of streamers and threadlike processes (Friedheim, 1929; Jones, Gey and Gey, 1943) .

The ultrastructure of the free surface of the syncytium at 9 to 10 weeks typifies that of a pinocytotic membrane (Wislocki and Dempsey, 1955).^ A profusion of microvilli of various shapes reach into the maternal blood. Some microvilli are long and slender with enlarged tips; others are short and thick; and occasionally peninsulas of cytoplasm studded with microvilli extend into the maternal blood space. Often in the marginal zone immediately beneath the microvilli there are large vesicles containing finely stippled or flocculent material. They are occasionally seen in the large tongues •of cytoplasm which protrude from the free surface. These vesicles are most likely formed as a result of pinocytotic activity.

Basal surface. The inner surface of the syncytium is approximated to the surfaces of the Ijanghans cells or the subjacent stroma of the placental villus (Fig. 15.9). As mentioned previously, the Langhans cells gradually diminish in number so that the syncytium eventually comes widely in contact with the chorionic stroma. However, even early in gestation, there are occasional gaps between the Langhans cells where the syncytium is in direct contact with the subjacent mesenchyma. Through these gaps metabolites traversing the placental barrier can by-pass the Langhans cells.

Cell organelles. Cytoplasmic basophilia is very intense in the broad inner zone of the syncytium, especially surrounding the nuclei (Figs. 15.7 and 15.10). Since this basophilia is abolished by ribonuclease, it has been attributed to the presence of ribonucleic acid (Dempsey and Wislocki, 1945). Further evidence supporting this identification is derived from: (1) the similarity of this basophilia to that of Nissl substance (Singer and Wislocki, 1948), (2) the metaohromasia of this region in young placentas (Wislocki and Dempsey, 1948), and (3) the <'oncentration of the endoplasmic reticulum (ergastoplasm) in the inner two-thirds of the .syncytium (Wislocki and Dempsey, 1955). This rich cytoplasmic basophilia, which has been shown to constitute the microsomal fraction of the biochemist (Palade and Siekevitz, 1956) points toward an active participation by these cells in protein synthesis.

• Bargmann and Knoop (1959) have also described the ultrastructure of the human placental

barrier. They emphasize the syncytial nature of the outer trophoblastic layer, and offer further description of the ultrastructure of the Langhans cells, Hofbauer cells, and stromal cells.

In contrast to the inner zone, the outer zone of the syncytium is strongly acidophilic, although the narrow outermost zone corresponding to the brush border is less acidophilic (Singer and Wislocki, 1948). This acidophilia suggests the occurrence of basic proteins in the outer zone. The ultrastructure of this region shows that there are relatively fewer ergastoplasmic elements but there is a concentration of large vesicles which most likely are the products of pinocytosis. Higher resolution electron microscopy of this region is needed. It should be mentioned at this point that protein-bound sulfhydryl groups are concentrated especially at the inner and outer borders of the syncytial cytoplasm (Figs. 15.33 and 15.34).

Mitochondria are abundant in the syncytium, occurring as small granules and rods (Figs. 15.26 and 15.30). There is the indication of high succinic dehydrogenase activity here at 6 weeks of gestation, although this histochemical determination was complicated by the presence of lipid in the syncytium (Figs. 15.37 and 15.38). With the electron microscope, it was observed that the mitochondria are smaller but more numerous than those in the Langhans cells (Wislocki and Dempsey, 1955).

The Golgi apparatus forms a dispersed network in the syncytium. This organelle has been described in the various cells of the placenta by Acconci (1912), Wislocki and Bennett (1943), and Baker, Hook and Severinghaus (1944).

Glycogen and other pas positive material. In the human placenta in the first month of gestation a moderate amount of glycogen is stored in the syncytial trophoblast. It disappears almost entirely by the end of the second month. Similar early storage and loss of glycogen occur in the Langhans cells and stromal fibroblasts.

PAS positive material which is resistant to digestion by saliva is conspicuous in the brush border and marginal cytoplasm of the syncytium (Figs. 15.29 and 15.54). A faint red stippling of reactive material is also visible in the deeper cytoplasm (Fig. 15.54). A dark red reaction occurs also in the basement membrane upon whicli the Langhans cells or syncytium rest (Fig. 15.54 1 .

Lipids. Birefringent, sudanophilic lipid droplets are abundantly present in both the inner and outer zones of the syncytium (Figs. 15.6 and 15.13). The droplets are acetone soluble. They react with phenylhydrazine (Wislocki and Bennett, 1943), give a positive Schiff reaction, exhibit yellowish green fluorescence (Dempsey and Wislocki, 1944; Rockenschaub, 1952), and give a positive naphthoic acid-hydrazide reaction (Fig. 15.18) (Ashbel and Seligman, 1949; Seligman, Ashbel and Cohen, 1951; Wislocki, 1952; Ashbel and Hertig, 1952). These histochemical reactions occur also in the lipid droplets of the gonads and adrenal cortex, and their occurrence in the syncytium suggests that it is the site of steroid hormonal synthesis in the placenta. Previous and later paragraphs discuss this problem more fully.

Enzymes. Alkaline phosphatase (Figs. 15.4, 15.41, and 15.42) occurs in the syncytial trophoblast (Buno and Curi, 1945), where it is slight in amount at 6 weeks, but it increases tremendously as gestation advances (Dempsey and Wislocki, 1945). It varies in degree of activity according to the substrate used, the enzymatic reaction being most intense following the use of fructose diphosphate and nucleic acid, and less so with glycerophosphate and adenylic acid (Dempsey and Wislocki, 1947). The reaction occurs earliest and reaches its maximal intensity in the brush border, although, as gestation proceeds, it spreads throughout the syncytial cytoplasm and also involves the nuclei. However, the localization of the enzymatic reaction within the nuclei may not represent the actual distribution in the living state, for investigations have shown that the reaction products are capable of migrating, especially from cytoplasm to nuclei (Martin and Jacoby, 1949; Leduc and Dempsey, 1951 ; Herman and Deane, 1953). This enzyme is a distinguishing feature of the great absorptive surfaces of the small intestine, proximal convoluted tubule of the kidney, and syncytial trophoblast of the placenta.

Acid phosphatase (Figs. 15.39 and 15.40) occurs in great intensity in the syncytium in both cytoplasm and nuclei (Wislocki and Dempsey, 1948). The nuclear staining may not represent the true location of the enzyme, because on localizing the enzyme by both histochemical and biochemical methods and comparing the results, Palade ( 1951 ) found in the case of the hepatic cells of the rat that the enzyme was confined almost entirely to the cytoplasm. Furthermore, recent biochemical evidence indicates that this enzyme is located in a particular cytoplasmic fraction, the lysosomes (DeDuve, 1959).

Plate 15. V

All of the drawings on tliis i)late (except Fig. 15.23) are of frozen sections of material fixed in 10 jier cent l^uffered formalin and stained by the Ashbel and Seligman method for carbonyl groups. Comi^are the illustrations on this plate with those on Plate 15.IV.

Fig. 15.18. The syncytial trophoblast of a human villus at 5 months of gestation, stained by the carbonyl method. The reaction is localized in the lipid droplets of the syncytium. Compare with Figure 15.13. X 7 ocular; X 90 objective.

Fig. 15.19. A human placental villus at full term illustrating the positive carbonyl reaction of the minute lipid droplets in the syncytium as well as a diffuse reaction of the entire syncytial cytoplasm. Compare with Figure 15.14. X 7 ocular; X 60 objective.

Fig. 15.20. The placental labyrinth of a cat (fetal crown to rump length, 75 mm.) showing an intense carbonyl reaction in lipid droplets located in the cytotrophoblasts of the placental lamellae. A diffuse lavender reaction is present in the contiguous trophoblastic syncytium. X 10 ocular; X 40 objective.

Fig. 15.21. The carbonyl reaction in the placenta of a rat of 18 days of gestation. Compare with Figure 15.16 illustrating the distribution of sudanophilic lipids in the rat's placenta, and with Figure 15.17 illustrating the carbonyl reaction in the labyrinth of the mouse. X 10 ocular; X 40 objective.

Fig. 15.22. The carbonyl reaction in the placenta of a pig (crown to rump length, 90 mm.). The reaction is localized in the uterine epithelium, whereas little staining is apparent in the chorion. The distribution of the reaction coincides with that of lipids revealed by sudan dyes (Fig. 15.15). X 10 ocular; X 40 objective.

Fig. 15.23. The placenta of a pig (fetal crown to rump length, 235 mm.). Bouin's fixation. Masson's stain. This figure is shown to illustrate the detailed structure and relative thinness of the pig's placental barrier. It shows "intra-epithelial" maternal capillaries {m.c.) in the uterine epithelium (ep), and "intra-epithelial" fetal capillaries (/.c.) in the chorionic syncytium (c/i) ; the distance separating the two sets of capillaries varies between 6 and 8 /x. Compare with Figures 15.28 and 15.64 which also illustrate the extreme thinness of the trophoblast and the narrow distance separating the maternal from the fetal capillaries in the placenta of the pig. X 70 ocular; X 40 objective.

High esterase activity is demonstrable in the troi)hoblast following the use of the method of Barrnett and Seligman ( 1952) on unfixed frozen sections ( Wislocki, 1953) . It could not be determined with certainty whether the intense but poorly localized crystalline reaction product was entirely in the syncytium or whether some was present in the Langhans cells (Fig. 15.36). With the method for esterase (Nachlas and Seligman, 1949) carried out on sections of acetone-fixed, paraffin-embedded material, the <trophoblast shows no reaction. Acetone fixation destroys cholinesterases and some aliesterase, as a consequence of which a tissue poor in aliesterase would be negative. The combination of the positive Barrnett and Seligman reaction on unfixed frozen sections and of the negative reaction by the method of Nachlas and Seligman on acetone-fixed sections indicates that cholinesterases and possibly a small amount of aliesterase are present in the trophoblast. Cholinesterase has been found in high activity in the human placenta by chemical means (Torda, 1942; Ord and Thompson, 1950). Acetylcholine has been reported by a histochemical method devised by Wen, Chang and Wong (1936) as occurring in large amounts in the border and on the surface of the syncytium.

The indophenol oxidase reaction was performed with the nadi reagents on fresh, unfixed, teased villi, and on unfixed, frozen sections of human placenta of the 6th week of gestation (Dempsey and Wislocki, 1944). Indophenol blue appeared in the syncytium (Fig. 15.5), where it was interpreted as revealing the presence of the cytochrome oxidase-cytochrome c system. In contrast to the syncytium, the stroma of the villi did not give the indophenol reaction.

A series of reduction-oxidation indicators was similarly applied to placental villi. The syncytium concentrated the dyes in their oxidized form, whereas the stroma of the villi reacted far less intensely. These results indicate that the syncytium is maintained in the air at a more positive reduction- oxidation potential than the stroma and are in keeping with the similar distribution of indophenol l)lue.

3. Stroma of the Chorionic Villi; Hofbauer Cells

The stroma of the villi consists of mesenchymal connective tissue composed of cells, argyrophil reticular fibers, and fetal blood vessels. The surface of the stroma upon which the syncytium and Langhans cells rest is condensed into a basement membrane assumed to be composed of argyrophilic reticular fibers and ground substance. This membrane stains deeply with connective tissue stains, such as Mallory's and Heidenhain's azan (Fig. 15.9). It is stained also with varying intensity by the PAS reagents (Figs. 15.29, 15.32 and 15.54), the reaction probably being attributable to a ground substance consisting of mucopolysaccharide or glycoprotein which surrounds the reticular collagenous fibers. The membrane is quite variable in intensity of staining and definition, being best differentiated at the distal ends of the growing secondary villi (Fig. 15.32). In the latter region the basement membrane also exhibits metachromatic staining with toluidin blue (Wislocki and Dempsey, 1948), a response interpreted as indicating the presence of acid mucopolysaccharide.

It is of interest that the basement membrane described here, as being rich in mucopolysaccharide, is also a site in which iron (Fig. 15.8) (Hofbauer, 1905; Zancla, 1912; Wislocki and Dempsey, 1946c), calcium (Schimig, 1929; Wislocki and Dempsey, 1946c ) , and variable amounts of alkaline phosphatase (Wislocki and Dempsey, 1946c) are concentrated. It should be noted that in the early months of gestation these substances are demonstrable principally toward the growing tips of the villi, a finding which suggests possible functional differences in various regions.

The surface of contact between the Langhans cells, the syncytium, and the stroma also merits discussion with reference to the basement membrane. With connective tissue stains, such as Mallory's, each of the clear, chromophobic Langhans cells seems to be enclosed by a deeply stained surface film or membrane which seems to be continuous with the subjacent reticular basement membrane (Fig. 15.9). However, unlike the latter, these encapsulating membranes of the Langhans cells are neither argyrophilic nor stained by the PAS reagents (Fig. 15.54). Grosser (1925a, b) concluded that the "capsules" represent a secretion liberated by the Langhans cells, which he designated as "fibrinoid" (Wislocki and Bennett, 1943). Electron micrographs of early and late human placentas demonstrate that these cells are not separated from the syncytium or encapsulated by a collagenous material (Wislocki and Dempsey, 1955). However, narrow clefts often occur between the plasma membranes of the .syncytial and Langhans cells and also between two contiguous Langhans cells.

The mesenchymal cells or fibroblasts of the chorionic villi contain a moderate quantity of glycogen in the first 2 months of gestation, but this substance declines thereafter. They possess numerous mitochondria (Fig. 15.26) and also give faint cytoi)la.smic reactions for sulfhydryl groups (Figs. 15.33 and 15.35), esterase (Fig. 15.36), and succinic dehydrogenase (Fig. 15.38). Gersh and Catchpole (1949) described glycoprotein granules in the mesenchymal cells, which increase in number up to 4 months of gestation, after which they decline. To these cells they ascribed the role of producing the ground substance of the villi. The endoplasmic reticulum is conspicuous in these cells, being highly branched and irregular in shape, and its matrix has unusually high density (Wislocki and Dempsey, 1955) .

Besides mesenchymal cells, the stroma contains peculiar, large, predominantly round, vacuolated cells designated as Hofbauer cells (Fig. 15.13). They are numerous in human villi in the first months of gestation but diminish in number afterwards. In the first 6 weeks of gestation these cells contain some glycogen which is located between the cytoplasmic vacuoles. On removal of the glycogen they exhibit a residual PAS

reaction in the cytoplasm surrounding the nucleus and vacuoles (Fig. 15.57). They contain a moderate number of mitochondria (Fig. 15.26), and succinic dehydrogenase activity has been demonstrated in these cells. They do not exhibit alkaline phosI)hatase activity, but give a moderately strong cytoplasmic reaction for acid phosphatase. Hofbauer (1905) described osmicated droplets in some of them. Lipids occur in finely granular form, when demonstrated by Sudan black. Ashbel and Hertig (1952), utilizing the Ashbel-Seligman reaction, reported that a carbonyl reaction occurs in some of the Hofbauer cells. Wislocki and Bennett (1943, Fig. 16) observed them to be filled occasionally with vacuoles or droplets deeply stained with iron hematoxylin.

The nature of the Hofbauer cells is uncertain. Some investigators have regarded them as degenerating cells (Mall and Meyer, 1921), whereas Lewis (1924) considered them to be macrophages because of the affinity of their cytoplasm for neutral red. The latter opinion also receives support from their observed increase in numbers in syphilitic infections (Hofbauer, 1925). Their histochemical properties enumerated above, particularly the presence of a positive PAS reaction after removal of glycogen, of finely dispersed lipids, and of acid phosl)hatase activity, are also compatible with similar reactions encountered in macrophages (Leblond, 1950; Doyle, 1950; Weiss and Fawcett, 1953j. The Hofbauer cells differentiate presumably from the mesenchymal cells of the villi, altiiough an origin from the Langhans cells has also been suggested; the latter, although unproven, is a distinct possibility. It seems quite unlikely, as some have thought, that they originate from either erythroblasts or endothelial cells (Mall and Meyer, 1921).

4- Age Changes in the Chorionic Villi

The chorionic y\\\\ undei-go changes as the placenta ages. They become more branched, numerous, and slender. The Langhans cells diminish in number and assume a flattened shape, while the syncytium becomes thinned out over the distended sinusoidal capillaries. The stroma becomes less cellular, denser, and more fibrous. The

HISTOCHEMISTRY OF PLACENTA

899

basement membrane iii)on which the trophobhist rests becomes increasingly thick and continuous; it contains colhigenous fibers which are strongly argyrojihilic and a ground substance which stains intensely with PAS reagents (r/. Figs. 15.29 and 15.50).

As the human choi-ionic villi age, the cytoplasmic basophilia of the syncytium, revealed by staining with methylene blue, undergoes a steady decline, whereas the affinity for acid dyes, as shown l)y the curve of staining with orange G, increases sharply up to the middle of pregnancy and then continues at a constant level until full term (Singer and Wislocki, 1948). Thus, as ribonucleoprotein undergoes a gradual decline, the basic protein component rises.

As ribonucleoprotein diminishes, alkaline phosphatase, which is minimal in amount in the first 2 months of gestation, increases tremendously, becoming maximal in the last trimester (c/. Figs. 15.4, 15.41, and 15.42). This reaction is confined mainly to the syncytium, especially to its outer border where it is extremely intense. It is apparent that the amounts of ribonucleoi)rotein and alkaline phosphatase in the syncytium arc inversely related, the former being seemingly peculiar to the period of rapid growth, wdiereas the latter is associated with the phases of maturity and aging of the placenta. Acid phosphatase is also i)resent in the cytoplasm of the syncytium, although it does not increase to the degree to which alkaline phosphatase does (c/. Figs. 15.39 and 15.40).

Glycogen is not detectable in the syncytium at full term, but a saliva-resistant PAS reaction is evident in the cytoplasm of its outer border (Fig. 15.50) (Wislocki, 1950). This marginal staining occurs in minute, stubble-like microvilli which have replaced the luxuriant brush border of the first half of pregnancy. This change in surface area is clearly evident in both light and electron microscopes.

With age the stroma of the villi has become more fibrous and less cellular, the fibci's being markedly argyrophilic (Wislocki and Bennett, 1943, Fig. 20). Associated with the fibers, there is a ground substance which is strongly periodic acid-Schiff positive and

is contlensed to form a conspicuous basement membrane upon which the syncytium rests, besides forming sheaths which enclose the sinusoidal fetal capillaries (Fig. 15.50).

The lipid droplets present in the syncytium in the first trimester (Fig. 15.13) become reduced in size and relative number (Wislocki and Bennett, 1943, Figs. 9, 10, 11, 12, and 21). Nevertheless, numerous, minute, sudanojihilic droplets are still demonstrable at full term (Wislocki and Bennett, 1943, Figs. 14 and 21), and these also give a positive Ashbel-Seligman reaction for carbonyl groups (Fig. 15.19), (Wislocki, 1952). Besides the lipid droplets which react with the Ashbel-Seligman reagents, the entire syncytium exhibits a diffuse reaction (Fig. 15.19). In contrast to the above findings, Ashbel and Hertig (1952) did not observe a carbonyl reaction in 9 placentas examined at term, although Ashbel and Seligman (1949) reported a sparse reaction in 1 case at term.

The syncytium at term exhibits a large number of mitochondria (Wislocki and Bennett, 1943, Fig. 19) and also gives a positive Bakei-'s acid hematein reaction for phospholii)ids (Fig. 15.24).

The consjMcuous sinusoidal capillaries of the chorionic villi in the last trimester become closely pi'csscd against the syncytium in many i)laces. The latter becomes stretched over the capillaries to form a thin membrane. These thinned out meml)ranous areas (Figs. 15.14, 15.24, and 15.50) (Wislocki and Bennett, 1943, Plates 5 and 6) have been termed epithelial plates" and have been equated with Bowman's capsule of renal glomeruli (Bremer, 1916). It has been suj^posed that through them at this period the most active transfer of substances occurs. The thinnest areas of the hemochoi'ial jilacental barrier, separating the maternal and fetal blood streams, are composed of the following layers: (1) a thin lamina of syncytium which rests on a thick basement membrane, (2) a connective tissue space which contains ground substance, and collagenous fibers, and (3) the wall of the sinusoidal capillary composed of another basement membi-ane which is lined internally by endothelium (Wislocki and Dempsey, 1955). The cytochemical and structural comjilexity of the syncytium makes it more

900

SPERM, OVA, AND PREGNANCY

comparable with the epithelia of the proximal convoluted tubule of the kidney and of the small intestine than with the pulmonary alveolar lining or the glomerular membrane.

Degenerative age changes that occur in the human placenta have been variously described. Tenny ( 1936a, b) and Tenny and Parker (1940) reported hyaline degeneration of the syncytium clothing the terminal villi of aging placentas, a change which according to them occurs to a greater degree, as well as prematurely, in toxemia, preeclampsia, and eclampsia. Wislocki and Dempsey (1946a) examined the placentas from two cases diagnosed respectively as severe preeclampsia and eclampsia. Both cases were atypical, because in each the condition was present at 4.5 months of gestation. Thinning of the syncytium with some nuclear deterioration was observed, as well as a premature decrease in basophilia associated with early accumulation of alkaline and acid phosphatases and an increased acidophilia. Some of the smallest villi showed hyaline necrosis of both syn

cytium and stroma. These changes were ascribed to premature aging of the chorionic villi. Subsequently, however, in a further series of placentas from cases of toxemia in the last trimester of gestation (unpublished data ) , changes from normal, based on loss of basophilia and an increase in phosphatases, were not apparent.

Tenney and Parker (1940) observed that a variable number of the terminal placental villi undergo hyaline degeneration in both normal and toxemic pregnancies. Wislocki and Dempsey ( 1948) noticed that the stroma of degenerating villi develops intense metachromasia (Figs. 15.51 and 15.57), in contrast to the ground substance of normal villi, which does not exhibit metachromatic staining, except as noted above in the basement membrane at the growing tips of secondary villi. The metachromatic transformation of the stroma of fibrous, degenerating villi is probably attributable to the formation and accumulation of an acid mucopolysaccharide. This development, it seems probable, is the equivalent of, or allied to, changes described in some

Plate 15. VI

Fig. 15.24. Human placental villus at full term. Frozen section stained by Baker's acid hematein method for phospholipids following fi.xation in formalin-calcium-chloride. Observe the positive reaction in the syncytium covering tlie villus. The reaction coincides with the presence of mitochondria revealed in similar distribution by appropriate methods (Wislocki and Bennett, 1943, Fig. 19). X 10 ocular; X 60 objective.

Fig. 15.25. Cells of the basal plate (cytotrophoblastic shell) of a human placenta of 3'/2 months gestation. Frozen section stained by Baker's acid hematein method for phospholipids. Observe the cytotrophoblasts which contain variable amounts of reactive material; the latter coincides with mitochondria demonstrable by other methods. X 10 ocular; X 60 objective.

Fig. 15.26. A portion of a human placental villus at 3'/2 months of gestation. Frozen section stained by Baker's acid hematein method for phospholipids. Notice the dense concentration of reactive material in the cytoplasm of the syncytial trophoblast. The distribution corresponds to similarly abundant mitochondria demonstrable by other means (Wislocki and Bennett, 1943, Figs. 15.13 and 15.14). Observe also the positive Baker reaction in the cytoplasm of the Langhans cells beneath the syncytium as well as in a group of Hofbauer cells in the stroma of the villus. X 10 ocular; X 60 objective.

Fig. 15.27. The junction of the trophoblastic shell (upper half) and the decidua (lower half) in the placenta of a monkey of 4 weeks gestation. Section stained by Pap's method for reticulum. Bodian's fixative No. 2. Observe that the reticular fibers of the decidually transformed endometrium cease abruptly at the border of the trophoblastic shell. As a result of this, the boundary between the decidua and the trophoblastic shell is sharply demarcated. The metachromatic ground substance of the decidua (c/. Fig. 15.58) also ceases abruptly at the line of junction between the decidua and the fetal elements. X 800. (Wislocki and Bennett, 1943.)

Fig. 15.28. A chorionic fold of the placenta of a pig (fetal crown to rump length, 64 mm.), illustrating two sinusoidal fetal capillaries (surrounded by black-stained reticular fibers), penetrating the chorionic epithelium and creating thin epithelial surface plate (arrows). Compare with Figures 15.23 and 15.64 which illustrate collectively the thinness of the placental barrier in the pig's epitheliochorial placenta. Pap's stain for reticulum; Bouin's fixative. X 10 ocular; X 90 objective. (Wislocki and Dempsey, 1946b.)

HISTOCHEMISTRY OF PLACENTA

901

mj7^

Plate 15. VI

902

SPERM, OVA, AND PREGNANCY

pathologic conditions as "mucoid degeneration." This placental change occurs in villi as early as the sixth week of gestation and is especially evident in the regressing villi of the chorion laeve.

Ashbel and Hertig (1952), using the carbonyl reaction of Ashbel and Seligman, have reported that otherwise normal-appearing placentas from 6 cases of toxemia of pregnancy with gestational ages of 34 to 39 weeks contained increased amounts of ketosteroid, indicating a distinct metabolic abnormality." On the basis of the belief that there is an increased degeneration of the syncytium in pre-eclampsia, Smith and Smith (1948) postulated a marked decline in the production and secretion of ketosteroids by this tissue. The findings of Ashbel and Hertig would not seem to be in harmony with the supposition of the Smiths. However, one might interpret the increase in carbonyl compounds observed by them as possibly representing increased storage associated with diminished secretion of ketosteroids. The observations of Deane, Shaw and Greep (1948) on the cells of the adrenal cortex provided the basis for this speculation ; they found that on stimulation of the adrenal cortex the lipid droi)lets of the cortical cells diminished in size, whereas on gradual reduction of normal stimulation the droplets first enlarged but subsequently diminished with extreme shrinkage of the cells. The extent and direction of the changes depended on the degree and duration of stimulation or its cessation. It should be recalled in this connection that Bienenfeld (19121 reported an unusually high content of lipid in the placenta in eclampsia.

The main difficulty in relating eclampsia and the toxemias of pregnancy to premature and excessive age changes in the placenta is that the alterations differ from those characterizing normal placental aging in degree only, there being no really distinctive qualitative or absolute quantitative histopathologic changes that can be readily ascertained. Inasmuch as the topography of the placental villi and cotyledons is complex and variable, and many of the age changes are spottily disseminated, the human placenta presents grave obstacles to adequate histologic sampling for the

purpose of establishing relative degrees of age changes, whether they be normal or pathologic. Methods depending on the use of frozen sections combined with drastic histochemical procedures are especially unsuitable. The suggestion is offered that the placenta should be investigated for possilDle structural and functional differences with relationship to the various segments or divisions of the villous tree. The possibility that there are differences which characterize various portions of the villi is suggested by several of the histochemical findings reported in the previous pages. There is also the possibility that the structure and function of different segments of the villi, in a regional pattern from the basal plate to the surface of the placenta, may be related to the direction and manner of flow of the maternal blood in the intervillous space. With respect to this, although investigations of the uteroplacental circulation in man (Spanner, 1935b) and rhesus monkey (Ramsey, 1949, 1954) have slied much light on the uterine decidual vessels, practically nothing is known about the character and flow of the maternal blood in the intervillous space in relation to the villous trees. Moreover, because definitive information on the topography of the placental villi is lacking, previous descriptions of the pattern of the fetal villous circulation have not seemed particularly helpful (Spanner, 1935b; Boe, 1953). When the pattern of the fetal villi is worked out and correlated with the pathways of both the fetal and maternal blood streams, differing functional segments of the villi may become apparent, thus permitting deeper insight into questions of placental exchange of metabolites and aging of the placenta, both normally and in the toxemias.

C. PERIPHERAL CYTOTROPHOBLAST

The peripheral cytotrophoblast of the first trimester of gestation comprises the trophoblastic cell columns, cell islands, and the trophoblastic shell (Figs. 15.2 and 15.3). A detailed account of the differentiation and growth of these structures in the rhesus monkey was given by Wislocki and Streeter (1938). Eventually the tropho

HISTOCHEMISTRY OF PLACENTA

903

blastic cell columns (primary villi) are converted into vascularized secondary and tertiary villi covered by Langhans cells and syncytium. However, the trophoblastic shell persists and, through the expansion and growth of the placental cotyledons, is transformed into a basal plate contiguous to the decidua basalis and placental septa forming the boundaries between the cotyledons (Fig. 15.43). The septa placentae consist of masses of cells associated with a ground substance termed "fibrinoid" by Grosser (1925a, b), which is distinguishable histologically from actual fibrin intermingled with it by the fact that with Mallory's connective stain, as well as with azan, it is colored blue, whereas the fibrin stains red.

Contrary to earlier opinion, the septa placentae are now regarded as being fetal in origin and the component cells as trophoblasts (Spanner, 1935a; Stieve, 1940; Stieve and von der Heide, 1941). On the other hand, the nature and origin of the elements of the basal plate have not been so well clarified. Grosser (1948) , for example, states that, in the mature placenta, decidua constitutes the foundation of the basal plate, but that it contains remnants of the "junctional" or "penetration" zone and that "the fetal side is clothed by trophoblastic cell derivatives in various stages of regression." However, more recent work has shown that the basal plate is composed mainly of functional trophoblasts (Wislocki, 1951). The so-called cell islands which preponderate in the first half of gestation (Stieve and von der Heide, 1941) also consist of clusters of trophoblasts. The cells in these various sites, from the beginning until the end of gestation, will be referred to in the present chapter as peripheral trophoblasts and the sum total of them as the peripheral cytotrophoblast. It has been demonstrated (Wislocki, 1951) that the peripheral trophoblasts are chromophilic and differ in several important cytochemical respects from the chromophobic Langhans cells of the secondary and tertiary chorionic villi. Furthermore, whereas the Langhans cells become inconspicuous and diminish by the end of gestation, the peripheral trophoblasts in the placental septa and basal plate persist in large numbers until term (Wislocki, 1951).

1. Trophoblasts Forming the Cell Columns and Trophoblastic Shell

At the junction of the primary villi with the stroma-containing secondary villi during the first 4 weeks of gestation the trophoblastic cells are large and chromophobic, resembling Langhans cells. The proximal ends of the cell columns constitute a germinal bed of cells exhibiting numerous mitoses. It is from these cells that the peripheral trophoblasts apparently arise. Moving outward from the germinal region, along the columns into the shell, the trophoblastic cells change their character. In the first months of gestation they are laden with glycogen as revealed by the PAS method (Figs. 15.31 and 15.32). By the end of the first month, however, a type of cell which is characterized by cytoplasmic basophilia and lower glycogen content has arisen and these cells increase in number. When fully developed, they contain clumps of strongly basophilic cytoplasmic material, reminiscent of the Nissl substance of neurones ( Fig. 15.11), and the staining of the basophilic substance is abolished after treatment of the sections with ribonuclease (Fig. 15.12), a result indicating that the basophilic substance contains ribonucleoprotein (Dempsey and Wislocki, 1945). The staining and nature of this basophilic substance have been confirmed by Ortmann (1949). The cytoplasm of these trophoblasts also stains moderately strongly with acid dyes (Figs. 15.46 and 15.47). When stained with the PAS reagents following exposure to saliva, many of the trophoblasts exhibit delicate red stippling against a diffuse reddish background (Figs. 15.49 and 15.55) (Wislocki, 1950), a staining reaction indicative of the presence of a mucopolysaccharide or glycoprotein. The cytoplasm does not stain metachromatically with toluidin blue, a finding which indicates the absence of strongly acid mucopolysaccharides. The cells do not contain triglycerides or lipids of the type occurring in the syncytium. However, sudan black and Baker's method for phospholipids differentiate minute granules in variable numbers in the cytotrophoblasts (Figs. 15.25 and 15.48). These particles correspond in size and number to mitochondria revealed by other methods. Succinic dehydro

904

SPERM, OVA, AND PREGNANCY

genase activity is demonstrable by the formation of blue formazan crystals in the cytoplasm (Fig. 15.38). The cytotrophoblasts contain little acid and alkaline phosphatase (Dempsey and Wislocki, 1947).

From the observations cited above, it is apparent that the peripheral trophoblasts differ cytologically in several significant respects from the Langhans cells of the chorionic villi (Wislocki, Dempsey and Fawcett, 1948). The latter are chromophobic, whereas the peripheral trophoblasts are distinctly chromophilic, exhibiting marked cytoplasmic basophilia and containing a diastase-resistant polysaccharide.

2. Ground Substance of the Trophoblastic Cell Columns and Shell

The nature of the ground substance (Figs. 15.12, 15.44, 15.45, 15.46, 15.47, 15.48, 15.49 and 15.55) lying between the peripheral trophoblasts has not been clearly established. Grosser (1925a, b) designated it "fibrinoid" and regarded it as a secretion of the trophoblastic cells which becomes variously admixed with fibrin of maternal origin. He observed that the fibrinoid substance stained blue with Mallory's connective tissue stain in contrast to the fibrin which was stained red. Spanner (1935a) concurred, stressing that the ground substance surrounding the trophoblasts does not contain a collagenous reticular network, in this respect differing from the matrix of the maternal decidua basalis which possesses well defined reticular fibers surrounding its cells (Fig. 15.27). Wislocki and Bennett ( 1943) concluded that the ground substance of the trophoblastic shell and cell columns is probably a mixture of maternal and fetal substances. They pointed out that with other triacid stains, such as Heidenhain's azan and Masson's stain, fibrin is distinguishable from the component which Grosser termed "fibrinoid." The ground substance is also stained by basic dyes (Fig. 15.11), but, unlike the cytoplasmic basophilia of the trophoblasts, that of the ground substance is not influenced by exposure to ribonuclease (Fig. 15.12). However, Singer and Wislocki (1948) were unable to distinguish two separate components on the basis of the affinity of the ground substance for methylene

blue and orange G as measured under conditions of controlled pH staining; under these conditions the reactions to both dyes were found to be similar to that of purified fibrin. On the other hand, PAS reagents stain the fibrin an intense scarlet (Fig. 15.55), whereas the "fibrinoid" component assumes a much paler color (Fig. 15.32), a result which again indicates a distinct difference between them. Acid and alkaline phosphatases are present only in traces. The Turnbull blue reaction for iron occurs intensely in the ground substance (Dempsey and Wislocki, 1944).

It is noteworthy that the ground substance of the peripheral trophoblast is neither metachromatic (Fig. 15.58) nor provided with argyrophilic reticular fibers (Fig. 15.27) (Wislocki and Dempsey, 1948). Thus, although fibrin of maternal origin appears to gain ready access to the ground substance of the cytotrophoblast, the metachromatic component of the decidua as such does not appear to diffuse into the troj^hoblastic shell.

3. Cytotrophoblast of the Basal Plate and Septa Placentae at Term

It is generally believed that the peripheral trophoblast undergoes regression and degeneration in the second half of gestation, no longer playing any significant functional role. Spanner (1935a) states that the septa consist of fibrinoid, containing lacunae from which the trophoblastic cells gradually disappear. Baker, Hook and Severinghaus (1944) maintained that both cell columns and cell islands are characteristic of early pregnancy, the former persisting as late as 3.5 months. Furthermore, according to them, some cells continue to show indications of secretory activity, but for the most part they degenerate into fibrinoid. Similarly, Grosser (1948) remarked that the basal plate in the ripe placenta is clothed on its fetal side with trophoblastic cells in various states of regression.

The results obtained by the use of histochemical methods differ in two major respects from the views expressed above. First, the occurrence of cytoplasmic basophilia, of a positive PAS reaction (Figs. 15.49 and 15.55), and of numerous mito

HISTOCHEMISTRY OF PLACENTA

905

chondria in the placental septa and basal plate at full term indicate the presence of large numbers of viable and functionally active trophoblastic cells (Wislocki, 1951). Very few of the trophoblasts appear to be undergoing degeneration.

Secondly, in normal human placentas delivered at full term, the basal plate is composed predominantly of masses of cells which are identifiable as trophoblasts by cytologic and histochemical means (Wislocki, 1951). Only a few maternal decidual cells which differ cytologically from trophoblasts are attached irregularly to the maternal surface of the expelled placenta. The ground substance of the basal plate gives an intense PAS reaction which seems to be mainly attributable to fibrin (Figs. 15.49, 15.45, and 15.55) , and contains no acid phosphatase (Fig. 15.40) and only traces of alkaline phosphatase (Fig. 15.42). In these respects it differs completely from the matrix of the decidua. However, after abnormal implantation, culminating in the condition termed placenta accreta, the basal plate and septa {ilacenta are modified by the unusually deep penetration of the trophoblast into the uterine wall, resulting in widespread intercrescence of the fetal and maternal tissues (Irving and Hertig, 1937) with obliteration of the usually well defined demarcation of the basal trophoblast from the decidua.

4. Cytologic Comparisons of the Cytotrophoblasts with Decidual Cells

The decidua, as revealed by histochemical means, is composed mainly of rather large cells exhibiting many reactions, some of which distinguish them clearly from the peripheral cytotrophoblasts. In the first months of gestation both the trophoblasts and the decidual cells contain large amounts of glycogen which make it difficult to tell them apart; as gestation advances glycogen diminishes in both cell types, although a considerable quantity persists in the decidual cells until term. On the other hand, the cytoplasm of the decidual cells does not have the strong cytoplasmic basophilia which characterizes the peripheral cytotrophoblasts. Furthermore, unlike the trophoblasts, most of the decidual cells contain droplets of neutral fat and give a strong

reaction for acid phosphatase (Fig. 15.53) (Wislocki and Dempsey, 1948). Also the stippling observed in the trophoblasts following staining with PAS reagents after exposure to saliva (Figs. 15.49 and 15.55) is not seen in the decidual cells (Fig. 15.52) ; instead they are stained a diffuse pink. Thus certain features sharply differentiate the two types of cells.

In contrast to the ground substance of the cytotrophoblastic cell columns, the matrix of the decidua is characterized by the presence of an argyrophilic collagenous reticulum surrounding the decidual cells (Fig. 15.27). The matrix also contains an amorphous ground substance which stains metachromatically with toluidin blue (Fig. 15.58) and is quite deeply stained by PAS reagents (Fig. 15.52). The metachromasia indicates the presence in the matrix of an acid mucopolysaccharide. These reactions characterize the matrix throughout the entire period of gestation, as has been demonstrated on pieces of decidua vera and basalis obtained at cesarian removal of human placentas at full term (Wislocki, 1953).

The basal plate of the delivered human placenta exhibits, as a rule, a sharp line of demarcation between a wide zone of trophoblasts and fibrin and an incomplete, irregularly narrow lamina of decidua attached to its outer surface. The decidual cells of this narrow strip or border, which represents the "junctional" or "penetration" zone, differ from those in the general bulk of the decidua in that the cells are markedly degenerated. This zone gives a strong enzymatic reaction for acid phosphatase (Figs. 15.40 and 15.53) and a faint, irregular, nonenzymatic reaction by the method for alkaline phosphatase, the latter attributable mainly to the presence of calcium salts in this region (Fig. 15.42) (Dempsey and Wislocki, 1946).

5. Cytolytic and Proteolytic Activities 0/ Peripheral Trophoblasts

In the first weeks of gestation the cytotrophoblast of the trophoblastic shell seems to produce proteolytic and cytolytic substances capable of attacking the endometrium (Wislocki and Bennett, 1943; Wislocki, Dempsey and Fawcett, 1948). In the early

906

SPERM, OVA, AND PREGNANCY

stages of normal gestation in primates, hemorrhage in the decidua coincides with the erosion of the uterine mucosa by the advancing trophoblast. However, the trophoblast elaborates a substance which initiates changes in the decidual tissue even before the ovum has become attached. In the rhesus monkey, evidence of such a chemical factor is seen in the fact that the epithelium at the secondary implantation site begins to proliferate before actual erosion of the uterine surface has taken place (Wislocki and Streeter, 1938) . Similarly, in a previllous human ovum of 11 days' ovulation age, an area of congestion and hemorrhage was found on the opposite endometrial wall which had merely been in close proximity to the implantation site (Hertig and Rock, 1941).

The histologic appearances at the margin of the growing trophoblastic shell of the human placenta in the early months of gestation suggest that the cells and matrix of the decidua are attacked and slowly destroyed by the action of the advancing, growing cytotrophoblast. In sections which have been impregnated by silver, the dissolution of the fibers of the reticulum can be observed. Immediately adjacent to the border of the trophoblast, the fibers become broken up and the individual bits dissolve apparently in the outermost part of the matrix of the trophoblastic shell (Wislocki and Bennett, 1943, Plate 9). Similar fragmentation of collagen fibers in the vicinity of the

trophoblast has been noted in the placenta of rodents (Wislocki, Deane and Dempsey, 1946, Plate 10) and cat (Wislocki and Dempsey, 1946a, Fig. 5). The metachromatic ground substance between the decidual cells, close to the cytotrophoblast, is also destroyed.

Similar proteolytic activity has been demonstrated experimentally in the presence of fertilized mouse ova transplanted to various extra-uterine sites including the anterior chamber of the eye (Runner, 1947; Fawcett, Wislocki and Waldo, 1947). In the eye of a mouse containing proliferating ova, leakage of blood from engorged vessels in the iris and cornea took place before a blastocyst had actually become attached to the wall of the anterior chamber. It is noteworthy also that blood began to accumulate behind the iris at the same time that it appeared in the anterior chamber. Inasmuch as the vessels on the back of the iris and ciliary body were not in contact with the blastocyst and hence had not presumably been disrupted by actual invasion, the most satisfactory way to account for the bleeding in the posterior chamber was to attribute it to some chemical released by the trophoblast. Although the trophoblast invades maternal vessels in later stages of implantation, direct observation and study of histologic sections of ova transplanted to the eye suggested that the interstitial hemorhage and edema in the early hours of nidation were the result of diffuse damage to

Plate 15.VII

Fig. 15.29. Human placental villi at 2I/2 months of gestation, stained by the periodic acidSchilT (PAS) procedure. Section exposed to saliva before staining. Zenker's acetic acid fixative. Observe the positive PAS reaction in the outer zone of the syncytium as well as the variable reaction of the ba.sement membrane on which the trophoblast rests. Note also the reaction in the cells of the stroma. Compare with Figure 15.33. X 150.

Fig. 15.30. Human placental villi at 31/2 months of gestation stained by Baker's acid hematein method for phospholipids. Formalin-calcium-chloride fixative. Observe the intense reaction of the syncytial trophoblast clothing the villi, as well as the staining of the stromal cells, particularly of large, vacuolated Langhans cells. Compare with Figure 15.26, which shows more details. X 240.

Figs. 15.31 and 15.32. Sections illustrating the tip of a secondary chorionic villus attached to the trophoblastic shell (or basal plate) of a human placenta at 2 months of gestation. PAS stain after Rossman's fixative. In Figure 15.31, glycogen is abundantly revealed in the cytotrophoblasts of the basal plate (left side of figure). In Figure 15.32, the section was immersed in saliva before staining it, to remove glycogen; as a result the trophoblasts (left side of figure) are now only very faintly stained, the outer zone of the syncytial trophoblast is moderately stained and there is an intense reaction visible in the basement membrane between the stroma of the secondary villus and the cytotrophoblastic cell column. These residual reactions are attributable to carbohydrates (glycoproteins, mucopolysaccharides) other than glycogen. X 300.

HISTOCHEMISTRY OF PLACENTA

907

PUTE 15.VU

SPERM, OVA, AND PREGNANCY

vessels by cytolytic substances emanating from the trophoblast. Further evidence of a cytolytic factor was demonstrated by the observation that ova placed in the eye were capable of developing in close proximity, but only until the most precocious one among them began to implant. Thereafter, the others quickly degenerated. In connection with the question of cytolytic substances produced by the placenta, it should be recalled that bits of human chorionic villi obtained from placentas of the first months of pregnancy and grown on plasma clots liquefied the medium (Grafenberg, 1909a, b; Friedheim, 1929; Caffier, 19291.

IV. The Structure of the Placental Barrier

Two concepts have dominated the subject of placental physiology. The first of these states that, as gestation advances and the placenta ages in a given species of mammal, the placental barrier becomes progressively more permeable to physiologic exchange between mother and fetus. The second concept proposes that the placentas of mammals can be arranged in an ascending order or phylogenetic series with reference to the relative facility and rapidity with which metabolites traverse them.

The placental barrier can be defined for Mammalia as "an apposition or fusion of the fetal membranes to the uterine mucosa for physiologic exchange" (Mossman, 1937). The membranes involved are the chorion, allantois, and yolk sac, w^hich in various relations to one another and to the uterine tissues give rise to a variety of placental structures, the nature of which will be discussed below.

A. grosser's classification

The relative permeability of the placenta in eutherian mammals, with reference to both ontogeny and phylogeny, has been generally related to four morphologic placental types defined by Grosser (1909). According to this doctrine, the chorioallantoic placentas of mammals can be arranged in an ascending order from the most primitive to the most advanced on the basis of the successive disappearance of 3 out of 6 layers which intervene between the maternal and fetal blood streams. The dis

appearance of the maternal layers is attributed to the invasive and aggressive properties of the trophoblast of the chorion. In the most primitive placentas the layers comprise (1) the uterine vascular endothelium, (2) the uterine stroma, (3) the uterine epithelium, (4) the fetal trophoblast, (5) the fetal stroma, and (6) the fetal capillary endothelium. When these layers are all present, they form the so-called epitheliochorial type of placenta encountered in some ungulates (e.g., pig, mare), Cetacea, and lemurs. Next, the uterine epithelium disappears through the invasive activity of the trophoblast, with the formation of the syndesmochorial type of placenta of other ungulates {e.g., cow, sheep). Then with the loss of the maternal connective tissue, the endotheliochoricd type of placenta of carnivores (dog, cat), sloths (Bradypodidae), Tupaiidae, some Insectivora (shrew% mole), and some bats (Chiroptera) arises. Finally, with the ultimate loss of the maternal endothelium, the hemochorial type of placenta of rodents, Tarsiidae, monkeys, anthroi:)oid apes, and man is formed. This series has been widely accepted as having phylogenetic significance, in that it is sujiposed to begin with a primitive, six-layered placental barrier which is the least permeable, and to l^rogress, by a successive reduction of the three maternal layers and a gradual diminution in width of the remaining fetal layers, to the most advanced evolutionary type in which transmission is most rapid and complete.

As an extension to Grosser's classification, Mossman (1926, 1937) sought to demonstrate that in lagomorphs and higher rodents, especially rabbit, rat, and guinea pig toward the end of gestation, the trophoblastic syncytium is normally, and very generally, lost, so that the placental membrane consists merely of fetal capillaries composed of endothelium. For the rabbit's placenta, he stated that after the 22nd day the syncytium disappears so that the placental barrier becomes reduced to a layer of "endothelium and a very thin Plasmodium, the latter entirely absent in many places." This he designated as a hemoendothelial type of placenta, and it has been accepted as the most advanced stage of the

HISTOCHEMISTRY OF PLACENTA

909

series both morphologically and functionally.

In some groups of eutherian mammals (carnivores, some insectivores ) a further provision for placental transfer exists in the paraplacental and central hematomas. These structures, illustrated by the paraplacental green" and "brown" borders in dog and cat, consist of extensive extravasations of maternal blood between the chorion and endometrium. The extravasated blood is absorbed by the chorionic epithelium, with the result that the iron of the j^hagocytized red blood cells becomes available to the fetus. These paraplacental structures, which are epitheliochorial according to Grosser's classification but which occur mainly in association with endotheliochorial placentas, represent another important but poorly studied route of nutritive exchange between mother and fetus which has not been adequately evaluated with reference to Grosser's doctrine.

Hemotrophe is the name given to the nutritive materials absorbed by the placenta or fetal membranes directly from the circulating maternal blood stream. Histotrophe, on the contrary, refers to secretions and degradation products of the endometrium, as well as extravasated maternal blood, which undergo absorption. According to Grosser (1927), there is a correlation between the kind of nutriment supplied to the fetus and the degree of association between the maternal and fetal blood streams. Thus, in epitheliochorial placentas, histotrophe in the form of secretions and transudations is stated to be the almost exclusive form of nourishment. However, the higher the organization of the placenta is, the more important hemotrophic nourishment is said to become, so that it is maximal in species with hemochorial placentas. In man, and possibly the hedgehog, transmission becomes exclusively hemotrophic, so that according to Grosser, the human placenta in this respect represents a developmental end-stage. Amoroso (1952) concluded similarly that "in the hemochorial and hemo-endothelial placentas of man and higher rodents, histotrophic nutrition is insignificant after the early stages of development and nourishment of the foetus becomes possible, largely

by direct absorption from the maternal blood." In regard to man, this conclusion seems entirely warranted, but in respect to rodents it should be borne in mind that they possess, in addition to a hemochorial placenta, a well developed yolk sac placenta which engages actively throughout gestation in the absorption of transudate and secretion, representing histotrophe derived from the endometrium. This illustrates again the paradox that, in the presence of what is regarded as the most highly developed and "efficient" type of placenta, namely, the hemochorial one of the higher rodents, a complex yolk sac placenta is also present which certainly functions principally by the absorption of histotrophe, a process which is generally regarded as being most primitive.

B. ULTRASTRUCTURE OF THE CHORIO-ALLANTOIC PLACENTA

A knowledge of the ultrastructure of the placental barriers of mammals is necessary to form hypotheses for the mechanism of transport across these membranes. Although our information on ultrastructure is fragmentary now, many significant observations have already been presented.

The ])lacenta of the pig is classified as epitheliochorial since it consists of a simple apposition of the chorion to the endometrial epithelium without erosion. Observations have been made on the ultrastructure of the definitive pig placenta in which the corrugated surfaces of the chorion and uterine mucous membrane interdigitate closely (Dempsey, Wislocki and Amoroso, 1955}. Chorionic ridges fit into matching uterine depressions (Figs. 15.23, 15.28, 15.64). These macroscopic interdigitations are further extended by the "submicroscopic" interdigitation of the free surfaces of the chorionic and uterine epithelia. In the regions of the chorionic ridges, the chorionic surface is thrown into numerous microvilli which align with the uterine microvilli to form simple interdigitations. The bovine fetal-maternal junction is similarly constructed (Bjorkman and Bloom, 1957). In the chorionic fossae, the fetal-maternal junction is similarly constructed; however, in addition, there are deep, thread-like invaginations of the fetal

910

SPERM, OVA, AND PREGNANCY

plasma membranes. Thus, there is far more intimate apposition than was previously realized and also these ultrastructural devices result in a great expansion of surface area. Furthermore, the uterine space at this maternal-fetal junction is so slight that the possibility of absorption of uterine secretions through the chorionic ridges and fossae seems to be excluded. Uterine secretions are most likely absorbed through the chorionic areolae which occur in regions where the uterine lumen is patent and filled with secretion. The special chorionic epithelium of the areolae shows great complexity at its surface. In addition to possessing irregularly shaped, often bulbous, projections which extend into the uterine lumen, these cells have complicated infoldings of the plasma membrane both apically and laterally. During the latter half of gestation, the fetal capillaries burrow into the chorionic epithelium where they acquire a so-called intra-epithelial position; fetal capillaries are especially rich in the chorionic ridges. The electron microscope has revealed that two basement membranes remain interposed between the chorionic epithelium and endothelium. Further interesting observations on the pig placenta concern the possibility of secretory activity by the cells lining the chorionic fossae and the occurrence of occasional cilia in the uterine glands.

The placenta of the cat, according to Grosser's classification, is of the endotheliochorial type in which erosion of the uterine epithelium and connective tissue occurs. The placenta proper consists of a series of roughly parallel trophoblastic plates or lamellae (Figs. 15.56, 15.60, and 15.61). In the center of the lamellae, maternal blood flows through closed capillaries surrounded by a relatively thick amorphous matrix and occasional giant decidual cells of maternal origin. The syncytiotrophoblast abuts on the amorphous matrix. The cytotrophoblast, which diminishes as gestation proceeds, is situated between the syncytiotrophoblast and the fetal connective tissue. The trophoblastic lamellae are separated from each other by a layer of fetal connective tissue through which fetal capillaries course.

Some observations on the ultrastructure of the placental and paraplacental regions of the cat's chorion were reported by Dempsey and Wislocki (1956). Substances passing from the maternal blood to that of the fetus first encounter the unusually thick and basophilic maternal endothelium which typifies the capillaries of the carnivore placentas. The endothelial cytoplasm evaginates to form surface projections and appears to have a well developed endoplasmic reticulum; however, inadequate preservation of these cells does not permit further

Plate 15. VIII

All of the photographs on this plate are of a normal human placenta of 6 weeks of gestation.

Figs. 15.33, 15.34 and 15.35. The localization of sulfhydryl groups by Barnett and Seligman's method. The reaction is most intense in the trophoblastic .syncytium where it is especially pronounced at the inner and outer borders (Fig. 15.33). The Langhans cells give the least reaction, being set off as clear cells which appear to be enclosed by dark capsules (Figs. 15.33 and 15.34). The peripheral cytotrophoblasts and the stromal cells of the chorionic villi are also moderately reactive (Fig. 15.35). X 300.

Fig. 15.36. The enzymatic reaction of esterase by Barrnett and Seligman's method is confined principally to the trophoblast clothing the villi, although small scattered particles of the reaction product are also present in the stroma. As explained in the text, it is believed that the reaction seen should be attributed to cholinesterase. X 175.

Fig. 15.37. The reaction of Seligman and Rutenberg for succinic dehydrogenase. Blue crystals of diformazan are produced which are indicative of enzymatic activity. This reaction product is soluble in lipids which it stains pink, but this color has been eliminated in the picture by photographing the section through a red filter (Wratten A 25). The succinic dehydrogenase reaction indicated by blue diformazan particles is present in the syncytium and Langhans cells. Very little reaction has occurred in the cells of the stroma. X 250.

Fig. 15.38. Illustrates the combined appearance of the blue diformazan particles and the pink-stained fat droplets. Compare with Figure 15.37 in which the color of the latter has been reduced. The stained lipid is identical with the sudanophilic droplets pre.sent in the syncytium (c/. Fig. 15.13). In the lower half of this figure clear nuclei of cytotrophoblasts are barely apparent, surrounded by cj^toplasm containing diformazan crystals. X 200.

HISTOCHEMISTRY OF PLAC^ENTA

911

-■■';• ■-^'

•V>

37

36

"-^

1

3»;

t^.>

Plate IS.VIII

912

SPERM, OVA, AND PREGNANCY

evaluation of their ultrastructure. The transported substance next traverses the amorphous perivascular ground substance which varies in amount regionally, being abundant, moderate, or entirely absent. In areas in which this ground substance is lacking, the syncytium abuts on maternal endothelium. The endothelial margin of the ground substance is regular, but its opposite border, which is in contact with the syncytium, is irregular and seems to be sculptured by the syncytium. The lipid-rich syncytium sends branched processes between the cytotrophoblast to end in foot-like expansions on the trophoblastic basement membrane. Later in gestation, as the cytotrophoblasts diminish in number, these processes extend through large extracellular spaces. After crossing the two trophoblastic layers and the chorionic basement membrane, a substance then passes through a thin connective tissue space containing fibroblasts and collagen fibers. From there it next passes through the basement membrane and endothelium of ordinary fetal capillaries. Thus, despite the loss of two uterine layers, the placental barrier in the cat is structurally complex.

The paraplacental or brown border of the cat's chorion is a specialized region composed of cells rich in iron and caj^able of phagocytosing red blood cells, trypan blue, and other substances (Figs. 15.62 and 15.63). The absorptive surface of these cells j bears a striking resemblance to that of the visceral endoderm of the rodent yolk sac. The surface plasma membrane of these columnar cells evaginates to form elaborate microvilli and invaginates to form a system of canals immediately beneath the cell surface. It is assumed that pinocytotic vesicles are farmed as ingested substances are segregated in the canalicular- system. In the paraplaQehtal cells, ingested erythrocytes in various stages of breakdown were frequently observed.

The hemochorial type of placenta, in which extensive erosion of the uterine wall occurs, has been studied in the human and rodents by several investigators (Boyd and Hughes, 1954; Wislocki and Dempsey, 1955a, b; Wislocki, Weiss, Burgos and Ellis, 1957; Bargmann and Knoop, 1959; Schieb

ler and Knoop, 1959). In the "definitive" hemochorial placenta the trophoblast is bathed directly by circulating maternal blood. Observations on the ultrastructure of the human placenta (Wislocki and Dempsey, 1955a) have been incorporated with the histochemical findings in Section III of this review. The syncytium of the human placenta resembles a pinocytotic epithelium having numerous pleomorphic microvilli and containing many large vesicles which probably represent engulfed material (Wislocki and Dempsey, 1955a, Plates 1-4). Although the Langhans cells diminish in number as gestation proceeds, some flattened cytotrophoblasts remain at term interposed between the syncytium and the trophoblastic basement membrane. At term, when the placental barrier is thinnest, a maternal substance encounters the following successive layers in reaching the fetal blood: syncytium, trophoblastic basement membrane, fetal connective tissue, basement membrane, and endothelium of the fetal sinusoidal capillary.

Many significant observations have been made on the fine structure of the chorioallantoic placenta of the rat. To aid in orienting the reader, a brief description of the histology of the rat placenta follows. In the established chorio-allantoic disc of this species, three general zones are recognizable in the fetal portion: (1) a trophoblastic labyrinth which is served by both maternal and fetal blood vessels and is considered to be the principal area of transport; (2) a spongiotrophoblastic zone which partially surrounds the labyrinth and which, although it is not penetrated by fetal blood vessels, is perfused by maternal blood; and (3) a meshwork of giant cells which caps the spongy zone, is permeated by maternal blood, and forms the frontier of the fetal tissue.

The first description of the ultrastructure of the labyrinth of the rat at 15, 17, and 21 days of gestation quickly established two important points (Wislocki and Dempsey, 1955b ) . It was demonstrated that the rat and rabbit placentas are hemochorial rather than hemoendothelial as had been proposed by Mossman (1926, 1937). Furthermore, it was shown that there is no syncytial tropho

HISTOCHEMISTRY OF PLACENTA

913

blast in the labyrinth, as had been described by Grosser (1908, 1909) ; the fetal blood vessels are clothed by two or three thin layers of overlapping individual cytotrophoblasts which together constitute a laminated membrane (Wislocki and Dempsey, 1955b, Plate 2). These trophoblasts are held together by small cytoplasmic pegs which fit into depressions in adjacent cell surfaces. Wislocki and Dempsey (1955b) observed that the labyrinthine lipid droplets are located mainly in the cytoplasm of the innermost trophoblasts. Schiebler and Knoop ( 1959) reported that there is a relatively wide space between the outer and middle layers which communicates with the maternal blood but that the deeper cells are closely apposed. The latter investigators also observed that pinocytotic vesicles occur in numerous trophoblasts and claimed that two kinds of trophoblastic cells can be differentiated in the labyrinth with the electron microscope.

According to Wislocki and Dempsey, the placental barrier in the labyrinth of the rat is composed of : ( 1 ) two or three sheets of laminated cytotrophoblast; (2) the basement membrane supporting the trophoblast; and (3j the basement membrane and endothelium of the fetal capillary. Schiebler and Knoop did not see two separate basement membranes and report the occurrence of a single basement membrane between the trophoblast and fetal endothelium.

Schiebler and Knoop (1959) also presented some interesting observations on the fine structure and histochemistry of the spongiotrophoblasts and giant cells. The cytotrophoblasts of the spongy zone have an intensely basophilic cytoplasm, and this is consonant with the presence of an extensive, highly oriented endoplasmic reticulum. This luxuriant endoplasmic reticulum is comparable in its arrangement and abundance to that of the pancreatic acinar cells and of the Nissl bodies of neurones. The function of the spongiotrophoblast remains unknown, but this observation points toward a special role in protein synthesis. Padykula (1958) reported a striking increase in acid phosphatase and adenosine triphosphatase activity in this zone during the last week of gestation.

In the same report, Schiebler and Knoop offered much new information about the fetal giant cells, and their observations suggest a dynamic role for these strategically placed cells. The giant cells are contiguous with the spongiotrophoblasts, and with the aid of the electron microscope they can be differentiated into several types. The nuclei of the giant cells are invaginated in many places, and these recesses contain cytoj^lasm. In some planes of section, this morphologic arrangement gives the false impression of intranuclear inclusions, especially when the invagination contains lipid or glycogen. However, the cytoplasmic mass enclosed by the nucleus maintains its connection with the main body of cytoplasm. The cytoplasm proper is highly differentiated. It contains a great complexity and variety of vesicles and membranes, and resembles the cytoplasm of phagocytic cells in several aspects of fine structure. The surface of the giant cells presents a complicated interwoven array of microvillus-like projections to the intercellular space. This space is filled with a material which contains mucopolysaccharide, is fibrous, and appears to be continuous in some regions with Reichert's membrane. In some regions the maternal blood spaces among the giant cells are lined by a thin layer of cytoplasm which is judged to be endothelium by Schiebler and Knoop. In this location a subendothelial basement membrane seems to be lacking. Elsewhere the surfaces of the giant cells are in direct contact with the maternal blood.

The fine structure of the granular cells of the metrial gland of the pregnant rat was described by Wislocki, Weiss, Burgos and Ellis (1957). The suggestion was offered that the basic protein granules of these cells contain relaxin.

The fine structure of the hemochorial placenta of the nine-banded armadillo (Dasypus novemcinctus) was recently reported by Enders (1960).

Some generalizations may be made concerning the fine structure of the placental barriers. Certainly the absorptive trophoblasts resemble the cells of the proximal convoluted tubule of the kidney and the absorptive cells of the small intestine more

914

SPERM, OVA, AND PREGNANCY

closely than the components of the pulmonary alveolar lining or the renal glomerular membrane. The trophoblastic cells are characterized by microvilli and other surface projections which are pleomorphic and often branched. Many observations suggest that absorption by pinocytosis occurs in many types of placental cells. Whether or not erosion of the uterine wall occurs, the placental barrier is structurally complex. Along with the cellular layers, basement membranes are regularly interposed between the maternal and fetal bloodstreams. Further discussion is presented in the section on yolk sac placentation where some experimental cytologic observations have been made during the process of absorption.

C. REDUCTION IN NUMBER OF THE LAYERS OF THE CHORIO-ALLANTOIC PLACENTA

The successive elimination of the maternal layers of chorio-allantoic placentas as envisioned by Grosser's scheme has met with general acceptance. However, Wislocki and Dempscy (1946a I pointed out that the endotheliochorial type of placenta is probably nonexistent, because in carnivores and sloths the endothelial-lined maternal blood vessels are surrounded by a basement membrane and in some species, as in the cat, large decidual cells are present in the labyrinth. Consequently, in some carnivores the placenta is syndesmochorial rather than endotheliochorial.

The hemoendothelial type of placenta postulated by Mossman finds no support in recent observations by Wislocki and Dempsey (1955b) and Schiebler and Knoop (1960) with the electron microscope. These

investigators found in the chorio-allantoic placental labyrinths of rat and rabbit late in gestation a complete trophoblastic membrane, consisting of 2 or 3 layers of flattened, imbricated trophoblastic cells. The presence of these layers is not detectable with the light microscope. These findings indicate that the placentas of these species are hemochorial and not hemoendothelial.

D. REDUCTION IN WIDTH OF THE LAYERS

OF THE CHORIO-ALLANTOIC

PLACENTA

The diminution in width of the tissues separating the maternal blood channels from the fetal capillaries, as postulated in Grosser's scheme of chorio-allantoic placentas, seems to be borne out by histologic observations of both his phylogenetic series and successive ontogenetic stages. However, some have pictured the layers in a schematic way (Huggett, 1944; Arey, 1946) with no regard for their relative widths and relationships. Actually, in progressing from ei)itheliochorial to hemochorial placentas, the gradual reduction in width of the thinnest areas is not nearly as striking as the theoretic concept of the removal of successive layers implies. This is due partly to the fact that in all species the connective tissue layers at the sites of the thinnest places consist only of basement membranes. Furthermore, the capillaries in the thin areas of all animals are pressed against the adjacent epithelia and in some, for example in the sow (Figs. 15.23, 15.28 and 15.64) and many ungulates, the fetal capillaries follow intra-epithelial courses in the trophoblast (Wislocki and Dempsey, 1946b: Amoroso, 1947, 1952). In addition, in the

Plate 15. IX

Fig. 15.39. Human placental labyrinth at 4 months, showing acid phosphatase activity in the syncytium and stroma of the chorionic villi. Gomori's method, using glycerophosphate as substrate at pH 4.7. X 140. (Wislocki and Dempsey, 1948.)

Fig. 15.40. The basal plate of a human placenta at full term, showing the presence of acid phosphatase in the chorionic villi (above), its almost complete absence in the basal plate (center), and a marked reaction at the line of junction of the basal plate with the decidua (below). Gomori's method using glycerophosphate as substrate at pH 4.7. X 175.

Fig. 15.41. Human placental labyrinth at full term, showing the activity of alkaline pho.sphatase in the syncytium clothing the chorionic villi. Gomori's method using glycerophosphatase as substrate at pH 9.4. X 220.

Fig. 15.42. The basal plate of a human placenta at full term, showing an inten,se alkaline phosphata,se reaction in the chorionic villi (above), its nearly complete absence in the basal plate (center) and a slight reaction at the line of junction of the basal plate with the decidua (below). Gomori's method using glycerophosphate as substrate at pH 9.4.

HISTOCHEMISTRY OF PLACENTA

9V

• A ^<L/i: ■.,;•. "v *C

39 v:» X x-->;'

c^

" r

.^'

Plate 15. IX

916

SPERM, OVA, AND PREGNANCY

SOW, Amoroso (1952) has described the trophoblast at one period as sending actual processes between and past the maternal epithelium into the region of the underlying maternal capillaries, establishing an endotheliochorial relationship. Despite the presence of six theoretic layers in the sow, provisions exist which tend to by-pass or materially reduce the width of several of them, thus diminishing the distance between the two blood streams. Similarly, in the cat, in the last half of pregnancy, Amoroso (1952) observed "that the foetal capillaries come to lie so near the surface of the lamellae that only the thinnest laminae of syncytial trophoblast separate them from the maternal tissues."

Measurements of the width of the maternal epithelium in the sow, made by Gellhorn, Flexner and Pohl (1941), show that its height changes from 18 fj. at midgestation to 10 fi just before term. Nevertheless, in many of the thinner places between the capillaries by midgestation (Fig. 15.23 of the present study), the width of the intervening cytoplasm of the combined chorion and uterine epithelium is reduced to no more than 6 or 8 fj.. In the sheep at 100 days of gestation Barcroft (1947) reported that none of the fetal and maternal capillaries is closer than 20 ix and none is separated by more than 120 /j.. This offers no clue, however, as to what the average distance may be. It is apparent, nevertheless, from some excellent figures of the sheep's placenta submitted by Wimsatt (1950, Figs. 54 and 56) , that the numerous maternal capillaries arc about 10 to 20 /x from the fetal capillaries at 100 days and less at 133 days. In the cat at term, the distance between the two blood streams is narrowed in many places to 6 or 8 fji. In the human at term, the thinnest places vary between 3 and 6 /x in width. More extensive and careful measurements of the distances between the blood streams should be obtained in various animals at different stages of gestation, in order to provide a better basis than now exists for comparisons. In making such measurements the degree of shrinkage and separation of the layers in preparing the tissues should be carefully evaluated.

A more important consideration than the

actual diminution in width of the layers in the thinnest regions might be the apparent much larger extent of thin areas in hemochorial placentas than in other placental types. Thus, for example, although the thinnest places in the sow's placenta do not seem to differ greatly from the human in respect to their actual widths, the relative extent of the thin areas is very much greater in the latter than the former. In this respect, hemochorial placentas differ greatly from epitheliochorial ones. This consideration, although possibly inherent in Grosser 's doctrine, has never been clearly brought out and documented, but instead has been subordinated to the prevailing concepts of the phylogcnetic reduction in number and widths of the layers.

A further point of interest concerns the placentas of rodents. The physiologic advantages obtained presumably by the reduction in width of the trophoblastic membrane and the increased extent of the thin areas would seem to be offset by the functional disadvantage of the laminated arrangement of the trophoblastic cells as revealed by electron microscopy. Here, where a syncytium with only inner and outer surfaces was believed to exist, the trophoblast is laminated, so that 4- or 6-cell surfaces extend across the placental barrier. Thus, with respect to cell surfaces and cell layers forming the placental barrier, the hemochorial placentas of rodents seem to be quite as complex as epitheliochorial and syndesmochorial placentas. The most important difference between them would seem to lie in the relatively greater extent of the thin regions, rather than in any extreme reduction of the number of cell layers in the rodent's hemochorial placenta.

Some degree of cytologic and histochemical simplification is apparent in successive stages of gestation in any given species, but striking cytologic differences between the placental membranes of Grosser's phylogenetic series at equivalent stages of gestation are not very evident. Even the thinnest regions of the different types of chorio-allantoic placentas possess a far greater cytochemical complexity than the glomerular and pulmonary membranes. Several writers have postulated that toward the end of

HISTOCHEMISTRY OF PLACENTA

917

pregnancy some of the human chorionic villi lose their syncytial covering entirely; the increase in placental permeability is attributed to this structural alteration. However, it should be pointed out that the nature and degree of degeneration and loss of the syncytium in human villi have not been carefully analyzed. Moreover, it is not known whether such altered villi are functionally active or dead and functionless. However, the assumption that a fraction of the villi becomes functionless would be consonant with an observation of Flexner, Cowic, Hellman, Wilde and Vosburgh (1948) that there is a sharp terminal decline in placental permeability after the 36th week of gestation.

V. Yolk Sac Placentation

In those lower vertebrates, such as some fishes, amphibians, and reptiles which are either ovoviviparous or viviparous, the yolk sac plays the principal role as the fetal membrane subserving the transfer of metabolic materials (Amoroso, 1952) . An exception to this is encountered in some reptiles (Weekes, 1935) in which chorio-allantoic placentation occurs. In marsupials, the yolk sac is very large, whereas the allantois is always relatively small, and in only three species does the latter vascularize a placenta. In accordance with Grosser's terminology, the chorio-allantoic placenta of Perameles is "endothelio-endothelial" in character, thus differing fundamentally from the types he defined. "In all other marsupials so far investigated the embryo is nourished exclusively through the yolksac and a definite yolk-sac placenta of somewhat complex character is present" (Amoroso, 1952). Thus it is apparent that Grosser's theory does not apply to placentation in the majority of lower placental vertebrates or to the Metatheria (marsupials).

In eutherian mammals, on the other hand, the most typical structure subserving physiologic exchange between mother and fetus and which is constantly present, is the chorio-allantoic placenta (Hamilton, Boyd and Mossman, 1952). The yolk sac in these mammals is the most variable of the fetal membranes. It may occur as a primitive bi

laminar yolk sac, or as a vascularized trilaminar yolk sac which develops early and is temporary. In some orders of mammals (rodents, bats, insectivores, armadillos), a very different and more complex structure, an "inverted" yolk sac placenta, develops. This usually increases in extent during gestation and in most species becomes covered with elaborately branched, vascularized villi which are in contact with the uterine mucosa (Amoroso, 1952; Hamilton, Boyd and Mossman, 1952). In ungulates, cetaceans, lemurs, sloths, and the Simiae (monkeys, apes, man), it has been assumed that the yolk sac, although present as a vesicle, plays little or no role in the metabolic exchange between mother and fetus. However, some histochemical findings on the human yolk sac challenge this assumption. In a histochemical study of 5-, 6-, and 7-mm. human embryos, McKay, Adams, Hertig and Danziger (1955a, b) localized the following substances in the yolk sac endoderm : glycogen, glycoprotein, ribonucleoprotein, acid and alkaline phosphatase, 5-nucleotidase, and nonspecific esterase. These investigators suggested that the large amount of glycogen in the yolk sac and its absence from the fetal liver may indicate that the yolk sac is supplying glucose to the embryo during the first weeks of embryonic life. McKay and his associates pointed out that there is no iron in the human yolk sac endoderm, whereas the rodent yolk sac is rich in this substance (Wislocki, Deane and Dempsey, 1946).

The inverted yolk sac placenta of rodents and lagomorphs has received more attention since Brambell and his associates (1948, 1949, 1951, 1957) demonstrated that maternal antibodies are transferred exclusively by this ancient membrane. A short description of the histology of the inverted yolk sac of the rat follows for the purpose of general orientation. A good diagram of the histology of the rat placenta was published by Anderson (1959). The yolk sac placenta of the rat is divided into two morphologic zones. (1) An outer, nonvascular parietal wall (bilaminar omphalopleure) consists of scattered cuboidal endodermal cells w^iich form an incomplete lining on the interior surface of Reichert's membrane.

918

SPERM, OVA, AND PREGNANCY

This thick and unusual basement membrane adheres externally to a meshwork of trophoblastic giant cells. Maternal blood flows through the interstices of this meshwork of giant cells and presumably is a regional source of some of the substances which gain entrance to the vitelline circulation. A portion of the parietal wall of the yolk sac is firmly attached to the fetal surface of the chorio-allantoic disc. (2) An inner, vascular, visceral wall (visceral splanchnopleure ) is composed of a simple columnar endodermal epithelium which rests on a mesenchymal layer which carries the vitelline blood vessels. A serosal basement membrane (Wislocki and Padykula, 1953) separates this mesenchymal layer from a narrow, basophilic layer of mesothelium which lines the exocoelom. As the allantoic vessels penetrate the placental labyrinth, portions of both visceral and parietal walls of the yolk sac are invaginated into the labyrinth, forming perivascular recesses which were called "endodermal sinuses" by Duval (1892). In the rat the parietal wall of the yolk sac breaks down on the 15th day of gestation, and this event makes the yolk sac cavity confluent with the uterine cavity and also puts the visceral endoderm into direct contact with the uterine contents.

On the basis of histophysiologic studies on the absorption of dyes, Everett ( 1935 1 concluded that the yolk sac of the rat is a significant organ of exchange and that it is more permeable to dyes than the labyrinth. Vital dyes, such as trypan blue, which are relatively large molecules, find their way rapidly into the yolk sac where they are absorbed and stored by the visceral endoderm. These dyes reach the yolk sac, apparently, either by way of the uterine mu

cosa or through that portion of Reichcrt's membrane covering the fetal surface of the allantoic placenta. Brambell and his coworkers (1948, 1950, 1951, 1957) have established experimentally in the rabbit and rat that antibodies find their way from the maternal circulation into the embryos, not by passage through the thin and supposedly more permeable layers of the chorio-allantoic placenta, but by way of the yolk sac placenta, the latter mode of entry necessitating transfer across several layers of cells and tissues, including the structurally elaborate vitelline epithelium. Histologic evidence in support of transport of antibodies and serum proteins by the yolk sac placenta comes from the localization of absorbed serum proteins labeled by fluorescent dyes (Mayersbach, 1958), and from autoradiographic studies (Anderson, 1959). Both investigations substantiate the impermeability of the labyrinthine trophoblasts to these labeled proteins and demonstrate the accumulation of proteins in the visceral endoderm. Brambell and Halliday (1956) and Alayersbach (1958) have suggested from different kinds of evidence that the endodermal sinuses of Duval may also particil^ate in antibody transport. Padykula (1958 » demonstrated a rise in the succinic dehydrogenase activity of the visceral component of the endodermal sinuses shortly before term in the rat.

The absorptive visceral endodermal cells are interesting from the points of view of both cytology and placentation, since these cells are capable of transporting certain large molecules, such as antibodies and serum proteins, and withholding and segregating other colloidal substances, such as trypan blue. In the latter respect, they func

Plate 15.x

Fig. 15.43. Semischematic drawing of a human placenta delivered at full term. Two cotyledons are illustrated, bounded above by the so-called chorionic or closing plate (c.p.) and on the sides and below by septa placentae (s.) and the basal plate (b.p.). Placental branches of the umbilical blood vessels are seen in the closing plate and in the anchoring villi (o.v.).

Fig. 15.44. A section through a delivered placenta at full term, showing a darkly stained placental septum extending up from the base of the placenta and forming the boundary between two cotyledons. At the top of the figure, blood vessels in the closing plate are apparent. Buffered formalin fixative. Azan stain. X 3.

Fig. 15.45. A photograph of a placenta at full term showing a portion of a placental septum at a higher magnification. Buffered formalin fixation. Periodic acid-Schiff stain. By this method the septum is seen to consist of darkly stained ground substance composed principally of fibrin in which there are numerous lacunae containing faintly stained individual trophoblast.* or colonies of them. X 90.

HISTOCHEMISTRY OF PLACENTA

919

'<%

Plate 15.X

920

SPERM, OVA, AND PREGNANCY

tion as phagocytes. The absorption of large molecules is believed to occur by the process of pinocytosis. The fine structure of these cells has been described in the guinea pig (Dempsey, 1953) and rat (Wislocki and Dempsey, 1955b). Some recent electron micrographs of the rat yolk sac (Figs. 15.8015.83) are presented in this review by Padykula. The free surface of these cells typifies that of a membrane engaged in pinocytosis. There are numerous surface projections or microvilli which are pleomorphic and branch frequently (Figs. 15.81 and 15.82). These projections form the brush border which has long been recognized with the light microscope, and which is rich in glycoprotein and, at certain times in gestation, in alkaline phosphatase. In both the rat and guinea pig these surface projections l)ccome simpler and shorter near term. In addition to these evaginations, the surface plasma membrane is invaginated in the form of minute anastomosing tubules which have a denser thicker wall than tiie microvilli (Fig. 15.82). It seems fairly certain that during pinocytosis a local enlargement of such a tubule is produced and a pinocytotic vesicle is formed. Vesicles fill much of the supranuclear cytoplasm, and there is considerable heterogeneity in the size and content of the supranuclear vesicles (Fig. 15.80). Filamentous mitochondria occur throughout the cytoplasm. The endoplasmic reticulum is most concentrated around tiio nuclei, although it is also diffusely distributed throughout the cytoplasm. The typical agranular membranes and vesicles of the Golgi apparatus can be recognized near the nucleus (Fig. 15.83). Glycogen is stored in the lower half of the cell, especially in the infranuclear region. Lipid droplets occur

throughout the cytoplasm, but the larger ones are usually infranuclear where they are often in close association with the basal surface of the nucleus (Figs. 15.81 and 15.83). The supranuclear lipid is often in the form of aggregations in complicated associations with membranes (Figs. 15.79 and 15.81). Minute lipid droplets are also found within the nucleus (Figs. 15.78 and 15.80). The lateral cell boundaries of these cells are closely apposed early in gestation, whereas near term large lateral intercellular dilatations occur. Between these dilations, where the plasma membranes are closely apposed, desmosomes are evident. The bases of the cells rest on a narrow basement membrane.

With the electron microscope, experimental cytologic analyses of absorption by the ^•isceral cndoderm of the rabbit and mouse have been made by Luse and her associates (Luse, 1957; Luse, Davies and Smith, 1959; Luse, Davies and Clark, 1959). The following materials were injected into the uterus of the rabbit and mouse: colloidal gold, egg albumin, lii)ids, saccharated iron, bovine y-globulin, and salivary gland virus. All of these materials entered cytoplasmic pinocytotic vesicles. However, more interestingly, iron, colloidal carbon, and salivary gland virus penetrated the nuclei. Further work suggests that pinocytosis by the nuclear membrane is the method of nuclear ]ienetration. Similar nuclear inclusions arise in the nuclei of newborn rat duodenum after suckling. This amazing intracellular pathway of absorption i)robably occurs naturally as witnessed by the occurrence of lipid in the nuclei of normal visceral endoderm, duodenum, and liver.

The inverted yolk sac placenta of the rat undergoes a striking differentiation during

Plate 15. XI

Fig. 15.46. A portion of a placental septum showing cytotroplioblasts surrounded by ground substance. Human placenta at 3% months of gestation. Masson's triacid stain. X 240.

Fig. 15.47. A cell island containing a group of cytotioi)liol)last8 surrounded by ground substance. Human placenta at 2 months of gestation. Azan stain. X 240.

Fig. 15.48. Cytotrophobla.sts of the trophoblastic slioU stained by Baker's acid hematein method for phospholipids. Human placenta at 3V2 months of gestation. Observe the intense reaction in the cytoplasm of the tiophoblasts. Compare with Figure 15.25. X 160.

Fig. 15.49. A portion of a placental septum of a human placenta at full term, stained by the periodic acid-Schiff method after exposure of the section to saliva. The cyto])l;ism of the cytotrophoblasts exhibits a faint reaction which should be compared with Figuio 15.55 which illustrates the cells at higher magnification. The clumps of cells are surrounded by masses of intensely stained fibrin. Compare with Figures 15.32 and 15.46 which illustrate the trophoblastic shell at 3 to 3M montlis of gestation before a great deal of fibiin has appeared. X 300.

HISTOCHEMISTRY OF PLACENTA

921

Plate 15. XI

922

SPERM, OVA, AND PREGNANCY

its brief life history (Padykula, 1958a, b) . A major architectural reorganization occurs with the loss of the parietal wall shortly after mid-gestation. A short period of lipid storage for from 10 to 15 days is succeeded by a phase of glycogen storage from 15 to 20 days. After the loss of the parietal wall, there is a sharp rise in certain enzymatic activities (alkaline phosphatase, adenosine triphosphatase, acid phosphatase, succinic dehydrogenase) . This burst of vitelline activity in this last third of gestation suggests greater functional activity after direct exposure of the visceral endoderm to the uterine contents. As in the case of the chorio-allantoic placentas, the reduction in the number of layers of the yolk sac probably increases the rate of absorption by the vascularized splanchnopleure. Shortly before term there is a sharp decline in glycogen content and certain enzymatic activities in the visceral endoderm. As these particular histochemical properties decline in the visceral yolk sac, they appear in the fetal liver with good temporal correlation. If the yolk sac functions in part as a fetal liver, then the terminal decrease in enzymatic activity should not be interpreted as placental aging, but rather as a redistribution of the functional activities of the placental-fetal complex (Padykula, 1958). Further morphologic aspects of aging in the placenta were discussed by Wislocki (1956).

With the ascendency of the chorio-allantoic placenta in eutherian mammals and its postulated progression in the sense of Grosser's series from a simple epitheliochorial placenta to the physiologically more "efficient" hemochorial type, one might have expected that the mammalia would have abandoned yolk sac placentation altogether. But that is not the case, for paradoxically the greatest placental development of the yolk sac, involving inversion, is associated with hemochorial placentas (rodents, lagomorphs, bats, some insectivores), whereas the least developed yolk sacs occur in animals possessing epitheliochorial placentas which are the most primitive, according to Grosser's scheme. The general adoption of Grosser's concept of the chorio-allantoic placenta has resulted in the almost complete exclusion of the yolk sac. For example.

Needham (1931, Table 227) attributes the entire transfer of substances in mammals to the chorio-allantoic placenta, without reference to other avenues of exchange. In view of the observations and experiments described here, it is evident that Grosser's doctrine will have to be re-evaluated and modified to include yolk sac placentation.

VI. Histochemistry with Reference to Comparative Placentation

It is beyond the scope of this chapter to describe in detail the structure and cytology of the placentas of various mammals. However, certain histochemical observations on lipids will be presented because they afford some clues to the probable sites of localization of placental steroid compounds. In addition, the localization of glycogen, other complex carbohydrates, and phosphatases in several types of placentas will be summarized. This histochemical information will serve as a basis for subsequent comparisons of placental structures and functions.

However, before proceeding to these matters, attention should be drawn, for readers who may wish to familiarize themselves with comparative placentation, to the compendia of this subject by Grosser (1925b), Mossman (1937), and Amoroso (1952). Recent papers on placental histochemistry of various animals will be listed here. These include investigations of the chemical morphology of the placentas of the pig (Wislocki and Dempsey, 1946b), sheep and cow (Wimsatt, 1950, 1951), shrews (Wislocki and Wimsatt, 1947), cat (Wislocki and Dempsey, 1946a), and bat (Wimsatt, 1948, 1949). In rodents histochemical observations are more numerous, including earlier investigations of fat, glycogen, and iron in placentas of the rabbit (Chipman, 1902) and rat (Goldmann, 1912). More recent studies describe glycogen in the rat's placenta (Szendi, 1933; Krehbiel, 1937; Bridgman, 1948; Bulmer and Dickson, I960), alkaline phosphatase m the guinea pig's placenta (Hard, 1946; Nataf, 1953), and in the pregnant uterus of the rat (Pritchard, 1947), and multiple histochemical reactions in placentas of rats, mice, guinea pigs, and rabbits (Wislocki, Deane and Dempsey,

HISTOCHEMISTRY OF PLACENTA

923

1946; Bridgman, 1948a, b; Wislocki and Padykula, 1953; Davies, 1956; Padykiila, 1958). The Ashbel-Seligman reaction for carbonyl groups has been briefly described in the placentas of various mammals (Ashbel and Seligman, 1949; Wislocki, 1952), as has also the PAS reaction (Wislocki, 1950) . Many histochemical observations on a variety of placentas were summarized by Starck (1945-50).

Present information on the histochemical localization of lipids in various placentas is fragmentary. The main effort has been directed toward localizing within the placenta the type of lipid droplets which occur in the steroid-producing cells of the adrenal cortex and gonads. These droplets are acetone soluble, birefringent, exhibit greenish fluorescence, and give positive Ashbel-Seligman and Schif! reactions for carbonyl groups. As has been discussed in the section on methods, these nonspecific reactions actually reflect the degree of unsaturation in the compounds comprising the fixed lipid droplets. Certainly further work is needed to characterize more fully the various lipids of placentas, especially in relation to the storage of cholesterol and triglycerides.

In the following descriptions it will be seen that the lipid reactions characteristic of steroid-producing cells usually occur in some part of the trophoblast. In the human placenta, only the syncytium contains the lipid droplets characteristic of steroid-producing cells. The cat possesses a so-called endotheliochorial type of placenta, consisting of sinusoidal maternal capillaries and "decidual" giant cells arranged in sheets alternating with lamellae of trophoblast, the latter enclosing the fetal stroma and capillaries. The trophoblast consists of an outer syncytial and an inner cellular layer. The cellular layer contains abundant lipid droplets of variable, but relatively large size. They are sudanophilic and birefringent (Fig. 15.59), exhibit greenish fluorescence, give an Ashbel-Seligman reaction for carbonyl groups (Fig. 15.20) and stain intensely with Schiff's reagent. The syncytium is negative in these respects, except for a mild diffuse coloration bv the Ashbel

Seligman carbonyl method (Fig. 15.20) and a greyish tint with sudan black B which is attributable probably to mitochondria. The "decidual" giant cells, generally regarded as of maternal origin, and the maternal endothelium give no lipid reactions beyond a delicate sudanophilia associated with the presence of mitochondria.

In rodents two types of placentas, a chorio-allantois of the hemochorial type and a yolk sac placenta, function concurrently throughout gestation. Lipids occur in many placental constituents of the rat: labyrinthine trophoblast, giant cells, parietal endoderm, visceral endoderm, decidua capsularis, and mesothelium lining the exocoelom. Bridgman (1948) pointed out that in the rat the labyrinthine trophoblast contains lipid from the 12th day onward and that it diminishes shortly before term. This lipid in rats and mice is birefringent and gives an intense carbonyl reaction (Figs. 15.16, 15.17, and 15.21) (Wislocki, Deane, and Dempsey, 1946; Ashbel and Seligman, 1949; Wislocki, 1952). This cellular layer is a logical suspect as the site of steroid hormonal synthesis. However, these reactions are present also in the visceral endoderm of the yolk sac where lipid droplets occur in great abundance from 9 to 17 days (Figs. 15.74-15.77) . These lipids which occur principally as large infranuclear droplets (Figs. 15.78 and 15.83) are birefringent, strongly fluorescent, and give a strong carbonyl reaction with the Schiff reagents (Wislocki, Deane and Dempsey, 1946). Further work has confirmed these observations and has also shown this acetone-soluble lipid gives a strong Ashbel-Seligman reaction and contains cholesterol. It immediately turns a brilliant blue-green in the Schultz test (Padykula, unpublished observations) . One difference between the labyrinthine and endodermal lipids is the color response in the Schultz test. The labyrinthine lipid turns red-brown but never the blue-green which indicates the presence of cholesterol. Lehner (1914) reported that intranuclear lipid droplets are abundant in the visceral endoderm of the mouse. This finding is confirmed in the rat by electron microscopy (see Figs. 15.78 and 15.80). The significance of this lipid in the yolk sac is not clear, al

924

SPERM, OVA, AND PREGNANCY

though the findings of Luse (1958), Luse, Davies and Smith (1959), Luse, Davies and Clark (1959) suggest that it is absorbed lipid. Further discussion of the lipids of the yolk sac was given in the section on yolk sac placentation.

In placentas of two species of shrews {Blarina brevicauda and Sorex fiimeus) Wimsatt and Wislocki (1947) described numerous coarse lipid droplets in the columnar trophoblastic epithelium forming the chorionic membrane. These droplets are sudanophilic and birefringent, exhibit greenish fluorescence, and give an intense reaction with Schiff's reagent. In the chorio-allantoic placenta, minute sudanophilic particles are observed in the placental trabecule, but birefringence and fluorescence are not evident and Schiff's reaction is feeble.

In the placenta of the bat {Myotis lucifugus lucifugus) Wimsatt (1948) observed sudanophilic lipids in nearly all placental constituents, but only those present in the columnar trophoblastic cells of the membranous chorion were birefringent, emitted a greenish-yellow fluorescence, and gave positive phenylhydrazine, Schiff's and Liebermann-Burchardt reactions. With respect to these reactions, it is apparent that the membranous chorion of shrews and bats is similar.

In the placenta of the Virginia deer {Odocoileus virginiayms borealis) in midgestation, lipid droplets which are birefringent and give an Ashbel-Seligman reaction are

present in a layer of epithelium lining the maternal crypts of the placentomas (Wislocki, 1952) . A further interesting feature of the deer's placenta is an intense reaction for lipids (sudanophilia, birefringence, positive Ashbel-Seligman carbonyl reaction) in the withered, degenerating peripheral ends of the maternal septa. This reactive material is evidently attributable to degeneration of a portion of the epithelium covering the maternal septa.

In the sheep, the epithelial layer clothing the maternal septa consists of syncytial trophoblast derived from the chorionic villi (Wimsatt, 1950; Amoroso, 1951, 1952). Although it has not been investigated, the epithelium lining of the maternal crypts of the deer's placenta may also be of fetal origin. In the sheep, Wimsatt (1951) remarks briefly that lipid droplets, which are birefringent and give positive Baker's acidhematein and Liebermann-Burchardt reactions, are present in the columnar trophoblastic cells, but no mention is made of the reaction of the syncytial tro]ihoblast lining the maternal crypts.

In view of the localization of these various lipid reactions in some part of the trophoblast, results obtained in the placenta of a pig (17 cm. crown to rump length) are an interesting exception (Wislocki, 1952). At this period of gestation, lipids are not encountered in the chorionic epithelium, except phospholipids of mitochondria which are demonstrable by means of sudan black B (Fig. 15.15) and Baker's acid hematein

Plate 15. XII

Fig. 15.50. Human placental labyrinth at full term, .-stained by the periodic acid-Schiff (PAS) method after exposure of the section to saliva. Zenker's acetic acid fixative. The walls of the sinusoidal fetal capillaries and the (reticular) basement membrane upon which the syncytium rests are deeply stained. The outer zone of the syncytium is also noticeably stained. Although many of the capillaries deeply indent the syncytium producing so-called "syncytial" or "epithelial plates," a narrow rim of syncytium and the PAS-stained wall of the subjacent capillary always intervene between the intervillous space and the lumen of the capillary. Compare with Figures 15.14. 15.19, 15.25, and 15.41. X 280.

Fig. 15.51. Degenerate placental villi at 2 months of gestation, illustrating their intense metachromatic staining with toluidin blue. Basic lead acetate fixative. Compare with Figure 15.57. X 240. (Wislocki and Dempsey, 1948.)

Fig. 15.52. Human decidua vera at 2'/2 months of gestation, stained by the PAS method. Rossman's fixative. Observe the pronounced staining of ground substance encapsulating the poorlv stained decidual cells. Two arterioles are visible near the center of the figure. Compare with Figures 15.53 and 15.58 which illustrate decidua stained by other means. X 240. (Wislocki and Dempsey, 1948.)

Fig. 15.53. Human decidua vera at 4 months of gestation, .showing the acid phosphatase reaction of the decidual cells. Gomori's method, using glycerophosphate as substrate at pH 47. X 240. (Wislocki and Dempsey, 1948.)

HISTOCHEMISTRY OF PLACENTA

925

».«^

^

t'^'if^*^^*'

i%

1. %

53

Plate 15.XII

926

SPERM, OVA, AND PREGNANCY

test. Instead, extremely minute, lipid droplets giving an Ashbel-Seligman reaction are present in the epithelium of the uterine mucosa (Figs. 15.15 and 15.22). Later in gestation, from about the 20-cm. stage on, large sudanophilic lipid droplets begin to appear in the basal ends of the columnar trophoblastic cells of the chorionic fossae (Wislocki and Dempsey, 1946b), but these have not been studied by other histochemical reactions for lipids.

B. GLYCOGEN AND CARBOHYDRATE CONTAINING MACROMOLECULES

Since Claude Bernard suggested in 1859 that the placenta may perform the glycogenic function for the embryo before the developing liver has acquired this function, considerable attention has been given to localizing this important metabolic reserve. In previous paragraphs, the localization and fluctuations in glycogen were reviewed for the human placenta. In this species, glycogen has a widespread distribution during the first 2 months, occurring in the syncytium, Langhans cells, stromal fibroblasts, Hofbauer cells, peripheral cytotrophoblasts, and decidual cells. After the second month, there is a sharp decline in glycogen storage in all these components, except in the de

cidual cells which retain glycogen until term. This decline has also been recorded biochemically by Villee (1953) whose measurements of glycogen content show a rapid drop after 8 weeks of gestation. Furthermore, glucose production is possible early in gestation but not at term (Villee, 1953). Concerning Bernard's hypothesis, it is interesting to note Villee's observation that the glycogenic storage function of the human fetal liver is acciuired at 7 to 8 weeks of gestation. Thereafter, the glycogen content of the fetal liver rises sharply, as placental glycogen content falls.

Glycogen storage in the rat placenta is also widespread, occurring in various parts of the maternal-fetal complex (Goldman, 1912; Krehbiel, 1937; Bridgman, 1948a, b; Padykula, 1958b; Bulmer and Dickson, 1960). Early storage during the first ten days is chiefly a decidual function. However, the trophoblastic ectoplacental cone and its later derivative, the spongy zone, contain some glycogen from implantation until term, with peak storage occurring the 15th day of gestation. The vascularized trophoblast of the labyrinth and the visceral endoderm of the inverted yolk sac placenta initiate glycogen storage at 14 days, reach a peak at 18 days, and have released most

Pl.\te 15. XIII

Fig. 15.54. A portion of a human chorionic villus at 21/2 months of gestation, stained by the periodic acid-Schiff (PAS) method (exposed to saliva). Orth's fixative. Observe the marked reaction of the outer zone of the syncytium, the delicate stippling of the deeper layer, the chromophobic appearance of the Langhans cells, the intense staining of the basement membrane and the strong response of the large vacuolated Hofbauer cell. Compare with Figure 1529. X 7 ocular; X 90 objective.

Fig. 15.55. Cytotrophoblasts from a human placental septum at full term, showing the cells partially surrounded by dark red stained fibrin. Orth's fixative. PAS stain. The cytoplasm of the trophoblasts contains a delicate stippling of PAS positive material as well as accentuated staining around the nuclear membrane. Compare with Figure 15.48. X 7 ocular; X 90 objective.

Fig. 15.56. A lamella of a cat's placenta, stained by the PAS method. Orth's fixative. Treatment with saliva. Observe the intense reaction in a narrow zone located between the maternal capillaries and giant decidual cells and the trophoblastic sjmcytium. In the trophoblast occasional large intensely stained droplets of "colloid" are visible. Compare with Figures 15.60 and 15.61. X 7 ocular; X 60 objective.

Fig. 15.57. A degenerating placental villus at 6 months of gestation, consisting of degenerating stroma which has become intensely metachromatic (red), surrounded by a mantle of bluish green-stained, hyalinized syncytium. Basic lead acetate fixative. Toluidin blue stain. Compare with Figure 15.51. X 7 ocular; X 20 objective.

Fig. 15.58. The decidua basalis of a human placenta at 2V2 months of gestation, illustrating the characteristic red metachromasia of the ground substance surrounding the decidual cells. A cleft in the decidua contains bluish green-stained fibrin. Basic lead acetate fixative. Compare with Figure 15.52, stained by PAS reagents. Toluidin blue stain. X 7 ocular; X 40 objective.

56 _

HISTOCHEMISTRY OF PLACENTA

927

of this material by the 21st day. Concerning tliese fluctuations, it may be said that the glycogen content of the fetal placenta is highest immediately preceding the great terminal growth spurt of the embryo and fetal placenta. It should also be noted that, as in the human, placental glycogen content is decreasing during the period when the fetal liver is beginning to store glycogen (Padykula and Leduc, 1955).

The distribution of glycogen in the rabl)it placenta is drastically different from that of the human and the rat. In this species glycogen is localized exclusively in the decidua of the maternal placenta ; the fetal placental tissue, including the yolk sac, is devoid of glycogen (Bernard, 1859; Chipman, 1902; Lochhead and Cramer, 1908; Loveland, Maurer and Snyder, 1931 ; Tuchmann-Duplessis and Bortolami, 1954; Davies, 1956). Glycogen content reaches a peak near the 17th day of gestation, and decreases until term. Several investigators (Lochhead and Cramer, 1908; TuchmannDuplessis and Bortolami, 1954) have further substantiated Claude Bernard's observation that the decline in placental glycogen correlates in time with the onset of the hepatic glycogenic function in the fetus. In the guinea pig a similar temporal correlation has been made for glycogen storage in the placenta and fetal liver (DuBois and Ducommun, 1955).

Saliva-insoluble carbohydrates, such as glycoproteins and mucopolysaccharides, revealed by the PAS reaction are demonstrable in the placentas of all animals which have been examined (Wislocki, 1951). In the pig's placenta a positive reaction is given by minute droplets in the apical ends of the uterine gland cells, in the glandular secretion (Fig. 15.66), and in the uterine surface epithelium. An intense reaction is given by the secretion (uterine milk) in the lumens of the chorionic areolae. Numerous ]iositively stained, delicate droplets are present in the distal cytoplasm of the columnar epithelium lining the chorionic fossae and areolae (Fig. 15.65). In the basal part of the tall columnar cells lining the chorionic fossae there are, in addition, large colloid" droplets (Fig. 15.65) which are stained intenselv red (Wislocki and Dempsev, 1946b).

In the numerous trophoblastic binucleate giant cells of the sheep and cow, Wimsatt (1951) reported the presence of many PASpositive cytoplasmic granules. The trophoblastic giant cells of the Virginia deer react similarly (Wislocki, unpublished observation) .

In the chorionic lamellae of the cat's placenta, deeply stained PAS-reactive material is present between the fetal trophoblast and the maternal vessels (Figs. 15.56 and 15.61). In addition, large, deeply stained colloid droplets are located irregularly in the trophoblastic syncytium (Fig. 15.56). In the placental "brown" border of the cat, a reaction is present in the chorionic epithelium, as well as in the secretion in the uterine lumen and in the surface and glandular uterine epithelium (Fig. 15.62) .

In the rat's placenta, the apical cytoplasm of the uterine epithelium, amorphous material in the uterine and vitelline cavities, the substance of Reichert's membrane and the apical cytoplasm of the vitelline epithelial cells (Fig. 15.72) all react strongly (Wislocki, and Padykula, 1953) . The trophoblast of the chorio-allantoic placenta gives a relatively faint reaction (Fig. 15.72).

C. METACHROMASIA

None of the structures in the placentas of the various animals cited above, which are strongly PAS positive, exhibits any metachromasia (Wislocki, 1953) , except that the binucleate cells of the sheep react faintly under some conditions of staining (Wimsatt, 1951). Also the ground substance of the stroma of the chorionic rugae of the pig's placenta is moderately metachromatic.

D. PHOSPHATASES

Because the methods for alkaline and acid phosphatase were among the first histochemical procedures for localizing enzymes, many observations have been made on this type of hydrolytic activity. The absorptive surfaces of the small intestine, kidney, and placenta, which are characterized by brush borders, contain strong alkaline phosphatase activity. In many forms the syncytial trophoblast is rich in alkaline phosphatase activity. In the human syncytium, alkaline phosphatase activity which is low early in gestation increases greatly

928

SPERM, OVA, AND PREGNANCY

later. This enzymatic activity is high in the syncytial trophoblast of the cat, rodents (Figs. 15.71 and 15.73), shrews, and bats. Alkaline phosphatase occurs in the visceral endoderm of the splanchnopleuric yolk sac of rodents (Fig. 15.70) (Hard, 1946; Wislocki, Deane and Dempsey, 1946; Pritchard, 1947; Padykula, 1958a), shrews (Wislocki and Wimsatt, 1947), and bats (Wimsatt, 1949). Thus, this enzyme is located at two major placental aljsorptive surfaces.

In the pig's placenta, alkaline phosphatase activity is high in the columnar epithelial cells of the chorionic fossae (Fig.

15.67) and in the stroma and blood vessel walls of the maternal endometrium (Fig.

15.68) (Wislocki and Dempsey, 1946b; Dempsey and Wislocki, 1947) . However, it is completely absent from the chorionic areolae and extremely low in the epithelium of the chorionic rugae. In the coiv and sheep, binucleate trophoblastic giant cells are ricli in alkaline phosphatase (Wimsatt, 1951). In the Virginia deer at midgestation, this enzyme was found in the binucleate giant cells, stroma and walls of the blood vessels of the maternal septa, epithelium clothing these septa, and at the surface of the trophoblast covering the chorionic \\\\\. In the

cat, alkaline phosphatase occurs also in material surrounding the capillaries and decidual giant cells (Fig. 15.60). In the paraplacental "brown" border of the cat, there is abundant alkaline phosphatase activity in the uterine glands and surface epithelium, in the uterine secretion, and in the outer parts of the columnar epithelial cells of the membranous chorion (Fig. 15.63).

Acid phosphatase occurs in the human syncytium and also in the labyrinthine trophoblast of the i^at where it increases in activity in the last week of gestation (Padykula, 1958). In the rat placenta, this enzyme appears in the cytotrophoblast of the spongy zone and in the giant cells on the 17th day and increases steadily until term. In the cat, acid phosphatase occurs in the trophoblast of the placental lamellae, in the uterine glands, in the uterine surface epithelium, and to some degree in the epithelium of the membranous chorion (Wislocki, 1953). In the pig, acid phosphatase activity is high in the uterine glands (Fig. 15.69), in the uterine milk occurring in the lumens of the chorionic areolae, and in the distal ends of the epithelial cells lining the arealae (Wislocki and Dempsey, 1946b; Dempsey and Wislocki, 1947) . The activity of the uterine surface epithelium is moderate. In

Plate 15.XIV

Fig. 15.59. Frozen section of the placental labyrinth of a cat, viewed under a polarizing microscope, to illustrate the strong birefringence encountered in the cellular trophoblast of the placental lamellae (fetal crown to rump length, 110 mm.). The birefringence is associated with numerous sudanophilic lipid droplets which exhibit greenish fluorescence and give a positive Ashbel-Seligman reaction for carbonyl groups. Compare with Figure 15.20 which illustrates the carbonyl reaction. X 280. (Wislocki and Dempsey, 1946a.)

Fig. 15.60. A placental lamella of a cat, illustrating the presence of alkaline phosphatase in the interstitial matrix around the maternal blood vessels and decidual giant cells and extending into the syncytial trophoblast (embryo length, 13 mm.). Gomori's method, using glycerophosphate as substrate at pH 9.4. X 800. (Wislocki and Dempsey. 1946a.)

Fig. 15.61. Placental lamellae of a cat (fetal crown to rump length, 45 mm.), illustrating the periodic acid-Schiff (PAS) reaction which is localized in the interstitial matrix surrounding the maternal blood vessels and decidual giant cells and intervening between them and the trophoblast. Compare with the similar localization of alkaline phosphatase in Figure 15.60 and also with Figure 15.56. Zenker's acetic acid fixative. Treatment with saliva. X 260.

Fig. 15.62. The paraplacental endometrium ("brown" border) of a pregnant cat (fetal crown to rump length, 45 mm.), illustrating the intense PAS reaction in the secretion of the uterine glands which is poured into the uterine lumen. A reaction is present also in the cytoplasm of the distal ends of the uterine epithelium. The paraplacental chorionic membrane (opposite the uterine epithelium at the extreme upper border of the photograph) consists of columnar cells the supranuclear cytoplasm of which contains PAS-stained droplets. Zenker's acetic acid fixative. Treated with saliva. X 260.

Fig. 15.63. The paraplacental "brown" border of a pregnant cat (fetal crown to ru.mp length, 45 mm.), illustrating the intense alkaline phosphatase reaction in the endometrium (right) and lesser reaction in the chorion (left). Gomori's method, using glycerophosphate as substrate at pH 9.4. X 120.

HISTOCHEMISTRY OF PLACENTA

929

Tlatk 15.XIV

^30

SPERM, OVA, AND PREGNANCY

the trophoblast covering the chorionic rugae, acid phosphatase activity is low or completely absent.

In the rat placenta the distribution of phosphatase activity toward adenosine triphosphate at alkaline pH has been described by Padykula (1958a).

VII. Evidence of the Possible Site

of Production of Placental

Steroid Hormones

In the section on methods, a procedure for characterizing lipids was outlined for localizing the sites of the synthesis of steroid hormones or their precursors in histologic sections. It was pointed out that none of the reactions involved in the procedure is specific for the identification of ketosteroids, but lipids possessing all of the properties enumerated have been found solely in those organs (adrenals, gonads, and placenta) in which steroid hormones are known to be produced.

In formalin-fixed, frozen sections of human placenta, birefringent, sudanophilic lipid droplets are abundantly present in the syncytial trophoblast throughout gestation. They are acetone soluble, react with phenylhydrazine (Wislocki and Bennett, 1943), give a positive Schiff reaction, exhibit yellowish-green fluorescence (Dempsey and Wislocki, 1944; Rockenshaub, 1952), and also react positively with the Ashbel-Seligraan reagents for carbonyl groups (Ashbel and Seligman, 1949; Seligman, Ashbel and Cohen, 1951; Wislocki, 1952; Ashbel and Hertig, 1952). The lipid droplets diminish in size and relative abundance as gestation advances, but are, nevertheless, still apparent in the syncytium at full term (Wislocki and Bennett, 1943, Figs. 10 and 11). The decrease in droplet size with age may not necessarily indicate a reduction of functional activity, for in the adrenal cortex and ovaries a diminution in the size of the lipid droplets accompanies active secretion (Deane, Shaw and Greep, 1948; Barker, 1951). Furthermore, since the total volume of the syncytium must increase considerably as the placental villi grow and branch, it seems reasonable to assume that the absolute quantity of the

lipids may not actually diminish. Thus, there is possibly no discrepancy between the increase in formation and excretion of steroid compounds in the course of gestation and the total amount of lipids in the syncytium at term.

Strands of syncytium which penetrate the trophoblastic shell and junctional zone in the first trimester of pregnancy (Wislocki and Bennett, 1943) and undergo degeneration are probably responsible for the presence in these regions of occasional patches of lipids giving these reactions.

The uterine glandular epithelium contains large sudanophilic droplets which are not birefringent, but stain with Schiff's reagent (Wislocki and Dempsey, 1945) and give a hydrazide reaction (Ashbel and Hertig, 1952). Wislocki and Dempsey (1945j erroneously equated Schiff's reaction with the "plasmal" reaction and speculated on its possible significance. It now seems more probable that Schiff's reagent, under the conditions of fixation utilized by them, reveals peroxides of unsaturated lipids (Nicander, 1951). Ashbel and Hertig (1952) attributed staining of the epithelium of the endometrial glands by the carbonyl procedure to "ketosteroids," a conclusion which Atkinson, in a discussion of their paper, found difficult to believe, since he observed that all of the cellular elements of the parietal decidua give a positive reaction for carbonyl groups. This objection seems valid, inasmuch as the hydrazide will react with the oxidation products of unsaturated groups.

In mammals, other than man and the rhesus monkey, the observations cited in a previous section of this review indicate that lipid reactions characteristic of steroidproducing cells are also usually located in some part of the trophoblast (cat, shrews, bat, rodents) . However, the pig's placenta is exceptional in that the sudanophilic lipid droplets giving the Ashbel-Seligman and Schiff reactions are located in the uterine epithelium. It is doubted that the reactions in the sow's uterine epithelium are indicative of steroidal synthesis, inasmuch as Wislocki and Dempsey (1945) encountered no birefringence in the epithelium. Estrogens have been detected by various means

HISTOCHEMISTRY OF PLACENTA

931

of assay in the placenta of the sow, being excreted between the 20th and 30th days of gestation and thereafter diminishing, to increase again around the 10th or 12th week and continuing to do so until term (Cowie, 1948). As a possible source of estrogens, the sudanophihc lipid droplets, present in the columnar cells of the sow's chorionic fossae in the latter half of gestation (Wislocki and Dempsey, 1946b), should be investigated further.

An histochemical method for visualizing steroid-3/3-ol-dehydrogenase activity by tetrazolium salts has proved useful in identifying steroid-producing cells in the rat adrenal, ovary, and testis (Levy, Deane, and Rubin, 1959) . Application of this technique to the rat placenta (Deane, Lobel, Driks, and Rubin, I960) has localized steroid-3/?ol-dehydrogenase activity in the trophoblastic giant cells. This activity is greatest between the 8th and 15th day, becomes low by 18 days, and is nearly absent by the 21st day of gestation. Further application of this technique to other placental types should be fruitful.

VIII. Evidence of the Possible Site

of Production of Placental

Gonadotrophic Hormones

A. HUMAN PLACENTA

Friedheim (1929), Sengupta (1935), Gey, Jones and Hellman (1938), Jones, Gey and Gey (1943), and Stewart, Sano and Montgomery (1948) have grown human placental trophoblast in tissue cultures. It has been observed that the cytotrophoblast rather than the syncytium proliferates and that the latter, in so far as it arises, seems to be derived from the cellular form, and is small in amount and atypical in appearance. Friedheim observed no conversion of cytotrophoblast into syncytium in actively growing cultures. Furthermore, Gey, Jones and Hellman (1938), Jones, Gey and Gey (1943) , and Stewart, Sano and Montgomery (1948) demonstrated that tissue cultures ■containing actively growing cytotrophoblast produce appreciable quantities of chorionic gonadrotrophic hormone, even after repeated transplantation over several months. These observations indicate that

the trophoblast, and more particularly the cytotrophoblast, is the Sburce of the hormone.

Stewart, Sano and Montgomery reported their inability to grow trophoblast from mature placentas of the 8th and 9th months. Inasmuch as the syncytium does not divide mitotically and the Langhans cells are numerically much decreased at this period, the result is not surprising. However, if they had cultured tissue containing peripheral trophoblasts, obtained specifically from the placental septa or basal plate, growth might have been anticipated.

Chorionic gonadotrophic hormone is, as a rule, al)undantly present in the urine of women suffering from hydatidiform moles or chorion epitheliomas (Tenney and Parker 1939, 1940; Rubin, 1941), and disappears promptly after the successful surgical removal of these tumors. Tenney and Parker noted that the amount of hormone corresponds roughly to the number of trophoblastic cells in a mole or chorion epithelioma and that a mole with cystic villi and slight trophoblastic proliferation gives a low titer. These findings also indicate that proliferating cytotrophoblast is the source of the hormone and that the syncytium is of less or of no importance.

Wislocki and Bennett (1943) emphasized that the curve of excretion of chorionic gonadotrophic hormone corresponds very well with the period of active proliferation of the trophoblastic shell. Nevertheless, a discrepancy seemed to exist in that the cytotrophoblast has generally been believed to degenerate and disappear in the last trimester, whereas the excretion of chorionic gonadotrophin continues throughout gestation (Venning, 1948). This apparent discrepancy is now understandable in the light of observations reported here which demonstrate that, although the Langhans cells which are chromophobic diminish greatly in number, the peripheral cytotrophoblasts which are chromophilic survive until full term in large numbers in the septa placentas and basal plate as viable, functional cells (Wislocki, 1951).

Baker, Hook and Severinghaus (1944) described blue granules in both the cytotrophoblast and the syncytium of the hu

932

SPERM, OVA, AND PREGNANCY

man placenta, demonstrable by a trichrome stain devised by Severinghaus (1932). According to them, these granules enlarge toward the surface of the syncytium and then seem to liquefy, forming vacuoles which liberate their contents through the brush border into the maternal blood stream. In late pregnancy, as the syncytium becomes thinner, these granules and vacuoles disappear. The investigators interpreted these findings as signifying that the trophoblast of early pregnancy performs a significant secretory function, and they emphasized that the period of activity was roughly contemporaneous with the time of greatest excretion of gonadrotrophin.

Bruner and Witschi (1947) and Bruner (1951), investigating tlie distribution of chorionic gonadotrophin in the human placenta by biochemical means, reported that it is found at all stages of pregnancy, the concentration being highest in the fetal portion of the placenta. It is evident, they stated, that the major part of the hormone is "released from the plasmotrophoblast into the maternal blood, whereas only a small fraction passes the inner placental barrier, the cytotrophoblast, to enter the fetal blood stream." They apparently believed that the hormone is formed by the synctium.

Similarities between the peripheral trophoblasts and the cells of the anterior lobe

of the hypophysis believed to produce the hypophyseal gonadotrophic hormones, suggested to Dempsey and Wislocki (1945) that the peripheral trophoblasts produce chorionic gonadotrophin. They demonstrated that the cytoplasm of the syncytium and peripheral trophoblasts of the human placenta contains a basophilic substance similar to that in the true basophilic cells of the anterior lobe of the hypophysis, which in both instances is abolished by digestion with crystalline ribonuclease. That the hypophyseal basophilic cells contain ribonucleoprotein was first demonstrated by Desclin (1940). Since ribonucleoprotein is generally concentrated in cells in which protein synthesis is actively taking place, Wislocki, Dempsey and Fawcett (1948) suggested that its presence in the peripheral trophoblast might be related to the formation tlierc of chorionic gonadotrophin. On the other hand, they thought that the ribonucleoprotein in the syncytium which is particularly abundant in the first months of gestation might represent the primary site of synthesis of fetal plasma proteins, a function taken over by the hepatic cells when the fetal liver becomes sufficiently differentiated. Wislocki (1951) observed that the basophilia of the peripheral trophoblasts persists until full term, coinciding with the continued production of chorionic gonadotrophin.

Plate 15.XV

Fig. 15.64. Cliorionic fold.s of a pig's placenta (fetal crown to rump length, 130 mm.), showing many intraepithelial blood capillaries (blood cells not visible), resulting in the formation of many extremely thin epithelial plates some of which appear to be quite as thin as the human "epithehal plates" seen at full term {cj. with Figure 15.50). Compare with Figures 15.23 and 15.28 which are also of pig's placenta. Zenker-formol fixative. Eosin and nifthylene blue stain. X 400. (Wislocki and Dempsey, 1946b.)

Fig. 15. 65. A chorionic fossa of a pig's placenta (fetal crown to rump length, 120 mm.), illustrating the presence of finely dispersed periodic acid-Schiff (PAS) positive material in the distal ends of the columnar cells and coarse PAS stained droplets in their proximal ends. Orth's fixative. X 150.

Fig. 15.66. Uterine glands in the maternal placenta of a pig (fetal crown to rump lengtli. 120 mm.), showing the strongly PAS-reactive secretion in the lumens. Orth's fixative. X 165.

Fig. 15.67. The chorion of a pig's placenta (fetal crown to rump length, 120 mm.) to illustrate the intense reaction of alkaline phosphatase in the columnar cells of the chorionic fossae located between the chorionic rugae. Gomori's method, using nuclei acid at pH 9.6. X 200.

Fig. 15.68. The endometrium of a pig's placenta (fetal crown to rump length, 120 mm.), to illustrate the intense alkaline phosphatase reaction in the endometrial stroma and blood vessel walls. The enzyme is essentially negative in the epithelium (ep) covering the endometrial folds as well as in the epithelium lining the uterine glands (g). Gomori's method, using fructose diphosphate at pH 9.4. X 200.

Fig. 15.69. Uterine glands in a pig's placenta (fetal crown to rump length, 125 mm.), illustrating the intense acid phosphata.se activity of the glandular cells. Gomori's method, using glycerophosphatase at pH 4.7. X 170. (Wislocki and Dempsey, 1946b.)

The gonadotrophic hormones of both pituitary and placenta are known to be glycoproteins (Bettelheim-Jevons, 1958). Purified gonadotrophins of pituitary and placenta contain hexose and hexosamine. Catchpole (1949 J reported that he found a glycoprotein constituent of the basophil cells of the hypophysis of the rat, demonstrable by means of the PAS stain. On the basis of the increase of this reaction after castration, as well as from other physiologic correlates, he concluded that a part of the material represents the follicle-stimulating hormone. Pearse (1949) likewise observed a positive PAS reaction in the pituitary basophils which he ascribed similarly to the gonadotrophic hormone. Moreover, he described a particular type of vesiculated chromophobe which he suggested might represent a phase in the secretory cycle of the basophils. Purves and Griesbach (1951a, b) established that there are two categories of glycoprotein-containing basophils in the rat pituitary, the gonadotrophs and thyrotrophs. (See also discussion in chapter by Purves.) These two groups of basophils can be distinguished on the basis of shape, geographical distribution, nature of granulation, and responses to changes in hormonal environment. Pearse (1949) noticed PAS-positive material in the form of granular masses, globules, and vesicles in "the trophoblast layer of the placenta and in the Langhans cells of chorionepithelioma." Inasmuch as the granular part of this material could be removed bv diastase, it seemed to consist of

glycogen. However, the globules and vesicles which were saliva fast he regarded as being probably of "mucoprotein nature" and as representing chorionic gonadotrophin.

Wislocki, Dempsey and Fawcett (1948) mentioned briefly that after fixation in Rossman's fluid and removal of glycogen they were unable to demonstrate any reaction in the cytotrophoblast by the PAS method. However, on further trials with other fixatives, including Zenker's acetic acid mixture and Orth's fluid, delicate granules were rendered visible (Wislocki, 1950) in the cytoplasm of some fraction of the peripheral trophoblasts throughout gestation (Figs. 15.49 and 15.55). These fine, but often indistinct, particles are more like those described in the pituitary basophils by Catchpole (1949) than the globules and vesicles mentioned by Pearse (1949) which were not seen in these preparations. Although the reactions observed in the peripheral cytotrophoblast may bear a relationship to the presence of chorionic gonadotrophin, it should not be overlooked that the syncytial trophoblast also exhibits a delicately stippled, variable reaction and that its outer surface and brush border are quite strongly stained (Figs. 15.29 and 15.54). Furthermore, it should be borne in mind that a variety of carbohydrate-containing substances, including various mucopolysaccharides, glocoproteins glycolipids, and glycoliproproteins react with the PAS reagents and that such reactive substances are widely distributed in cells and tissues. As a result of this, the possible identification of chorionic gonadotrophin by this single reaction must be accepted with considerable reservation.

Plate 15.XVI

Fig. 15.70. The placenta of a guinea pig (fetal crown to rump length, 75 mm.), illustrating the distribution of alkaline phosphatase. The reaction is extremely intense in the placental cotyledons (fine-meshed s>'ncytium) and diminishes abruptly in the coarse, interlobular syncytium. The villous portion of the yolk sac (at top of figure above the placenta) is also rich in alkaline phosphatase. The subplacenta beneath the placental labyrinth is negative, but the junctional and decidual zones are quite reactive. Gomori's method, using gh^cerophosphate as substrate at pH 9.4. X SVi. (Wislocki, Deane and Dempsey, 1946.)

Fig. 15.71. The placental labyrinth of a mouse, on the 15th day of gestation illustrating the intense alkaline phosphatase reaction in the trophoblastic syncytium. Gomori's method, using glycerophosphate as substrate at pH 9.4. X 170. (Wislocki, Deane and Dempsey, 1946.)

Fig. 15.72. The chorio-allantoic placenta and yolk sac of a rat on the 21st day of gestation, illustrating the periodic acid-Schiff reaction of the epithelium of the villous portion of the yolk sac and of Reichert's membrane. Orth's fixative. Treatment with saliva. X 175. (Wislocki and Padykula, 1953.)

Fig. 15.73. A detail of the guinea pig's placenta shown in Figure 15.70, illustrating the intense alkaline phosphatase reaction in the trophoblastic syncytium enclosing the maternal vascular channels. X 830. (Wislocki, Deane and Dempsey, 1946.)

From the foregoing summary of the evidence regarding the localization of chorionic gonadotrophic hormone, it seems probable that in the human the trophoblast is the site of its formation. However, it is not clearly established whether it is localized in the Langhans cells, syncytium, peripheral cytotrophoblast, or in several of these elements. Evidence favors the peripheral trophoblasts. The Langhans cells are a less likely site because they are chromophobic and decline perceptibly in size and number by the beginning of the last trimester, whereas chorionic gonadotrophin and the peripheral trophoblasts persist until full term. Involvement of the syncytium seems unlikely, because it is the probable site of formation of the placental steroid hormones, and because it gradually loses its cytoplasmic basophilia, whereas the production of chorionic gonadrotrophin continues until full term.

In attempting to evaluate the possible nature of granules, vacuoles, lipid droplets, mitochrondria, and other organelles in the syncytium, the entire role of the placental barrier must be kept in mind. The syncytial trophoblast is chiefly involved in the absorption and transfer of metabolites from the maternal to the fetal blood stream, besides serving as a means of excreting certain waste products. Many of the organelles which its cytoplasm contains are related in some manner to these functions, although it is not possible at present to assign specific roles to many of them. Wislocki and Streeter (1938) and' Wislocki and Bennett (1943) suggested that a considerable number of the vacuoles seen in the syncytium, especially early in gestation, might be related to absorption. They based their opinion on the probability that much of the syncytial cytoplasm is in a state of motion and flux, with the likelihood that maternal plasma is absorbed by a process of pinocytosis in the manner visualized by Lewis (1931) in cells growing in tissue culture. Recent observations of the human placenta with the electron microscope (Boyd and Hughes, 1954; Wislocki and Dempsey, 1955a) bear out this interpretation. On the other hand, waste

products of fetal metabolism, such as creatine, creatinine, and urea, are so readily diffusible that their excretion would in all probability not be associated with the formation of granules which liquefied and formed vacuoles. There is considerable justification for associating the formation of placental steroid hormones with the sudanophilic, birefringent, lipid droplets present in the syncytium, but there is no evidence that steroid hormones are liberated from cells, in the adrenal glands or elsewhere, in a visible sequence of liquefying granules and discharging vacuoles. In regard to placental gonadotrophin, the slight evidence which can be assembled regarding its localization tends to place it in the peripheral trophoblasts rather than in the Langhans cells and syncytium. On the other hand, early in gestation during implantation and subsequent invasion of the uterine wall by the trophoblast, cytolytic substances and enzymes are quite possibly released by the trophoblast and these probably account for some of the granules and vacuoles seen in parts of the trophoblast at that period (Wislocki and Bennett, 1943).

Considering all of the evidence, chorionic gonadotrophic hormone of the human is most likely produced by the peripheral trophoblasts. This opinion is based on the observed results of culturing trophoblast, as well as on several histochemical similarities of the basophils of the pituitary gland with the peripheral trophoblasts (cytoplasmic basophilia, PAS reaction). Of importance for this concept is the observation that, whereas the majority of the Langhans cells diminish in size and number by the 6th month of gestation, many of the peripheral trophoblasts remain viable and functional until full term (Wislocki, 1951), thus coinciding with the continued production of chorionic gonadotrophin. According to these observations, the Langhans cells, clothing the secondary villi and located in the proximal ends of the trophoblastic cell columns, represent a germinal bed composed of chromophobic trophoblasts, from which the lineages of the syncytium and the chromophilic peripheral trophoblasts are separately derived. A further parallelism with the pituitary is apparent here, in that in both organs chrouiophobic cells are postulated as being the precursors of the chromoi:)hilic elements. Little is known about the site of placental gonadotrophin production in infrahuman mammals. The PAS reaction provides little information about possible sites of formation of the hormone in various animals, because the reaction occurs in many placental components. This undoubtedly reflects the presence of other carbohydrate-containing substances besides the gonadotrophic hormone. Several reports on the mare (Cole and Goss, 1943; Rowlands, 1947; Amoroso, 1952; and Clegg, Boda and Cole, 1954) indicate that the equine gonadotrophin is produced by special parts of the endometrium of the maternal placenta called endometrial cups. Observations of Clegg, Boda and Cole (1954) suggest that the glandular epithelium is the site of gonadotrophic hormone production. A recent detailed discussion of the comparative aspects of hormonal functions was presented by Amoroso (1960).

IX. Significance and Relationships of Some Placental Constituents

A. Ribonucleoprotein

Most of the cytoplasmic basophilia encountered in the trophoblast, uterine surface epithelium, and cells of the endometrial glands is due to the presence of ribonucleoprotein. Because most embryonic or rapidly growing cells are rich in this nucleoprotein, it has been difficult to separate the ribonucleoprotein associated with grow^th from that related to the synthesis of specific proteins by placental cells. Intense cytoplasmic basophilia occurs in the trophoblast of the pig (Wislocki and Dempsey, 1946b), cat (Wislocki and Dempsey, 1946a), rodents (Wislocki, Bunting and Dempsey, 1947), and man. Basophilia is extremely intense in the early part of gestation and diminishes in the second half of pregnancy, except in the pig in which it remains constant and in the bat in which, according to Wimsatt (1949), it becomes more pronounced. In the human, it has been proposed (Wislocki, Dempsey and Fawcett, 1948) that one of the functions of the ribonucleoprotein of the syncytium is to synthesize the proteins of the fetal blood plasma before the fetal liver becomes sufficiently differentiated to assume that activity, whereas the ribonucleoprotein in the peripheral trophoblasts might conceivably be related to the formation there of gonadotrophic hormone. Consonant with the former thought is the gradual decline in cytoplasmic basophilia as pregnancy progresses.

Ribonucleoprotein is closely associated with the secretory activities of uterine glands. The surface epithelium and glands of the uterus, just before and during gestation, are rich in basophilic substance, greatest in amount in the pregnant sow, intermediate in carnivores, and least in rodents and in pregnant women (Wislocki and Dempsey, 1945). In the epitheliochorial placenta of the sow, all of the nutritive substances obtained by the fetus must either traverse or be secreted by the uterine surface epithelium or the uterine glands. The glands which are rich in ribonucleoprotein release a copious secretion which reacts intensely with PAS reagents (mucopolysaccharide), but is not metachromatic. It gives a strong reaction for acid phosphatase. This secretion, designated as uterine milk, seems to be absorbed mainly by the cells of the chorionic fossae and areolae. In the cat, the markedly basophilic paraplacental uterine glands release a secretion w^hich is rich in mucopolysaccharide, glycogen, and phosphatases and is absorbed l)y the columnar cells of the brown border of the chorion. The subplacental glands of the cat are strongly basophilic, but react only faintly with PAS reagents. In rodents, the uterine glands and surface epithelium also seem to secrete nutriment which is absorbed through the yolk sac placenta. These secretory cells are basophilic, and their distal cytoplasm and secretion react intensely for mucopolysaccharide with PAS reagents. In the human, at the time of implantation and for a considerable period thereafter, the paraplacental and subplacental glands are moderately basophilic and contain glycogen, PAS-reactive mucopolysaccharide, lipids, and phosphatases. Moreover, unlike the uterine glands of various animals, their secretion contains some metachromatic mucin (Wislocki and Dempsey, 1948). The secretion of the glands located in the basal decidua may supply the growing peripheral trophoblast (trophoblastic shell and cell columns) with nourishment.

B. Alkaline Phosphatase

From the study of the placentas of man, cats, pigs, and rodents, Wislocki and Dempsey (1945, 1946a, b) and Wislocki, Deane and Dempsey (1946) concluded that a layer of alkaline phosphatase intervenes between the maternal and fetal placental circulations. Wislocki and Wimsatt (1947) found this to be true also in shrews, and Wimsatt (1949) observed a layer located similarly between the maternal and fetal circulations in the placenta of the bat. In hemochorial and endotheliochorial types of placentas, the enzyme is usually present in the outermost layer of the trophoblast, whereas in the epitheliochorial placenta of the pig it is present mainly in the stroma of the maternal placenta. Dempsey and Wislocki (1945) pointed out that many substances may depend on phosphatases for their transfer across cellular boundaries, and they suggested that a layer of different phosphatases located in the placental barrier may participate in the transfer of metabolites. Wimsatt (1949) remarked that this view is consonant with a variety of metabolic processes which must be carried out at the placental barrier and accords with the interpretation of the barrier as a "selective" membrane. This distribution of phosphatase in the outer zone and brush border of the human syncytium resembles very strikingly the location of the enzyme in the epithelium of the small intestine and in the convoluted tubules of the kidney. These three absorptive surfaces also have well developed brush borders composed of numerous microvilli. Hence in the placenta, as in these latter sites, alkaline phosphatase may be associated in the microvilli with the absorption of phosphorylated compounds. Alkaline phosphatase is also a component of the barrier in the yolk sac placenta of rodents (Hard, 1946; Wislocki, Deane and Dempsey, 1946; Padykula, 1958), although it fluctuates in the vitelline epithelium at different periods of gestation. As in the human syncytium, the epithelium of the visceral layer of the yolk sac of rodents possesses a brush border.

In various parts of the endometrium and fetal placentas of lower mammals and man, Dempsey and Wislocki (1947) observed differences in alkaline phosphatase reactions following the use of a variety of substrates. Some structures reacted with many substrates whereas others reacted with only one or two. It was concluded that the observed differences could be accounted for most reasonably by the assumption that the tissues contain multiple enzymes of varying specificity which frequently do not coincide in their distribution.

Plate 15.XVII

Lipids of the rat yolk sac

All of the sections illustrated in Figures 15.74 to 1577 were fixed in 10 per cent buffered formalin and stained with sudan black B.

Fig. 15.74. Parietal wall and nonvillous visceral wall of the yolk sac at 13 days of gestation. The parietal endoderm {p) contains minute lipid droplets and rests on an unstained Reichert's membrane (r). Across the yolk sac cavity larger lipid droplets occur in the nonvillous visceral endoderm (v) ; however, the lipid here is less abundant than in the villous portion shown in Figure 15.75. In the left border of the photograph, lipid occurs also in giant cells and decidual cells of the capsularis. X 275.

Fig. 15.75. Villous visceral splanchnopleure at 13 days of gestation. Peak storage of lipid by the visceral endoderm occurs at this time (12 to 13 days). Note that the position of the lipid droplets is principally infranuclear. Lipid is much less abundant in the mesenchyme and mesothelium of the visceral splanchnopleure. Compare with Figures 15.76 and 15.77. Higher magnification of these endodermal cells is provided in Figures 15.78 to 15.81. X 275.

Fig. 15.76. Villous visceral splanchnopleure at 17 days of gestation. By the 17th day of gestation, the villi have elongated and branched, and the lipid content of each endodermal cell has decreased. The cells toward the tip of the villus tend to have more lipid than those at the base. The droplets remain infranuclear. Compare with Figures 15.75 and 15.77. X 275.

Fig. 15.77. Villous visceral splanchnopleure at 19 days of gestation. The endodermal cells are quite free of lipid droplets. There is a background sudanophilia which is mostly concentrated in the mitochondria. Compare with Figures 15.75 and 15.76. X 275.

In the placentas of man, cats, rodents, and shrews alkaline phosphatase increases in the course of gestation as cytoplasmic basophilia decreases. In the human trophoblastic syncytium, alkaline phosphatase first appears in the outer eosinophilic (alkaline) zone and seems to advance into the syncytium as the basophilia recedes. In the placentas of other animals listed above alkaline phosphatase is also usually localized in regions which are acidophilic. Wimsatt (1949) reported an exception to this inverse relationship of basophilia and phosphatase, because in the bat both substances are plentiful during the first half of gestation, whereas, in the second half, phosphatase activity declines while basophilia persists. Wislocki and Dempsey (1945) observed from study of the placentas of various animals that a layer of phosphatase intervenes between the maternal blood stream and regions where glycogen accumulates. Hard (1946) observed a similar spatial relationship in the placenta of the guinea pig between alkaline phosphatase, on the one hand, and accumulation of glycogen and lipid on the other. Wimsatt (1948, 1949) reported similar relations between the two substances in the placenta of the bat. Whether or not these two phenomena are related has been debated (Wimsatt, 1949; Pritchard, 1947). In this connection it is of interest, as more has been learned about the localization of glycogen by the use of the PAS reaction, that the cellular elements of the secondary and tertiary villi of the human placenta up to 6 weeks of gestation are moderately rich in glycogen, at a time when extremely little alkaline phosphatase has made its appearance in the placental barrier. And later, as alkaline phosphatase increases tremendously in amount, stainable glycogen disappears completely from the chorionic villi.

Wimsatt (1949) observed that the trophoblastic cells of the membranous chorion of the bat "contain heavy deposits of neutral fat, phospholipids, and cholesterids" and concluded that phosphatases, abundantly present in the adjacent decidua, "may be involved in the lipid metabolism and transport in this portion of the placental barrier." Regarding this he called attention to the i:)ossibility that phosphatases in the placental barrier provide a mechanism whereby ciruclating fats are phosphorylated, thereby facilitating the absorption and transmission of lipids. In this connection, it is interesting to note that Green and Meyerhof (1952) have shown that these enzymes can phosphorylate some compounds under certain circumstances.

C. Alkaline Phosphatase and the Periodic Acid-Schiff (PAS) Reaction

Moog and Wenger (1952) reported that neutral mucopolysaccharide, as demonstrable by the PAS reaction, occurs at sites of high alkaline phosphatase activity and they cite examples in various tissues and organs. They suggested that the mucopolysaccharide may serve as a cytoskeleton for the enzyme. They described the placental labyrinth of the mouse as an illustration of this relationship, stating that the trophoblastic syncytium is rich in both substances. However, in the similarly constructed placental labyrinth of the rat, it has been observed that the alkaline phosphatase reaction predominates in the trophoblast, whereas the PAS reaction occurs mainly in the adjacent basement membrane which supports the syncytium and encloses the fetal capillaries. Thus, although the two reactions are closely associated, they are by no means located in the same tissue elements. Despite this discrepancy, there is no question but that in many tissues the thesis Moog and Wenger have stated holds true. Indeed, numerous examples which bear out their conclusion can be found in the placentas of various animals. Thus, in the human placenta, intense reactions for alkaline phosphatase and for PAS-positive substances occur in the outer zone and brush border of the syncytial trophoblast. In the cat's placental labyrinth, both occur together with great intensity in the perivascular sheaths intervening between the maternal blood vessels and the trophoblast, as well as in the columnar chorionic epithelium of the paraplacental brown border. In the rat and guinea pig the two reactions are encountered in the epithelium of the visceral layer of the yolk sac placenta. Finally, in the pig's placenta, the two reactions coincide in the distal ends of the columnar cells lining the chorionic fossae.

Plate 15.XVIII

Visceral endoderm of the rat yolk sac

Figs. 15.78 and 15.79. Cytologic distribution of lipids in the visceral ondoderm. Frozen section, sudan black B. In Figure 15.78 the large clear nuclei have tiny clinlcslciol containing droplets which are truly intranuclear. The infranuclear cytoplasm is packed uitli lar^irr dioplets which are also rich in cholesterol. At the surface of these cells note I he ddicalc sudanophilia of the brush border and also the narrow sudanophobic band immediately beneath this border. The supranuclear cytoplasm contains many unstained vacuoles (Fig. 15.78) and also clusters of tiny lipid droplets (Fig. 15.79). In Figure 15.79 the plane of section runs obliquely through the apical cytoplasm. Three large lipid clusters are shown. X 1950.

Fig. 15.80. Electron micrograph of a visceral endodermal cell at 12 days of gestation. A large homogeneous cholesterol rich droplet is located in the cytoplasm near the polymorphic nucleus. Two minute droplets (d), which are also rich in cholesterol, occur within the nucleus. The apical cytoplasm contains vacuoles which are heterogeneous with respect to size and content. The limiting membrane of these vacuoles is usually incomplete. The free surface of this cell is formed of numerous microvilli. Beneath the surface, many small canaliculi with dense walls can be seen. X 7600.

D. Relationship of Lipids to the Placental Barrier

In reference to the question of lipids demonstrable in the placental barrier, Wislocki and Bennett (1943) emphasized, and Dempsey and Wislocki (1944) offered further substantiating evidence to show that, although lipids are demonstrable histologically in great abundance in the human trophoblastic syncytium, they are probably, for the most part, lipids associated with mitochondria and with the production of steroid hormones, rather than lipids in process of transmission across the placenta. Huggett and Hammond (1952) summarized the question of the manner of transmission of fat from mother to the fetus. In their opinion the chemical results do not show how the fat actually traverses the placenta ; "they neither prove nor disprove the possibility of hydrolysis at the placental membrane and subsequent resynthesis in the deep syncytium." All that the results show is "that particular or labeled fatty acids, originally on the maternal side, appear later in the fetal tissues."

Regarding phospholipids, Popjak and Beeckmans (1950) concluded from a study of rabbits that the placenta does riot transmit unhydrolyzed phospholipid molecules to the fetus. Similarly, glycerophosphate which is a phosphorus-containing degradation product of lecithin does not pass unhydrolyzed. In a previous investigation Popjak (1947) showed that fetal phospholipids in rats, rabbits, and guinea pigs are formed by synthesis within the fetal tissues. Popjak and Beeckman's findings offer a situation where phosphatases might serve as the dephosphorylating agents.

E. Fibrinoid

Earlier in this review the histologic properties of the ground substance of the human trophoblastic cell columns and shell were described. Grosser (1925a) called this ground substance "fibrinoid," adopting a term which had been introduced previously to describe substances occurring in a variety of pathologic lesions. Fibrinoid was defined originally as a somewhat refractile, homogeneous, intercellular substance with an affinity for acid dyes and with histologic resemblances to fibrin (Neumann, 1880). Recently, Altshuler and Angevine (1949, 1951) maintained that fibrinoid stains metachromatically with toluidin blue and reacts with the PAS reagents. From this, they concluded that fibrinoid consists of an acid mucopolysaccharide containing mucoitin sulfuric acid. On the basis of metachromatic staining, they identified placental fibrinoid in Nitabuch's membrane, the subchorial plate, and degenerating chorionic villi. Wislocki and Dempsey (1948) described metachromatic staining of the stroma of degenerating villi (Fig. 15.57) and of the ground substance of the decidua (Fig. 15.58), but they did not observe metachromasia of the ground substance of the peripheral trophoblast. However, it is specifically in the latter that Grosser (1925a) placed the placental fibrinoid, distinguishing it from fibrin and indicating that he did not believe it is related to fibrinoid elsewhere in the body. Altshuler and Angevine use the term "fibrinoid" differently from Grosser and other investigators. The staining seen by them in Nitabuch's layer is probably referable to metaehromasia of the ground substance of the decidua and is most likely physiologic rather than pathologic in nature, inasmuch as similar metachromatic ground substance occurs in the endometrium during the normal menstrual cycle (Bensley, 1934; Wislocki and Dempsey, 1948).

Despite the fact that fibrin, placental fibrinoid, and collagen seem to be identical in reference to staining with acid and basic dyes and isoelectric points (Singer and Wislocki, 1948; Sokoloff, Mund and Kantor, 1951) both fibrin and collagen are distinguishable in a number of respects from placental fibrinoid. Moreover, although fibrinoid of pathologic lesions may be metachromatic, the ground substance of the peripheral trophoblast is not and the only placental substance comparable to pathologic fibrinoid is possibly the metachromatic ground substance of the stroma of degenerating chorionic villi.

X. Placental Permeability with Respect to Morphologic Types

The evidence that the placental barrier in individual species becomes more permeable to some substances as pregnancy progresses seems to be adequately established. Utilizing rabbits, it has been shown that permeability to antibodies (Rodolfo, 1934), phenolsulfonphthalein (Lell, Liber and Snyder, 1932), neoarsphenamine (Snyder, 1943), and radioactive sodium (Flexner and Pohl, 1941a) increases during the course of gestation. In the rat, a similar increase in permeability to insulin (Corey, 1932) and radioactive sodium (Flexner and Pohl, 1941b) has been demonstrated, and the same is true for sodium in the guinea pig, sow, goat, and cat (Flexner and Pohl, 1941a-d; Pohl and Flexner, 1941). A progressive increase in rate of transfer of heavy water has also been observed in the guinea pig and man (Gellhorn and Flexner, 1942; Hellman, Flexner, Wilde, Vosburgh and Proctor, 1948). Flexner and his associates related their results to the progressive thinning of the chorio-allantoic placenta of the individual species studied with respect to the reduction of the number and width of the layers in the course of gestation. This is a natural conclusion in view of the morphologic observations reported in preceding passages which show that in individual species there is a diminution in both width and number of layers of the chorio-allantoic barrier in the course of gestation. For the human, this correlation is well illustrated by a series of drawings presented by Flexner, Cowie, Hellman, Wilde and Vosburgh (1948).

Little comparative information exists with respect to the relative permeability of different types of placentas. What data are available concern mainly the over-all exchange through the fetal membranes and give few, excepting inferential, clues to the exact regions and cytologic means of transfer. Moreover, most of the substances followed have been readily diffusible ones and proteins of various kinds which afford no histochemical means for their detection. The manifold combinations of placental structures in various animals and their cytologic complexities have already been outlined to some degree. However, the majority of investigators, speculating on the comparative aspects of physiologic exchange across the placental barrier, have generally ignored all placental structures, excepting the chorio-allantois. The popularity of Grosser's morphologic scheme of the progressive differentiation of the chorio-allantoic placenta, to the almost complete exclusion of the consideration of all other routes of exchange, has been due doubtlessly to its relative simplicity and its adaptability to a concept, widely held until recently, that all placental transmission can be explained on the basis of diffusion and filtration. Moreover, the fact that little is known of the physiologic activity of the placental structures other than the chorio-allantois has also contributed to their neglect.

It has been generally held that various proteins are readily transmitted by the hemochorial placentas of man and rodents, whereas their passage is slow or entirely prevented in epitheliochorial and syndesmochorial types of placentas (Needham, 1931, Table 227). To give a well known example, immune bodies are not transmitted through placentas of ungulates, whereas in animals possessing hemochorial placentas their transfer takes place readily (Kuttner and Ratner, 1923; Ratner, Jackson and Gruehl, 1927; Ratner, 1943). However, the hemochorial placenta of the rat and rabbit is not involved in antibody transfer; the inverted yolk sac placenta handles this function exclusively in these species. In the monkey which lacks a yolk sac placenta, Bangham, Hobbs and Terry (1958) obtained experimental evidence that the hemochorial chorio-allantoic disc handles antibody transfer. Thus, the routes of antibody transfer are different in primates and rodents.

Flexner and his associates have related the variations in rates of transfer of sodium per unit weight of placenta in pigs, goats, cats, rodents, and man to the four morphologic types of Grosser. On this basis they found that approximately 320 times as much sodium passes across a unit weight of the rat's placenta per hour as across the sow's placenta, and that between the sow and the goat the difference is approximately 16 times, whereas between the goat and the cat, both of which possess syndesmochorial placentas (see Section IV C), the difference is slight (Gellhorn, 1943; Flexner, Cowie, Hellman, Wilde and Vosburgh, 1948j. The inference is that the fewer the layers of tissue intervening between the circulations the greater is the rate of transfer per hour of a readily diffusible substance across a unit weight of placenta. However, these results are based on transfer per unit weight instead of transfer per unit absorbing surface of the placental barrier. Weight as a unit of measurement would be more acceptable if the placentas of different animals were essentially alike in their internal structure, so that unit weights contained approximately similar amounts of transmitting surfaces. Actually a gram of pig's placenta contains a large amount of edematous chorionic stroma and much endometrium, including uterine glands, and hence cannot be truly compared or equated with a gram of human placenta which contains relatively closely packed chorionic villi. As a consequence, any representative part of pig's placenta by weight contains little effective absorbing surface and a large amount of extraneous tissue, whereas a similar amount by weight of human placenta contains a relatively large amount of effective absorbing surface. Thus, comparisons of the relative rates of transmission of substances across various types of placental barriers based upon units of weight are relatively unsatisfactory.

Measurements based on the relative areas of the effective transmitting surface would be more significant. However, the rates of transmission of substances exchanged b} diffusion would depend not alone on the surface area of the placental membrane but more exactly on the areas of contact with the maternal and fetal capillaries variously associated with the membrane. To obtain accurate measurements of these complex surfaces would be well nigh impossible and no attempts to do so have been made, with the exception of the surface area of the human placenta, for which crude measurements vary from six square yards (Dodcls, 1924; Rech, 1924) up to twice that figure (Christoffersen, 1934) . At present, there would seem to be no clear demonstration of a correlation between Grosser's four morphologic types and the relative rates at which placental exchange takes place.

Plate 15.XIX

Fig. 15.81. Election micTograph of a \'isceral endodermal cell of the rat .yolk sac at 13 days. The junction of two endodermal cells occurs at J. Only a portion of the cell at the right is shown. Cholesterol rich lipid occurs in different forms both above and below the nucleus. A portion of a large homogeneous droplet (D) is seen in close association with the basal surface of the nucleus. Above the nucleus, lipid clusters {LC) of various sizes are conspicuous. The indi\idual lipid units which comprise the cluster are irregular in shape and are bound together by membranes. Long filamentous mitochondria (M) are oriented roughly parallel to the long axis of this columnar cell. The perinuclear cytoplasm is rich in typical elements of the endoplasmic reticulum, which are often longitudinally oriented. A vacuole (V) similar to those seen in Figure 15.80 occurs near the free surface. The supeificial cytoplasm is composed of pleomorphic, anastomosing microvilli. The plasma membrane of the microvilli {MV) is continuous with that which lines the canaliculi (C). These canaliculi which lie beneath the surface have dense walls, and they anastomose in a complicated fashion. The free surface is further enlarged in Figure 15.82. X 12,800.

It is, of course, quite probable that readily diffusible substances, such as oxygen, water, and salts, are exchanged through the thinnest parts of the placental membranes which are usually confined to the chorioallantoic placenta. However, it is probably not true, as previously generally believed, that almost all substances capable of transmission, including proteins, traverse the thinnest regions. It would seem likely that all of the substances requiring regulation, including most carbohydrates, lipids, and proteins, are exchanged through thicker and more specialized regions.

Cunningham (1920, 1922) was one of the first to discern, from experiments on the differential permeability of the placental barriers of cats and rabbits to potassium ferrocyanide and iron ammonium citrate, that substances which traverse the placenta are divisible into three categories. (1) Those which are diffusible and which meet with no mechanism in the placenta capable of acting on them. These pass by diffusion from mother to fetus, or in the reverse direction, without any mediation on the part of the placenta. (2) Those which meet with a definite preformed, regulatory mechanism. These include most of the substances which are designed for the fetal metabolism, including iron compounds. (3) Finally, those to which the maternal or fetal surfaces of the placental barrier are impermeable. These include most formed substances, such as cells and particulate matter.

From all of these considerations it would seem most likely that readily diffusible substances, such as water, oxygen, and some salts, are exchanged through the thinner parts of chorio-allantoic placentas, whereas more complex substances are transferred mainly through thicker regions, including the paraplacental borders and yolk sac placentas of animals which possess them. If it is true that readily diffusible substances traverse principally the thinner portions of the placental barrier, then with respect to them, the sequence of the several placental types defined by Grosser would continue to be significant. Nevertheless, it is apparent, as Huggett and Hammond (1952) and others have recently emphasized, that the exchange of each substance will have to be individually investigated with reference to its mode of transfer and the factors affecting it, before a clear picture of placental physiology can be drawn.

In view of the results of Brambell and his associates which indicate that antibodies are transmitted through the yolk sac placenta of rodents rather than through the thin placental membrane of the allantoic placenta, one wonders what prevents antibodies and some other proteins from traversing the syndesmochorial and epitheliochorial placentas of ungulates. Is it so much that the placental barrier in ungulates is too thick to permit their passage, or is it perhaps mainly that the barrier lacks the particular provisions which in the rodent's yolk sac facilitate their passage?

The human placenta is interesting in that it i)rovides only one general avenue for the transfer of substances from mother to fetus. The chorionic villi transmit both readily diffusible substances and those requiring chemical mediation of various kinds for their transfer. This raises the question as to whether, here, all substances follow the same morphologic route through the placental membrane. This cannot be answered, except to suggest that possibly they do not, because of slight histologic differences between the chorionic villi in various segments of the villous tree and of possible differences in the relationships of the segments to the maternal circulation. Similarly, the arrangement of the fetal blood vessels and the mode of circulation of the blood within them might also result in differences in permeability and functional activity in different regions of the villous trees. These are questions which should be investigated further. It seems reasonable to anticipate that future biochemical investigations will define transport mechanisms in the placenta, as they have in the kidney tubules and small intestine.

Another jioint of interest concerning the placental barrier is the fact that the trophoblast of the human placenta forms a syncytial sheet completely devoid of intercellular spaces or cement. Unlike capillaries, in which diffusion and filtration of water-soluble substances are believed by some to occur solely or mainly through the intercellular spaces, this continuous sheet of cytoplasm affords the only possible route of placental transfer. In the sow, to take another example, it is interesting that the thinnest epithelium covering the chorionic rugae is syncytial in character; it is particularly through these rugae that the transmission of gases, water, and salts is believed to occur. The relative inability of leukocytes to traverse the placental barrier may possibly be related to the absence of intercellular spaces. In contrast to many chorio-allantoic placentas, yolk sac placentas are composed of discrete epithelial cells with well defined intercellular spaces which appear with great prominence under the electron microscope (Dempsey, 1953).

XI. Suiiiinarizing Reflections on Comparative Placentation and Placental Permeability

The chorion, chorio-allantois, and yolk sac of mammals become variously apposed to the uterine mucosa to give rise to the "placental barrier" which mediates the physiologic exchange between the mother and the fetus. The chorio-allantoic placenta of eutherian mammals undergoes changes in the course of gestation. This aging process is characterized by a gradual diminution in width and cytologic simplification of the various layers of the placental barrier, and by a progressive elimination of one or more of the maternal layers in most groups of mammals. This gradual diminution in width of the chorio-allantoic placental barrier is believed to account for the fact that placental transmission of some readily diffusible substances, such as oxygen, water, and various salts, increases as gestation proceeds.

In those groups of mammals possessing an inverted yolk sac placenta, an elimination of several fetal, instead of maternal, layers occurs in the course of gestation. The principal remaining layer, the visceral endodermal epithelium, undergoes some degree of cytologic aging but does not diminish in width. Some mammals possess still other structures which mediate exchange between the mother and fetus. The principal of these, the various central and paraplacental hematomas of carnivores, persist throughout gestation, without apparent cytologic changes or any reduction in width or number of layers.

In mammals possessing only a yolk sac placenta, such as some marsupials, or solely a chorio-allantoic placenta, such as man, physiologic exchange must be mediated entirely through one type of placenta. In many groups of mammals both types of placentas develop and exist concurrently for varying periods of time, in which event placental exchange seems to be divided between them. However, the respective functional roles of each of these very different placental structures in the transmission of various substances has not been extensively investigated.

Until recently it was generally assumed that nearly all substances which traverse the placenta do so by diffusion through the thinnest parts of the chorio-allantoic placenta. However, it now seems more likely that most substances are regulated in their passage through the placental barrier and that possibly only some readily diffusible substances, such as oxygen, water, and some salts whose rates of placental exchange increase during the course of gestation, are transmitted mainly through the thinner parts of the chorio-allantoic placenta. Proteins, lipids, and carbohydrates which seem to be regulated in their passage are probably transmitted in many different ways and are acted on variously by the enzymes and complex organelles present in the cytoplasm of the cells of the barrier. Recent experimental observations indicate that in the visceral endoderm of the rodent yolk sac the nucleus may be involved in some transfers. In man and monkeys, which possess only a chorio-allantoic placenta, this regulation must take place solely in the trophoblast, but in other groups of animals which possess, in addition, either a yolk sac placenta or placental hematomas, transfer of some substances evidently occurs through the latter structures. Recent investigations indicate that in the rabbit and rat some proteins and dyes, administered experimentally, are transferred by way of the yolk sac. In those marsupials which possess only a yolk sac placenta nothing is known about the mode of physiologic exchange, except that it is sufficient to support fetal development.

Grosser arranged the chorio-allantoic placentas of eutherian mammals in a phylogenic series comprising four placental types dependent on a progressive decrease in the number of layers intervening between the maternal and fetal circulations. According to his theory, the most primitive placental barrier consists of six layers whereas the most highly developed barriers have been reduced to three layers. Although Grosser recognized yolk sac placentas and hematomas as supplemental means of transfer of some nutritive materials from mother to fetus, only his chorio-allantoic placental types have been generally adopted to explain the passage of nearly all substances from mother to fetus. By the removal of a succession of three more or less functionally equivalent, maternal layers in a phylogenic sequence, the placental barrier has been envisioned as becoming progressively narrower, or thinner, and increasingly more permeable to the passage by diffusion of an increasingly larger number of substances. This scheme of the phylogenic simplification of the placental barrier seemed to be repeated in an ontogenic sense by the observed thinning of the placental barrier in the course of gestation in individual species. According to this belief, substances of larger molecular size, particularly proteins, are the last which are enabled to diffuse across the placental barrier in the postulated phylogenic and ontogenic series of stages.

With the growing recognition that the passage of a great many substances across the placental barrier is chemically regulated (Huggett and Hammond, 1952) and that the cytoplasm of the barrier contains a host of enzymes and numerous organelles, it seems increasingly evident that Grosser's doctrine must be reevaluated and modified. It has become necessary to consider each substance individually, with respect to its place of passage and the fatcors regulating its exchange. Thus, the regional cytologic and histochemical organization of the barrier, of which Grosser's doctrine takes little cognizance, should assume much greater importance; and, in addition to the relative thickness and width of the barrier, the relative extent and nature of its absorbing surfaces will have to be more carefully explored. The electron microscope should be of considerable value in ascertaining the structure of the absorbing surfaces and interior of the barrier. Studies of the human placenta in early months of gestation, recently begun with the electron microscope, have revealed a multitude of microvilli on the surface of the trophoblast and vesicles in the syncytium. This suggests that absorption from the intervillous space takes place to a considerable extent by the process of pinocytosis. Indications of absorption by pinocytosis have also been reported by means of electron microscopy in the epithelium of the yolk sac of the guinea pig and rat. A major role in placental physiologic exchange may eventually have to be assigned to pinocytosis. It seems apparent that pinocytosis is a process which predominates in the first part of human gestation but subsequently diminishes as the placental barrier becomes thinner and more simplified. It seems probable that as the rate of transfer of simple substances by diffusion across the barrier increases, the rate of absorption of more complex substances by pinocytosis declines. In rodents' placentas, on the other hand, although the trophoblastic cells of the placental labyrinth become thinner and their cytoplasm more simplified, pinocytosis seems to persist throughout gestation in the epithelium of the yolk sac thereby affording a continuous means for the absorption of substances of larger molecular size. Comparative studies of the placentas of animals with the electron microscope are just beginning, so that, although it seems that pinocytosis might play an important role in placental physiologic exchange, no broader generalizations regarding its significance can at present be offered. It is apparent that in the future more attention will have to be paid to the chemical, histochemical, and cytologic structure of every part of the barrier with respect to the mode of passage of different kinds of substances. JMoreover, a comprehensive theory of placental exchange, from the comparative and phylogenic point of view, will have to take cognizance of all of the structures forming the placental barrier, instead of confining itself to the chorio-allantoic placenta as defined in terms of Grosser's four placental types.

Plate 15. XX Electronmicrographs of the rat visceral endoderm at 13 days of gestation

Fig. 15.82. Free surface of the visceral endoderm. The junction of two cells occurs at J which is marked also by the dense cytoplasmic condensation of the terminal bar. Note that the microvilli (mv) are penetrated by tiny tubules. At point x, the plasma membrane of the microvilli is continuous with the denser membrane which forms the walls of the canaliculi. Near the center of the photograph, the canaliculi are in the form of a figure 8, showing the anastomosis of the system of superficial canals. X 28,000.

Fig. 15.83. Basal surface of the visceral endoderm. The close association of a large lipid droplet with the concave basal surface of the nucleus is illustrated here. Two typical clusters of Golgi membranes (g) are located near the basal region of the nucleus. Mitochondria and elements of the endoplasmic reticulum are abundant. X 13.200.

XII. References

AccoNci, G. 1912. Di alcune fini particolarita di struttura della mucosa uterina, della decidua e dell 'uovo. Folia Gynec, 7, 25.

Altshuler, C. H., and Angevine, D. M. 1949. Histochemical studies on the pathogenesis of fibrinoid. Am. J. Path., 25, 1061-1077.

Altshuler, C. H., and Angevine, D. M. 1951. Acid mucopolysaccharide in degenerative disease of connective tissue, with special reference to serous inflammation. Am. J. Path., 27, 141156.

Amoroso, E. C. 1947. The vascular relations in the placenta of the sow. Proc. Physiol. Soc, 18, 1 ; 47.

Amoroso, E. C. 1951. The interaction of the trophoblast and endometrium at the time of implantation in the sheep. J. Anat. (Proc), 85, 428-429.

Amoroso, E. C. 1952. Placentation. In Marshall's Physiology oj Reproduction, 3rd ed., A. S. Parkes, Ed., Vol. 2, pp. 127-311. New York: Longmans, Green and Company.

Amoroso, E. C. 1960. Comparative aspects of the hormonal functions. In The Placenta and Fetal Membranes, C. A. Villee, Ed., pp. 3-28. Baltimore: The Williams & Wilkins Company.

Anderson, J. W. 1959. The placental barrier to gamma-globulins in the rat. Am. J. Anat., 104, 403-430.

Arey, L. B. 1946. Developmental Anatomy, 5th ed. Philadelphia: W. B. Saunders Company.

AsHBEL, R., AND Hertig, A. T. 1952. Histochemical demonstration of ketosteroids in normal, abnormal and neoplastic placentae. J. Nat. Cancer Inst., 13, 221-222.

AsHBEL, R., AND Seligman, A. M. 1949. A new reagent for the histochemical demonstration of active carbonyl groups. A new method for staining ketonic steroids. Endocrinology, 44, 565-583.

Baker, J. R. 1946. The histochemical recognition of hpine. Quart. J. Microscop. Sc, 87, 441-470.

Baker, B. L., Hook, S. J., and Severinghaus, A. E. 1944. The cytological structure of the human chorionic villus and decidual parietalis. Am. J. Anat., 74, 291-325.

Banghan, D. R., Hobbs, K. R., and Terry, R. J. 1958. Selective placental transfer of serumproteins in the rhesus. Lancet, 2, 351.

Barcroft, J. 1947. Researches on Pre-natal Life. Springfield, 111.: Charles C Thomas.

B.-^rgmann, W., and Knoop, A. 1959. Elektronenmikroskopische Untersuchungen an Plazentarzotten des Menschen (Bemerkungen zum Synzytiumproblem). Ztschr. Zellforsch., 50, 472-493.

Barker, W. L. 1951. A cytochemical study of lipids in sows' ovaries during the estrous cycle. Endocrinology, 48, 772-785.

B.\RRNETT, R. J., AND Seligman, A. M. 1952. Histochemical demonstration of protein-bound sulfhydryl groups. Science, 116, 323-327.

Bennett, H. S. 1940. The life history and secretion of the cells of the adrenal corte.x of the cat. Am. J. Anat., 67, 151-227.

Bensley, S. H. 1934. On the presence, properties and distribution of the intercellular ground substance of loose connective tissue. Anat. Rec, 60, 93-109.

Bergeron, J. A., and Singer, M. 1958. Metachromasy: an experimental and theoretical evaluation. J. Biophys. Biochem. Cytol., 4, 433-458.

Bernard, C. 1859. Recherches sur I'origine de la glycogenie dans la vie embryonnaire ; nouvelle fonction du placenta. Compt. rend. Soc. biol., 11, 101-7.

Bettelheim-Jevons, F. R. 1958. Protein-carbohvdrate complexes. Advances Protein Chem., 13, 35-105.

Bienenfeld, B. 1912. Beitrag zur Kenntnis des Lipoidgehaltes der Placenta. Biochem. Ztschr., 43, 245-255.

Bjorkman, N., and Bloom, G. 1957. On the fine structure of the foetal-maternal junction in the bovine placentome. Ztschr. Zellforsch., 45, 649659.

BoE, F. 1953. Studies on the vascularization of the human placenta. Acta obst. et gynec. scandinav., Suppl., 32, 57.

Boyd, J. D., and Hughes, A. F. W. 1954. Observations on human chorionic villi using the electron microscope. J. Anat., 88, 356-362.

Bracket, J. 1953. The use of basic dyes and ribonuclease for cytochemical detection of ribonucleic acid. Quart. J. Microscop. Sc, 94, 1-10.

Brambell, F. W. R., Hemmings, W. A., and RowL.\NDS, W. T. 1948. The passage of antibodies from the maternal circulation into the embryo in rabbits. Proc. Roy. Soc, London, ser. B, 135, 390-403.

Brambell, F. W. R., Hemmings, W. A., Henderson, M., .\ndRo\vl.\nds, W.T. 1950. The selective admission of antibodies to the foetus by the yolk-sac splanchnopleur in rabbits. Proc. Roy. Soc, London, ser. B, 137, 239-252.

Br.^mbell, F. W. R., Hemmings, G. P., Hemmings, W. A., Henderson, M., .■^nd Rowlands, W. T. 1951. The route by which antibodies enter the circulation after injection of immune serum into the exocoel of foetal rabbits. Proc. Roy. Soc, London, ser. B, 138, 188-204.

Brambell, F. W. R., and Halliday, R. 1956. The route by which passive immunity is transmitted from mother to foetus in the rat. Proc. Roy. Soc, London, ser. B, 145, 179-185.

Br-^mbell, F. W. R. 1957. The development of fetal immunity. Macy Foundation Conferences on Gestation, 4, 143-201.

Bremer, J. L. 1916. The interrelations of the mesonephros, kidney, and placenta in different classes of mammals. Am. J. Anat., 19, 179-209.

Bridgman, J. 1948a. A morphological study of the development of the placenta of the rat. I. An outline of the development of the placenta of the white rat. J. MorphoL, 83, 61-85.

Bridgman, J. 1948b. A morphological study of the development of the placenta of the rat.

II. An histological and cytological study of the development of the chorioallantoic placenta of the white rat. J. MorphoL, 83, 195-223.

Bruner, J. A. 1951. Distribution of chorionic gonadotropin in mother and fetus at various stages of pregnancy. J. Clin. Endocrinol., 11, 360-374.

Bruner, J. A., and Witschi, E. 1947. Distribution of chorionic gonadotropin. Anat. Rec. (Abstr.), 99, 662.

Bulmer, D., and Dickson, A. D. 1960. Observations on the carbohydrate materials in the rat placenta. J. Anat., 94, 46-58.

BuNO, W., and Curi, C. R. 1945. Distribucion de la fosfatase en la placenta humana. Ciencia, 6, 59.

Caffier, p. 1929. Die proteolytische Fahigkeit von Ei und Eibett. Zentralbl. Gynak., 53, 24102425.

Catchpole, H. R. 1949. Distribution of glycoprotein hormones in the anterior pituitary gland of the rat. J. Endocrinol., 6, 218-225.

Chipman, W. W. 1902. Observations on the placenta of the rabbit, with special reference to the presence of glycogen, fat and iron. Studies Royal Victoria Hosp., Montreal, 1, 1.

Christoffersen, a. K. 1934. La superficie des villosites choriales du placenta a la fin de la grossesse (etude d'histologie quantitative). Compt. rend. Soc biol., 117, 641-644.

Clegg, M. T., Boda, J. M., and Cole, H. H. 1954. The endometrial cups and allantochorionic pouches in the mare with emphasis on the source of equine gonadotrophin. Endocrinology, 54, 448-463.

Cole, H. H., and Goss, H. 1943. The source of equine gonadotrophin. In Essays in Biology. Berkeley and Los Angeles : University of California Press.

Corey, E. L. 1932. Placental permeability to insulin in the albino rat. Physiol. ZooL, 5, 36-48.

CowiE, A. T. 1948. Pregnancy diagnosis tests: a review. Commonwealth Agricultural Bureau, Joint Publication, No 13, Great Britain.

Cunningham, R. S. 1920. Studies in placental permeability. I. The differential resistance to certain solutions offered by the placenta in the cat. Am. J. Physiol., 53, 439-456.

Cunningham, R. S. 1922. Studies in placental permeability. II. LocaHzation of certain physiological activities in the chorionic ectoderm in the cat. Am. J. Physiol., 60, 44&-460.

Davies, J. 1956. Histochemistry of the rabbit placenta. J. Anat., 90, 135-142.

Deane, H. W. 1952. Histochemical observations on the ovary and oviduct of the albino rat during the estrous cycle. Am. J. Anat., 91, 363414.

Deane, H. W., and Creep, R. 0. 1946. A morphological and histochemical study of the rat's adrenal cortex after hypophysectomy with comments on the liver. Am. J. Anat., 79, 117145.

Deane, H. W., Lobel, B. L., Driks, E. C, and Rubin, B. L. 1960. Etudes supplementaires de I'activite de I'enzyme steroide-3/3-ol deshydrogenase des organes reproductifs de la rate. Proceedings of the 1st International Congress of Histochemistry and Cytochemistry, p. 111. Paris: Pergamon Press.

De.ane, H. W., and Morse, A. 1948. The cytological distribution of ascorbic acid in the adrenal cortex of the rat under normal and experimental conditions. Anat. Rec, 100, 127-141.

Deane, H. W., and Seligman, A. M. 1953. Evaluation of procedures for the cytological localization of ketosteroids. Vitamins & Hormones, 11, 173.

De.^ne, H. W., Shaw, J. H., and Creep, R. O. 1948. The effect of altered sodium or potassium intake on the width and cytochemistry of the zona glomerulosa of the rat's adrenal cortex. Endocrinology, 43, 133-153.

De Duve, C. 1959. Lysosomes, a new group of cytoplasmic particles. In Subcellular Particles, T. Hayashi, Ed., pp. 128-159. New York: The Ronald Press Company.

Dempsey, E. W. 1953. Electron microscopy of the visceral volk-sac epithelium of the guinea pig. Am. J. Anat., 93, 331-363.

Dempsey, E. W., and Bassett, D. L. 1943. Observations on the fluorescence, birefringence and histochemistry of the rat ovary during the reproductive cycle. Endocrinology, 33, 384-401.

Dempsey, E. W., and Wislocki, G. B. 1944. Observations on some histochemical reactions in the human placenta, with special reference to the significance of the lipoids, glycogen and iron. Endocrinology, 35, 409-429.

Dempsey, E. W., and Wislocki, G. B. 1945. Histochemical reactions associated with basophilia and acidophilia in the placenta and pituitary gland. Am. J. Anat., 76, 277-301.

Dempsey, E. W., and Wislocki, G. B. 1946. Histochemical contributions to physiology. Physiol. Rev., 26, 1-27.

Dempsey, E. W., and Wislocki, G. B. 1947. Further observations on the distribution of phosphatases in mammalian placentas. Am. J. Anat., 80, 1-33.

Dempsey, E. W., Wislocki, G. B., and Amoroso, E. C. 1955. Electron mici'oscopy of the pig's placenta, with especial reference to the cell membranes of the endometrium and chorion. Am. J. Anat., 96, 65-102.

Dempsey, E. W., and Wislocki, G. B. 1958. Electron microscopic observations on the placenta of the cat. J. Biophys. Biochcm. Cytol., 2, 743754.

DeiMpsey, E. W., Bunting, H., Singer, M., and Wislocki, G. B. 1947. The dye-binding capacity and other chemo-histological properties of mammalian mucopolysaccharides. Anat. Rec, 98, 417-429.

Desclin, L. 1940. Detection de substances pentosenucleiques dans les cellules du lobe anterieur de I'hvpophyse du rat et du cobave. Compt. rend. Soc. bioL, 133, 457-459.

Dodds, G. S. 1924. The area of the chorionic villi in the full term placenta. Anat. Rec, 24, 287294.

Doyle, W. L. 1950. The distribution of phosphatases in the rabbit appendix after x-irradiation. Am. J. Anat., 87, 79-117.

DuBois, a. M., and Ducommun, S. 1955. Variations de la teneur en glycogene et en graisse du placenta de cobaye, au cours de la gestation. Acta Anat., 25, 398.

Duval, M. 1892. Le Placenta des Rongeurs, Felix Alcan, Ed. Paris: Ancienne Libraire Germer Bailliere et Cie.

Enders, a. C. 1960. Electron microscopic observations on the villous haemochorial placenta of the nine-banded armadillo {Dasypus novevicinctus) . J. Anat., 94, 205-215.

Everett, J. W. 1935. Morphological and physiological studies of the placenta of the albino rat. J. Exper. ZooL. 70, 243-285.

Fawcett, D. W., Wislocki, G. W., and Waldo, C. M. 1947. The development of mouse ova in the anterior chamber of the eye and in the abdominal cavity. Am. J. Anat., 81, 413-443.

Flexner, L. B., and Pohl, H. A. 1941a. The transfer of radioactive sodium across the placenta of the white rat. J. Cell. Comp. Physiol., 18, 49-59.

Flexner, L. B., and Pohl, H. A. 1941b. The transfer of radioactive sodium across the placenta of the rabbit. Am. J. Physiol., 134, 344349.

Flexner, L. B., and Pohl, H. A. 1941c. Transfer of radioactive sodium across the placenta of the guinea pig. Am. J. Physiol., 132, 594-611.

Flexner, L. B., Cowie, D. B., Hellman, L. M., Wilde, W\S., AND VosBURGH, G.J. 1948. The permeability of the human placenta to sodium in normal and abnormal pregnancies and the supply of sodium to the human fetus as determined with radioactive sodium. Am. J. Obst. & Gynec, 55, 469-480.

Friedheim, E. a. H. 1929. Die Ziichtung von menschlichen Chorionepithel in vitro. Ein Beitrag zur Lehre vom Chorionepitheliom. Arch. Path. Anat., 272, 217-244.

FujiMURA, G. 1921. Cytological studies on the internal secretory functions in the human placenta and decidua. J. Morphol., 35, 485.

Gellhorn, a. 1943. Placental transmission of radioactive isotopes. In Proceedings of the Conference on Problems oj Human Fertility, E. T. Engle, Ed., pp. 158-166. Menasha, Wise: Banta Publishing Company.

Gellhorn, A., and Flexner, L. B. 1942. Transfer of water across the placenta of the guinea pig. Am. J. Pliysiol., 136, 750-756.

Gellhorn, A., Flexner, L, B., and Pohl, H. A. 1941. The transfer of radioactive sodium across the placenta of the sow. J. Cell. Comp. Physiol., 18, 393-400.

Gersh, I., AND Catchpole, H. R. 1949. The organization of ground substance and basement membrane and its significance in tissue injury, disease and growth. Am. J. Anat., 85, 457-521.

Gey, G. 0., Jones, G. E., and Hellman, L. M. 1938. The production of a gonadotropic substance (prolan) by placental cells in tissue culture. Science, 88, 306-307.

Goldmann, E. 1912. Die ilussere und innere Sekretion des gesimden und kranken Organismus im Lichte der "vitalen Fiirbung," Teil II. Beitr. klin. Chir., 78, 1-108.

GoMORi, G. 1941a. The distribution of phosphatase in normal organs and tissues. J. Cell. Comp. Physiol., 17, 71-84.

GoMORi, G. 1941b. Distribution of acid phosphatase in the tissues under normal and under pathological conditions. Arch. Path., 32, 189199.

GoMORi, G. 1945. The microtechnical demonstration of sites of lipase activity. Proc. Soc. Exper. Biol. & Med., 58, 362-364. "

GoMORi, G. 1950. Methods of study for tissue lipase. In Menstruation and Its Disorders, E. T. Engle, Ed., pp. 52-60. Springfield, 111.: Charles C Thomas.

Grafenberg, E. 1909a. Der Antitrypsingehalt des mutterlichen Blutserums wahrend der Schwangerschaft als Reaktion auf tryptische Pinfliisse der Eioberfliiche. Miinchen. med. Wchnschr.. 56, 702-704.

Grafenberg, E. 1909b. Beitrage zur Physiologie der Eieinbettung. Ztschr. Gebuitsh. Gynak., 65, 1-35.

Green, H., .and Meyerhof, 0. 1952. Synthetic action of pliosphatase. III. Transphosphorylation with intestinal and semen phosphatase. J. Biol. Chem., 197, 347-364.

Grosser, O. 1909. Vergleichende Anatomie und Entwicklungsgeschichte der Eihdute und der Placenta. Wien-Leipzig : W. Braumiiller.

Grosser. 0. 1925a. tJber Fibrin und Fibrinoid in der Placenta. Ztschr. Anat., 76, 304-319.

Grosser, 0. 1925b. Entwicklungsgeschichte des Menschen von der Keimzelle bis zur Ausbildung der iiusseren Korperform unter Beriicksichtigung ihrer vergleichend-entwicklungsgeschichtlichen Grundlagen. Vergleichende und menschliche Placentationslehre. In Biologie und Patholugie des Weibes, J. Halban and L. Seitz, Eds., Vol. 6, p. 1. Berhn: Urban and Schwarzenberg.

Grosser, 0. 1948. Gnindis der Entwicklungsgeschichte des Menschen. Berlin: Springer-Verlag.

HAMiLTON, W. J., AND BoYD, J. D. 1951. Observations on the human placenta. Proc. Roy. Soc. Med., 44, 489.

Hamilton, W. J., and Boyd, J. D. 1960. Development of the human jilacenta in the first three months of gestation. J. Anat., 94, 297328.

Hamilton, W. J., Boyd, J. D., and Mossman, H. W. 1952. Human Embryology, 2nd Ed. Baltimore : The WiUiams & Wilkins Company.

Hard, W. L. 1946. A histochemical and quantitative study of phosphatase in the placenta and fetal membranes of the guinea pig. Am. J. Anat., 78, 47-77.

Hellman, L. M., Flexner, L. B., Wilde, W. W., VosBURGH, G. J., AND Proctor, N. K. 1948. The permeability of the human placenta to water and the supply of water to the human foetus as determined with deuterium oxide. Am. J. Obst. & Gynec. 56, 861-868.

Herman, E., and Deane, H. W. 1953. A comparison of the localization of alkaline glycerophosphatase, as demonstrated by the GomoriTakamatsu method, in frozen and in paraffin sections. J. Cell Comp. Physiol., 41, 201-224.

Hertig, a. T., and Rock, J. 1941. Two human ova of the previllous stage, having an ovulation age of about 11 and 12 days respectively. Contr. Embryol., Carnegie Inst. Washington, 29, 127156.

Hertig, A. T., and Rock, J. 1945. Two human ova of the previllous stage, having a developmental age of about 7 and 9 days respectively. Contr. Embryol., Carnegie Inst. Washington, 31, 65-94.

Hertig, A. T., Rock, J., .and Adams, E. C. 1956. A description of 34 human ova within the first 17 days of development. Am. J. Anat., 98, 435494.

Hill, J. P. 1932. The developmental history of the primates (Croonian Lecture). Philos. Tr. Roy. Soc, London, 221, 45-178.

HoFBAUER, J. 1905. Grundzuge einer Biologie der menschlichcn Plazenta. Wien-Leipzig: W. Braumiiller.

HoFBAUER, J. 1925. The function of the Hofbauer cells of the chorionic villus particularly in relation to acute infection and svphilis. Am. J. Obst. & Gynec, 10, 1-14.

HoLZAEPFEL, J. H., AND Barnes, A. C. 1947. Placental metabohsm of vitamin C. II. Histochemical analysis. Am. J. Obst. & Gvnec, 53, 864-868.

HoTCHKiss, R. D. 1948. A microchemical reaction resulting in the staining of polysaccharide structures in fixed tissue preparations. Arch. Biochem., 16, 131-141.

Huggett, a. St. G. 1944. Influence of diet on pregnancy. I. The role of the placenta in foetal nutrition. Proc Nutrition Soc, 1, 227-237.

Huggett, A. St. G., and H.ammond, J. 1952. Physiology of the placenta. In Marshall's Physiology of Reproduction, 3rd ed., A. S. Parkes, Ed., pp. 312-397. New York: Longmans, Green and Company.

Irving, F. C, AND Hertig, A. T. 1937. A study of placenta accreta. Surg. Gvnec. & Obst., 64, 178200.

Jones, G. E., Gey, G. O., and Gey, M. K. 1943. Hormone production by placental cells maintained in continuous culture. Bull. Johns Hopkins Hosp., 72, 26-38.

K.arnovsky, M. L., and Deane, H. W. 1954. Alterations of adrenal cortical lipides by formalin fixation as determined chemically and histochemically. J. Histochem. (Abstr.), 2, 478.

Karnovsky, M. L., and Deane, H. W. 1955. Aldehyde formation in the lipide droplets of the adrenal cortex during fixation, as demonstrated chemically and histochemically. J. Histochem., 3, 85-102.

Krehbiel, R. H. 1937. Cytological studies of the decidual reaction in tlie rat during early pregnancy and in the production of deciduomata. Physiol. ZooL, 10, 212-234.

Kuttner, a., and R.atner, B. 1923. The importance of colostrum to the newborn infant. Am. J. Dis. Child., 25, 413-434.

Leblond, C. p. 1950. Distribution of periodic acid-reacti\'e carbohydrates in the adult rat. Am. J. Anat., 86, 1-49.

Leduc, E. H., and Dempsey, E. W. 1951. Activation and diffusion as factors influencing the reliability of the histochemical method for alkaline phosphatase. J. Anat., 85, 305-314.

Lehner, J. 1914. Dottersack der weLssen Maus. Anat. Anz., 46, 182-186.

Lell, W. a.. Liber, K. E., .and Snyder, F. F. 1932. Quantitative study of placental transmission and the permeability of foetal membranes at ^'arious stages of pregnancy. Am. J. Physiol., 100,21-31.

Levy, H., Deane, H. W., and Rubin, B. L. 1959. Visualization of steroid-3/3-ol dehydrogenase activity in tissues of intact and hypophysectomized rats. Endocrinology, 65, 932-943.

Lewis, W. H. 1924. Hofbauer cells (clasmatocytes) of the human chorionic villus. Bull. Johns Hopkins Hosp., 35, 183-185.

Lewis, W. H. 1931. Locomotion of lymphocytes. Bull. Johns Hopkins Hosp., 49, 29-36.

LocHHE.'iD, J., .\ND Cramer, W. 1908. The glycogenic changes in the placenta and the foetus of the pregnant rabbit: a contribution to the chemistry of growth. Proc. Roy. Soc, London, ser. B, 80, 263-284.

LOVELAND, G., MaURER, E. E., AND SnYDER, F. F.

1931. The diminution of the glycogen store of the rabbit's placenta during the last third of pregancy. Anat. Rec, 59, 265-275.

Low, F. N. 1953. Pulmonary alveolar epithelium of laboratory mammals and man. Anat. Rec, 117, 241-263.

Luse, S. a. 1957. The morphological manifestations of uptake of materials by the yolk sac of the pregnant rabbit. Macy Foundation Conferences on Gestation, 4, 115-142.

Luse, S., Davies, J., and Smith, M. 1959. Electron microscopy of experimental inclusions in cytoplasm and nuclei of yolk sac cells. Fed. Proc, 18, 491.

LxjSE, S., Davies, J., .wd Cl.ark, S. L., Jr. 1959. Electron microscopy of nuclear inclusions. Am. J. Path., 35, 686.

Mall, F. P., and Meyer, A. W. 1921. Hofbauer cells in normal and pathologic conceptuses. Contr. Embryol., Carnegie Inst. Washington, 12, 301.

Martin, B. F., and J.\coby, F. 1949. Diffusion phenomena complicating the histochemical reaction for alkaline phosphatase. J. Anat., 83, 351-363

Mayersbach, H. 1958. Zur Frage des Proteiniiberganges von der Mutter zum Foeten. I. Befunde an Rattan am Ende der Schwangerschaft. Ztschr. Zellforsch. mikroskop. Anat., 48, 479-504.

McKay, D. G., Ad.ams, E. C, Hertig, A. T., and Danziger, S. 1955a. Histochemical horizons in human embiyos. I. Five millimeter embryo — Streeter horizon XIII. Anat. Rec, 122, 125151.

McK.\Y, D. G., Adams, E. C, Hertig, A. T., and Danziger, S. 1955b. Histochemical horizons in human embrj^os. II. Six and seven millimeter embryos — Streeter horizon XIV. Anat. Rec, 126, 433-463.

McManus, J. F. A. 1945. Granules of the human polymorphonuclear leucocyte. Nature, London, 156, 173.

Michaelis, L. 1947. The nature of the interaction of nucleic acids and nuclei with basic dj^estuffs. Cold Spring Harbor Symposia Quant. Biol., 12, 131.

Mitchell, A. J., and Wislocki, G. B. 1944. Selective staining of glycogen by ammoniacal silver nitrate: a new method. Anat. Rec, 90, 261-266.

Moog, F., and Wenger, E. L. 1952. The occurrence of a neutral mucopolysaccharide at sites of high alkaline phosphatase activity. Am. J. Anat., 90, 339-377.

MossMAN, H. W. 1926. The rabbit placenta and the problem of placental transmission. Am. J. Anat., 37, 433-497.

MossMAN, H. W. 1937. Comparative morphogenesis of the foetal membranes and accessory uterine structures. Contr. Embryol., Carnegie Inst. Washington, 26, 129.

Nachlas, M. M., and Seligman, A. M. 1949. The histochemical demonstration of esterase. J. Nat. Cancer Inst., 9, 415-425.

Nataf, B. 1953. Evolution des activites phosphatasiciue alcaline et arginasique du placenta de Cobaye a differents stades de la gestation. Compt. rend Soc. biol., 147, 1564-1565.

Needham, J. 1931. Chemical Embryology. Cambridge, England: The Macmillan Company.

Neum.ann, E. 1880. Die Picrocarminfarbung und ihre Anwendung auf die Entzundiingslehre. Arch, mikroskop. Anat., 18, 130.

Newton, W. H. 1938. Hormones and the placenta. Physiol. Rev., 18, 419-446.

Nic.'VNder, L. 1951. The plasmal reaction of Feulgen and Voit with special reference to the adrenal body. Acta Anat., 12, 174-197.

Ord,M.G., .and Thompson, R.H.S. 1950. Nature of placental cholinesterase. Nature, London, 165, 927-928.

Ortmann, R. 1949. Uber Kernsekretion, Kolloidund Vakuolenbildung in Beziehung zum Nukleinsiiuregehalt in Trophoblast-riesenzellen der menschlichen Placenta. Ztschr. Zellforsch. mikrobiol. Anat., 34, 562-583.

Padykula, H. a. 1952. The locahzation of succinic dehydrogenase in tissue sections of the rat. Am. J. Anat., 91, 107-132.

Padykula, H. A., and Leduc, E. H. 1955. Etudes Histochimiques du foie prenatal et postnatal. VI° Congres Federatif Internat. d'Anatomie, Paris.

Padykula, H. A. 1958a. A histochemical and quantitative study of enzymes of the rat's placenta. J. Anat., 92, 118-129.

Padykula, H. A. 1958b. Histochemistry of the rat's placenta. In Environmental Influences on Prenatal Development, Beatrice Mintz, Ed., pp. 34-38. Chicago: University of Chicago Press.

Palade, G. E. 1951. Intracellular locahzation of acid phosphatase. A comparative study of biochemical and histochemical methods. J. Exper. Med., 94, 535.

Palade, G., AND SiEKEViTZ, P. 1956. Liver microsomes. An integrated morphological and biochemical study. J. Biophys. Biochem. Cytol., 2, 171-200.

PALAY, S. L., -AND Karlin, L. J. 1959a. An electron microscopic study of the intestinal villus. I. The fasting animal. J. Biophys. Biochem. Cytol., 5, 363-372.

PALAY, S. L., AND Karlin, L. J. 1959b. An electron microscopic study of the intestinal villus. II. The pathway of fat absorption. J. Biophys. Biochem, Cytol., 5, 373-384.

Pease, D. 1955. The fine structure of the kidney seen bv the electron microscope. J. Histochem., 3, 295-308.

Pearse, a. G. 1949. The cytochemical demonstration of gonadotropic hormone in the human anterior hypophysis. J. Path. & Bact., 61, 195202.

Pe.\rse, a. G. E. 1960. Histochemistry, Theoretical and Applied. Boston: Little, Brown and Company.

PoHL, H. A., AND Flexnor, L. B. 1941. Transfer of radioactive sodium across the placenta of the cat. J. Biol. Chem., 139, 163.

Pollock, W. F. 1942. Histochemical studies of the interstitial cells of the testis. Anat. Rec, 84, 23-29.

Popj.'VK, G. 1947. Synthesis of phosi^holipids in foetus. Nature. London, 160, 841-842.

Popj.AK, G., AND Beeckmans, M. L. 1950. Are phospholipins transmitted through the placenta? Biocliem. J., 46, 99-103.

Pritchard, J. J. 1947. The distribution of alkaline phosphatase in the pregnant uterus of the rat. J. Anat., 81, 352-364.

PuRVES, H. D., aND Grie.sb.ach. W. E. 1951a. The site of thyrotrophin and gonadotrophin production in the rat pituitary studied by McManusHotchkiss staining for glycoprotein. Endocrinology, 49, 244-264.

PuRVES, H. D., AND GRIE.SBACH, W. E. 1951b. The significance of the Gomori staining of the basophils of the rat pituitary. Endocrinology, 49, 652-662.

Ramsey, E. M. 1949. The \ascular pattern of the endometrium of the pregnant rhesus monkey (Macaca mulatta). Contr. Embryol., Carnegie Inst. Washington, 33, 113-147.

Ramsey, E. M. 1954. Circulation in the maternal placenta of primates. Am. J. Obst. & Gynec. 67, 1-14.

Ratner, B. 1943. The passage of native proteins through the placenta. In Proceedings of the Conference on Problems of Human Fertility, E. T. Engle, Ed., pp. 167-182. Menasha, Wise: Banta Publishing Company.

R.atner, B., Jackson, H. C., and Gruehl, H. L. 1927. Transmission of protein hypersensitiveness from mother to offspring. I. Criticiue of placental permeability. J. Immunol., 14, 249265.

R.WIN, H. a., Zacks, S. I., .and Seligman, A. 1953. The histochemical localization of acetylcholinesterase in nervous tissue. J. Pharmacol, & Exper. Therap., 107, 37-53.

Reale, E., -AND Piping, G. 1957. Les variations de la repartition histochimique de la succinodehydrogenase au cours de la placental ion chorion-allantoidienne et vitelline etudiees dans le Mus musculus albinus. Compt. rend. A. Anat., XLIV, Reunion-Leyde.

Reale, E., and Piping, G. 1959. La distribuzione della succinodeidrogenasi nella placenta di alcuni mammiferi. Studio istochimico. Arch, ital. Embriol., 64, 318-356.

Rech, W. 1924. L^ntersuchungen iiber die Grosse der Zottenoberflache der men.scldichen Plazenta. Ztschr. Biol., 80, 349-358.

Rock, J., and Hertig, A. T. 1948. The human conceptus during the first two weeks of gestation. Am. J. Obst. & Gynec, 55, 6-17.

RoCKENSCH.AUB, A. 1952. Eigenfluoreszenze und Hormonbildung in der Plazenta. Mikroskopie, 7,56.

Rodglfo, a. 1934. A study of the permeability of the placenta of the rabbit to antibodies. J. Exper. Zool., 68, 215.

Rowlands, I. W. 1947. Anterior pituitary -like hormones. J. Endocrinol., 5, xx-xxiii.

Rubin, I. C. 1941. Two unusual cases of chorioepithehoma. Am. J. Obst. & Gvnec, 41, 10631068.

Runner, M. 1947. The development of mouse eggs in the anterior chamber of the eve. Anat. Rec, 98, 1-17.

ScHONiG, A. 1929. iiber den Kalktransport von Mutter und Kind und iiber Kalkablagerungen in der Plazenta. Ztschr. Geburt.sh. Gynak., 94, 451-465.

Schiebler, T. H., and Knoop, A. 1959. Histocliemische und Elektronenmikroskopische Untersuchungen an der Rattenplazenta. Ztschr. Zellfonsch., 50, 494-552.

Scott. G. H. 1933. A critical study and review of the method of microincineration. Protoplasma, 20, 133-151.

Seligman, A. M., and Ashbel, R. 1951. Histochemical demonstration of ketosteroids in virilizing tumors of the adrenal cortex. Endocrinology, 49, 110-126.

Seligman, A. M., and Rutenburg, A. M. 1951. The histochemical demonstration of succinic dehydrogenase. Science, 113, 317-320.

Seligman, A. M., Ashbel, R., and Cohen, R. F. 1951. Histochemical demon.stration of active carbonyl groups in normal and neoplastic tissues. J. Nat. Cancer. Inst., 12, 226.

Sengupta, B. 1935. Plazenta in der Gewebekultur. Arch, exper. Zellforsch., 17, 281.

Severinghaus, a. E. 1932. A cytological technicjue for the study of the anterior lobe of the hypophysis. Anat. Rec, 53, 1-5.

Singer, M. 1952. Factors which control the staining of ti.ssue sections with acid and basic dyes. Internal. Rev. Cytol., 1, 211-255.

Singer, M., and Morrison, P. R. 1948. The influence of pH, dye and salt concentration on the dye binding of modified and unmodified fibrin. J. Biol. Chem., 175, 133-145.

Singer, M., and Wislocki, G. B. 1948. The affinity of .syncytium, fibrin and fibrinoid of the human placenta for acid and basic dyes under controlled conditions of staining. Anat. Rec, 102, 175-193.

Smith, G. V., and Smith, O. W. 1948. Internal secretions and toxemia of late pregnancy. Physiol. Rev., 28, 1-22.

Snyder, F. F. 1943. Placental transmission in relation to the stage of pregnancy. In Proceedings of the Conference on Problems of Human Fertility, E. T. Engle, Ed., pp. 144-157. Menasha, Wise: Banta Publishing Company.

SoKOLOFF, L., MuND, A., AND Kantor, T. G. 1951. The affinity of fibrinoid substances for acid dyes. Am. J. Path., 27, 1037-1045.

Spanner, R. 1935a. Beitrag zur Kenntnis des Baues der Plazentarsepten, gleichzeitig ein Versu. 11 zur Douiung ihrer Eustchung. Morphol. Jalub.. 75, 374.

Spanner. R. 1935b. Miitteilicher iind kindlicher Kreislauf der menschlichen Placenta imd .seine Stiombahnen. Ztschr. Anat., 105, 163-242.

Spanner, R. 1940. Betrachtungen zum Placentakreislauf des Menschen. Zentralbl. Gynak., 64, 2002-2011.

Starck, D. 1945-1950. Vergleichende Entwicklungeschichte der Wirbeltiere. Fort.schr. Zool., 9, 249-367.

Stewart. H. L., Jr., Sano. M. E., and Montgomery, T. L. 1948. Hormone .secretion by human placenta grown in tissue culture. J. Clin. Endocrinol.. 8, 175-188.

Stieve, H. 1940. Die Entwicklung und der Bau der menschlichen Placenta. I. Zotten, Trophoblastinseln und Scheidewiinde in der ersten Hjilfte der Schwangenschaft. Ztschr. mikrcanat. Forsch., 48, 287-358.

Stieve, H. 1941. Bemerkungen uber den Blutkreislauf in der Placenta des Menschen. Zentralbl. Gynak., 65, 370.

Stieve, H. 1944. Vergleichendanatomisclie Untersuchungen iiber das Zottenraunigitter in der Primatenplacenta. Ztschr. mikro.-anat. Forsch., 54, 480-542.

Stieve, H., and von der Heide. I. 1941. Uber die Entwicklung der Sei^ten in der menschlichen Plazenta. Anat. Anz., 92, 1-16.

SzENDi, B. 1933. Die Wege des Glykogens durch die hamochoriale Placenta. Ztschr. Anat., 101, 791-798.

SzENDi, B. 1934. Experimentelle Studicn iil)er die entgiftende Funktion der D(H-idua. Arch. Gynak., 157, 389-399.

Telkka, A., and Lehto, L. 1954. Histochemically demonstrable succinic dehydrogenase activity of placental tissue.s. Ann. Med. exper. Biol. Fenniae, 32, 292-296.

Tenney, B. 1936a. A clinical and pathological study of one hundred and fifty ca.ses of tubal pregnancy. New England J. Med.. 214, 773776.

Tenney, B. 1936b. Syncytial degeneration in normal and pathologic placentas. Am. J. Obst. & Gynec, 31, 1024-1028.

Tenney, B., and Parker, F. 1939. Hydatidiform mole and chorionepithelioma. New England J. Med., 221, 598-601.

Tenney, B., .\nd P.arker, F. 1940. The placenta in toxemia of pregnancy. Am. J. Obst. & Gynec, 39, 1000-1005.

Tonutti. E., and Pl.\te, E. 1937. Uber das Vitamin C in der menschlichen Placenta. Arch. Gynak., 164, 385-397.

ToRDA, C. 1942. Choline esterase content of tissues without innervation (the placenta). Proc. Soc. Exper. Biol. & Med., 51, 398-400.

TUCHMANN-DUPLE.SSIS, H., .AND BORTOLAMI, R. 1954. Les constituants histochimiques tie I'allentoplacenta du lapin. Bull. Micro. AppL, 4, 73-88.

Venning, E. 1948. The excretion of various hormone metabolities in normal pregnancy. In The Normal and Pathological Physiology of Pregnancy. Proceedings of the Conference of the Committee on Human Reproduction of the National Research Council, Baltimore: The Williams & Wilkins Company.

Villee, C. a. 1953. Regulation of blood gluco.se in the human fetus. J. Apjjl. Phy.siol., 5, 437.

Watanabe, H. 1923. Uber die Lipoidsubstanzen der Placenta in verschiedenen Schwangerschaftsmonaten mit besonderer Beriicksichtigung ihrer Mengenverhiiltnisse. J. Biochem. Tokyo, 2, 369.

Weekes, H. C. 1935. A review of placentation among reptiles with particular regard to function and evolution of the placenta. Proc. Zool. Soc, London, 2, 625.

WEIS.S. L. P., AND Fawcett, D. W. 1953. Cytochemical observations on chicken monocytes, macrophages and giant cells in tissue culture. J. Histochem. 1, 47-65.

Welss, L. p., Tsou, K. C, and Seligman, A. M. 1954. Histochemical demonstration of proteinbound amino grou]5s. J. Histochem. 2, 29-49.

Wen, I. C, Chang, H. C, .and Wong, A. 1936. Studies on tissue acetylcholine in cytological considerations of the chorionic villous epithelium of the human i)lacenta. Chino.^e J. Physiol. 10, 559.

WiMS.ATT, W. A. 1948. The nature and distribution in the placenta of the bat {Myotis lucifugrts lucifugus), with observations of the mitochondria and Golgi apparatus. Am. J. Anat., 82, 393-467.

Wi.MSATT. W. A. 1949. Cytochemical observations on the fetal membranes and placenta of the bat, Myotis Incifngus lucifvgus. Am. J. Anat., 84, 63-141.

WiMsATT. \V. .\. 1950. Xew histological observations on the placenta of the sheep. Am. J. Anat., 87, 391-457.

WiMSATT, W. A. 1951. Ob.servations on the morphogenesis, cytochemistry and significance of the binucleate giant cells of the placenta of ruminants. Am. J. Anat., 89, 233-281.

WiMSATT, W. A., AND Wi.SLocKi, G. B. 1947. The placentation of the American shrews, Blarina brevicauda and Sorex fumeus. Am. J. Anat., 80, 361-435.

WisLOCKi, G. B. 1929. On the placentation of primates, with a consideration of the phylogeny of the placenta. Contr. EmbryoL, Carnegie Inst. Washington, 20, 51.

WisLocKi, G. B. 1950. Saliva-in.soluble glycoproteins, stained by the periodic acid-Schiff procedure, in the placentas of pig, cat, mouse, rat and man. J. Nat. Cancer Inst., 10, 1341.

WiSLOCKi, G. B. 1951. The histology and cytochemistry of the basal plate and septa placentae of the normal human placenta deli\-ered at full term. Anat Rec (Ab.str.), 109, 359.

VViSLOCKi, G. B. 1952. The A.shbel-Seligman reaction for carbonyl groups in the placentas of man and animals. Anat. Rec, 112, 438.

WiSLOCKi, G. B. 1953. Succinic dehydrogenase, esterases and protein-linked sulfhydryl groups in human placenta. Anat. Rec (Abstr.), 115, 380-381.

WisLOCKi, G. B. 1956. Morphological aspects of ageing in the placenta. Ciba Foundation Colloquia Ageing, 2, 105-114.

WisLocKi, G. B., AND Bennett, H. S. 1943. The histology and cytology of the liuman and monkey placenta, with special reference to the trophoblast. Am. J. Anat., 73, 335-449.

WiSLOCKi, G. B., and Dempsey, E. W. 1945. Histochemical reactions of the endometrium in pregnancy. x4m. J. Anat., 77, 365-403.

WisLOCKi, G. B., AND Dempsey, E. W. 1946a. Hi.stochemical reactions in the placenta of the cat. Am. J. Anat., 78, 1-45.

WisLOCKi, G. B., AND Dempsey, E. W. 1946b. Histochemical reactions of the placenta of the pig. Am. J. Anat., 78, 181-225.

WisLocKi, G. B., AND Dempsey, E. W. 1946c. Histochemical age-changes in normal and pathological placental villi hydatidiform mole, eclampsia). Endocrinology, 38, 90-109.

WiSLOcKi, G. B., AND Dempsey, E. W. 1948. The chemical histology of human placenta and decidua with reference to mucoproteins, gh'cogen, lipids and acid phospliata.se. Am. J. Anat., 83, 1-41.

WiSLOCKi,G.B., and Dempsey, E.W. 1955a. Electron microscopy of the human i)lacenta. Anat. Rec, 123, 133-167.

WISLOCKI, G. B., and Dempsey, E. W. 1955b. Electron microscopy of the placenta of the rat. Anat. Rec, 123, 33-63.

WiSLOCKi, G. B., and Padykula, H. a. 1953. Reichert's membrane and the yolk sac of the rat in\estigated by histochemical means. Am. J. Anat., 92, 117-151.

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. Embrvol., Carnegie Inst. Washington, 27, 1.

WiSLOCKI, G. B., AND WlMSATT, W. A. 1947. Chemial cytology of the placenta to two North American shrew.s ( Blarina brevicauda and Sorex jumcK.'^). Am. J. Anat., 81, 269-307.

WisLOCKi, G. B., Bunting, H., and Dempsey, E. W. 1947. Metachromasia in mammalian tissues and its relationship to mucopolysaccharides. Am. J. Anat., 81, 1-37.

WiSLOCKi, G. B., Deane, H. W., and Dempsey, E. W. 1946. The histochemistry of the rodent's placenta. Am. J. Anat., 78, 281-345.

WisLocKi, (i. B.. Dempsey, E. W., and Fawcett, D. W. 1948. Some functional activities of the placental tropliol^last. Obst. & Gvnec. Surv., 3, 604-614.

WisLOCKi, G. B., Weiss, L. P., Burgos, M. H., and Ellis, R. A. 1957. The cytology, histochemistry and electron microscopy of the granular cells of the metrial gland of the gravid rat. J. Anat.. 91, 130-140.

Yamaua, E. 1955. The fine structure of the renal glomerulus of the mouse. J. Biophvs. Biochem. CytoL, 1, 551-566.

Zancla, L. 1912. Contributo alio studio istochiniico del ferro i)lacentare. Folia Gynec, 6, 297.

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)

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

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

---

Cite this page: Hill, M.A. (2024, April 16) Embryology Book - Sex and internal secretions (1961) 15. Retrieved from https://embryology.med.unsw.edu.au/embryology/index.php/Book_-_Sex_and_internal_secretions_(1961)_15

What Links Here?

© Dr Mark Hill 2024, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G