Paper - Histochemical reactions of the endometrium in pregnancy

From Embryology
Embryology - 23 Apr 2024    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Wislocki GB. and Dempsey EW. Histochemical reactions of the endometrium in pregnancy. (1945) Amer. J Anat.

Online Editor  
Mark Hill.jpg
This historic 1945 paper by Wislocki and Dempsey is an early description of the histology of teh pregnant uterus.

Modern Notes: Uterus Development

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

Histochemical Reactions of the Endometrium in Pregnancy

George B. Wislocki and Edward W. Dempsey

Department of Anatomy, Harvard Medical School, Boston, Massachusetts

Five Plates (Twenty figures)

This investigation we supported in part by a grant from the Committee on Research in Endocrinology, National Research Council.


For some time we have been interested in this laboratory in the application of a variety of histochemical procedures to the study of the placenta. Our interest has focussed mainly on the cytology of the trophoblast which constitutes the essential barrier between the mother and fetus. However, as a byproduct of this main interest we have had an opportunity to observe certain histochemical reactions in the endometrium of a variety of pregnant animals.

We have investigated the endometrium for the presence of iron, calcium, glycogen, lipoids, acid and basic phosphatases, and cytoplasmic basophilia. Further we have explored the occurrence in the endometrium and significance there of the Schiff’s “plasmal” reaction as Well as the Bodian reaction. These investigations have been carried out on a limited number of uteri of pregnant rats, guinea pigs, cats, sows and human beings. With the exception of lipoids, as demonstrable by osmic-acid or Sudan III, and glycogen, none of the substances listed above has been investigated heretofore in the endometrium. As our study of the pregnant uterus progressed, We found it necessary to become acquainted with the antecedent preimplantation endometrium. Hence our investigations had to be enlarged in that direction also. Thus, between the number of stages involved, the various species of animals examined, and the variety of histochemical reactions utilized, we were confronted by an increasingly complicated problem, certain aspects of which we have so far worked out no more than superficially. Nevertheless, some of our findings and the considerations attached thereto seem to us to be sufficiently interesting to be reported in the present paper. The need suggests itself of extending some of the observations in future to more closely seriated stages, with greater regard to the number of regions investigated and the number of species employed.

Material and Methods

For general histological orientation, sections prepared by a number of different methods were utilized. The tri-acid Masson and Azan staining procedures were used on sections 5 u in thickness obtained after fixation in Zenker’s, Helly’s or Bouin’s fluid. Occasional sections were also stained with hematoxylin and eosin. For the purpose of studying basophilia of the cytoplasm, eosin—methylene blue or eosin—toluidin blue stains were applied to sections after Zenker, Holly or Dawson fixation.

For determining the presence of iron, representative blocks of tissue were fixed routinely in a mixture of absolute alcohol and formalin, as recommended by Scott ( ’33). The Turnbull blue reagents were applied to sections made after embedding in paraffin. Contiguous sections were microincinerated and observed for the orange or reddish ash which is indicative of the presence of iron oxides.

For demonstrating the presence of glycogen, tissues were fixed in a solution of absolute alcohol, formalin and picric acid. Paraffin sections were prepared and stained by the Bauer-Feulgen method and by the silver method developed by Mitchell and Wislocki (’44). Occasional sections were also stained with Best’s carmine. The most satisfactory results were regularly obtained by the Bauer and silver procedures.

Alkaline phosphatase was demonstrated by a slight modification of Gromori’s (’41 a) method. Tissuesiwere fixed in either cold acetone or 80% alcohol. The sections were incubated routinely for 3 hours at 37°C. in a solution of sodium glycero—phosphate, the pH of which was adjusted to 9.4 with the aid of a glass electrode. For acid phosphatase, Gomori’s (’41 b) method, as modified by Wolf, Kabat and Newman (’4-3) was employed. Sections of tissues which had been fixed in cold absolute acetone were incubated for 24 hours at 37°C. in a phosphate solution at_ pH 4.7.

For the study of lipoids, unstained preparations fixed in Champy’s fluid and frozen sections stained with Sudan III were occasionally used. Plasmal, the lipoidal material which recolorizes the Schiff reagent (Lison, ’36), was investigated in frozen sections of fresh tissues, or in tissues which had been fixed in 10% formalin or a mixture of formalin and mercuric chloride. In the latter case involving the use of fixed material, the sections were carefully washed to remove any excess of formalin.

The Bodian protargol method, as modified by Dawson and Barnett (’44), was applied to tissues which had been fixed variously in Bodian’s fixative no. 2, Bouin’s fluid or 10% formalin.

The tissues on which this study is based were obtained from 3 rats, 6 guinea pigs, 6 sows, 10 cats, and 19 women. The human material was obtained through the kindness of Drs. Arthur Hertig and John Rock of the Boston Lying—In Hospital and the Free Hospital for Women, Brookline, Mass.


Our results will be presented in descriptive passages relating to each of the histochemical methods. The sequence of topics is purely random. Such correlations of the data as seem important to us will be presented in a discussion at the end of the paper.

Basophilia. The affinity of cell nuclei for basic dyes has long been attributed‘ to the presence of nucleoprotein. More recent investigations have shown that the nuclear nucleoprotein contains a sugar of the desoxyribose type, whereas only ribonucleoprotein has been demonstrated in the cytoplasm. The cytoplasmic ribonucleoprotein, like the nuclear desoxyribonucleoprotein, has a strong affinity for basic dyes. The characteristic cytoplasmic basophilia of several different varieties of cells has now been accounted for by their content of ribonucleoprotein (Brachet, ’4c0; Caspersson, ’40; Gersh and Bodian, ’44; Dempsey and Wislocki, ’45). Consequently, it appears that observations of basophilia are now of considerable significance in the analysis of the chemical content of cells.

The epithelial tissues of the uterus exhibited basophilic staining of the cytoplasm in all of the species of animals examined. Nevertheless, the intensity of staining varied greatly in the different species. In general, the heaviest cytoplasmic basophilia was observed in the epithelium of the uterine glands of the cat and sow. Variations in the degree of basophilia were also observed in different parts of the epithelium of the reproductive tract. For example, in the sow the uterine glands were more heavily stained than was the surface epithelium, whereas in the guinea pig and cat, the surface epithelium was the more strongly basophilic. Lastly, the staining reactions of the uterine epithelium showed differences at different stages in the reproductive cycle. For example, only moderate basophilia characterized the uterine epithelium of the estrous cat, while during pregnancy thesame cells stained much more deeply. We present two illustrations of cytoplasmic basophilia in figures 3 and 4. figure 3 illustrates a typical gland from the endometrium of a sow in the middle of pregnancy, whereas figure 4 shows the cells of a paraplacental gland from the uterus of a guinea pig in mid-gestation.

The connective tissue of the uterus is normally almost totally devoid of basophilia material in all of the species examined. However, just prior to implantation, or in regions remote from the implantation site, the superficial stromal cells in the cat’s uterus round up, hypertrophy, and their cytoplasm takes on a definitely bluish color. A similar intensification of the cytoplasmic staining occurs before implantation in the rat and guinea pig. The subsequent transformation of the stroma into decidual tissue in the several species examined is accompanied by a reduction in basophilia. Consequently the cytoplasm of the decidual cells of the human being, cat, rat, and guinea pig exhibits negligible staining by basic dyes.

Iron reactions. In the uterine glands of women, sows, rats, and guinea pigs, iron—containing substances are revealed during pregnancy by application of the Turnbull blue reagents. It would seem that the presence of iron in the uterine glands of the animals enumerated was a consequence of pregnancy, since in all of these species We have failed to obtain positive reactions in the endometrium of non-pregnant individuals. In the endometrial glands of the cat, in contrast to the preceding animals, we have been unable to demonstrate any iron whatsoever.

In the species in which positive reactions are obtainable, the cytological localization of iron is essentially similar. The iron is contained, for the most part, in the glandular epithelium, whereas the surface epithelium contains little or none at all. As illustrated in a uterine. gland of the sow in figure 1, the apical portions of the glandular cells exhibit the densest concentration of the iron compounds. The appearance of these cells in the sow’s uterine glands is such as to suggest secretion, because there is a progressive increase in amount of iron as the apical surface of the cell is approached, and because iron is demonstrable in the lumens of the glands as well as in the uterine milk located in the lumen of the uterine cavity. The size of the particles which exhibit the iron reaction varies considerably in different animals. figure 1 illustrates also the delicate, dust-like character of the Turnbull blue reaction in a. typical uterine gland of a sow during early pregnancy, whereas figures 2 and 5 respectively depict the coarser character of the iron reaction in a uterine gland of a rat near term and in a gland from a human pregnancy of 16 weeks.

Iron, alone of all the minerals present in tissues, may be detected by the microscopical examination of incinerated sections because of the reddish or orange color of the iron oxides. This fact permits an alternative method for studying the distribution of iron in cells. The localization of iron, as determined by examination of microincinerated sections, is in complete agreement with the above observations based upon sections stained by.the Turnbull blue method. Iron deposits could be detected in the uterine glands from pregnant women, guinea pigs, rats and sows, but not from non-pregnant individuals of the same species. The glandular epithelium from both pregnant and non-pregnant cats’ uteri was consistently negative. figure 20 illustrates a microincinerated section of the endometrium of a pregnant sow photographed with a red filter. In the print the iron in the uterine glands shows up as white.

The Bodian reaction‘. The Bodian protargol method, originally developed to impregnate nerve fibers, is also capable of impregnating other structures. Employing this method, Bennett (’41) found argyrophil granules in the cells of the adrenal medulla, and Dawson and Barnett (’44) observed specific granulations in other locations such as the pituitary gland, the islets of Langerhans and the argentaffine cells of the ga_strointestinal tract.

The Bodian stain, when applied to sections of pregnant uteri from the different species examined in this study, reveals argyrophilic granulations in the epithelial cells in some species but not in others. In the sow, both the lining epithelium and the uterine glands are excessively rich in argyrophilic granules, and figure 12 depicts the appearance of the Bodian-positive inclusions in a uterine gland of this animal, while figure 10 illustrates their occurrence in the surface epithelium. The argyrophilic inclusions are striking, also, in the epithelial cells of the cat’s uterus (figs. 9 and 11), although not so uniformly distributed nor so abundant as in the epithelium of the sow. In the rat’s epithelium they are much less impressive and must be carefully sought for. On the other hand, careful searching for the granules failed to reveal any at all in the uterine epithelial cells of either the guinea pig or human being.

Not only is there a species Variability in regard to the presence or absence of the argyrophilic -elements in the uterine epithelium, but also the localization of the granules within the cells varied somewhat i11 the different species which exhibited positive reactions. In the uterine glands of the sow, as is shown by figure 12, the granules are mostly supranuclear, and a similar distribution is also observed in the uterine glands of the cat (fig. 11). In the surface epithelium of the cat, on the contrary, the granules are concentrated near the apical surface (fig. 9), whereas in the pig the granules are widely scattered throughout the cells (fig. 10). The uterine stroma exhibited no granular argyrophilic materialin any of the species examined with the notable exception of the junctional zone of the fetal and maternal tissues in the pregnant uteri of guinea pigs, rats and human beings. In the junctional zone of these animals definite accumulations of argyrophil precipitate and granular detritus could be observed in the interstices between the degenerating decidual cells.

The presence of argyrophilic substances in the junctional zones is important in that it provided a clue which led to an interpretation of the significance of the Bodian reaction in the particular cases described in this paper. The junctional zone of the human placenta is known frequently to be a site of calcification. This thought led to the further reflection that calcium deposits in general, as exemplified in bones and teeth, show a strong argyrophilia. -This led in turn to the hypothesis that- the Bodian reaction might be indicative of calcium deposits in the localities in which the reaction occurs in the uterus. Fortunately, an experimental proof of this hypothesis suggested itself, since the well-known acid-solubility of inorganic calcium salts should provide a method of abolishing the argyrophilia of the tissues.

In view of the above considerations, the Bodian stain was applied to sections which had previously been fixed in aqueous acid fixatives such as Bouin’s fluid. Following fixation in acid media the argyrophilic granular material could not be detected in the uterine epithelial cells or in the junctional zone. As a variant of the above procedure, sections fixed in the Bodian fixative‘ were subsequently immersed overnight in a 5% aqueous solution of acetic acid. This treatment also abolished the granular argyrophilia in the epithelium as well as in the junctional zone. Lastly, microincinerated sections of some of the uteri were examined. The uterine glands and surface epithelium of the pig’s and cat ’s uterus proved to contain deposits of whitish, acid—soluble ash, the distribution of which coincided with the distribution of the argyyrophilic granules. From these experiments we concluded that the granulations of the endometrium—b-oth intracellular and intersti.tial——which are revealed by the Bodian method represent, in all probability, formed deposits of calcium salts.

Alkaline phosphatase. The localization of alkaline phosphatase in the endometrium varies considerably among the various species. At one stage or another, phosphatase is present in the endometria of all of the species examined. However, the amount and localization of the enzyme vary between wide limits.

During early human pregnancies, the endometrium exhibits phosphatase reactions only in the capillary endothelium. The decidua, stroma and epithelium are entirely devoid of the enzyme. In later stages, beginning at about the fourth month of pregnancy, the endothelium also becomes negative so that no demonstrable phosphatase reaction is obtained.

In the pregnant sow, the glandular and surface epithelia are negative for alkaline phosphatase, but the vascular endothelium exhibits a pronounced reaction. On the other hand, the uterine stroma contains considerable ‘amounts of the enzyme, localized interstitially for the most part, but occasionally demonstrable within the stromal-cells.

In pregnant rats and guinea pigs both the glandular and surface epithelia exhibit moderate amounts of phosphatase. The reaction is most intense in the apical margins of the cells, and this coupled with the presence of phosphatase free in the glandular lumens as well as in the cavity of the uterus suggests that phosphatase is being actively secreted by the uterine epithelium. In addition to the epithelial reaction for phosphatase, intense reactions are demonstrable in the connective tissue of the tunica propria. ‘figure 18 illustrates the marginal phosphatase reaction in the surface epithelium of the guinea pig’s endometrium, besides demonstrating the dense, interstitial deposits located in the tunica propria. Upon occassion, the tissues of several pregnant mice were examined and figure 17 illustrates the distribution of the alkaline phosphatase in the uterine mucosa of one of these animals at term. It is similar in every respect to the rat and guinea pig.

The endometrium of the cat, both before and immediately after implantation, contains alkaline phosphatase which is limited to the outer zone of the mucous membrane. It is abundantly present in the surface epithelium and the outer segments of the uterine glands. It is also present in the stroma beneath the surface epithelium. Unlike the rat and guinea pig, the reaction in the cat is confined to the outer portion of the mucosa, the inner two-thirds of the endometrium being free of phosphatase.

In the course of implantation in the cat the trophoblast becomes adherent to the surface of the uterus, thus sealing off the orifices of the glands. Later the surface epithelium at the implantation site and the underlying glands become gradually eroded. In spite of this progressive destruction of the endometrium, the glandular layer is never fully destroyed even at term. Pari passu with the recession of the -uterine glands, the alkaline phosphatase reaction extends deeper and deeper into the previously negative segments of the uterine glands, until at term the few basal remnants of the formerly extensive glandular. (system exhibit a positive reaction for alkaline phosphatase. Thus, as pregnancy advances, the phosphatase reaction overtakes progressively the more deeply situated glandular cells. figures 13 and 14.il1lustrate the distribution of alkaline phosphatase in the paraplacental endometrium of two pregnant cats; the first specimen from early pregnancy and the second from the latter half of gestation.

Acid phosphatase. ()f the several species examined in the present investigation, positive reactions for acid phosphatase have been obtained only in the endometrium of the sow and the cat. In the sow, the non—pregnant endometrium is devoid of the enzyme, whereas during pregnancy excessively dense reactions occur in the glandular epithelium and much fainter reactions characterize the cells of the surface epithelium. figure 19 illustrates the distribution of acid phosphatase in the sow ’s endometrium. The glandular phosphatase is especially concentrated in the apical regions of the cells. Phosphatase is frequently encountered free in the glandular lumens, and in some instances, acid phosphatase has been identified in the secretion in the uterine lumen,

In the eat, there are similar but more complicated changes associated with acid phosphatase than is the case in the sow. The uterus of the cat in estrus previous to implantation is devoid of acid phosphatase. On the other hand, by the time the blastocyst has reached a diameter of 1 cm. the deeper portions of the glands have assumed an intense reaction as illustrated in figure 15. At later stages of pregnancy, as the glands become eroded from the surface downward, the regions which display the acid phosphatase reaction become more and more restricted. By mid—pregnancy, only occasional ones of the most deeply situated glands contain any of the enzyme, traces of which persist until term.

Lipoids and the Schtfl or plasmal reaction. Fatty substances stainable by osmic acid or Sudan III have been demonstrated iii the endometrium of a Variety of pregnant animals (ungu— lates, Bonnet, 1882; carnivores, Melissenos, ’O6; Heinricius, ’14; rabbit, Chipman, ’02; man, Froboese, ’24, Aykroyd and Gatenby, ’41; Gillman, ’41). Fat varies considerably in manner of distribution and amount in various regions of the endometrium and at different stages of pregnancy in sundry species of animals, but no one so far has attempted to make any detailed or systematic survey of it. Lipoids arise after ovulation but before implantation, as witnessed by a number of detailed studies of the preimplantation endometrium, or its equivalent in experimentally induced deciduomas (rat, Krehbiel, ’37; macaque, Rossman, ’40; cat, Dawson a11d Kosters, ’44; human, Froboese, ’24; Aykroyd and Gatenby, ’41, Gillman, ’41).

Lipoids occur according to our own observation in the endometrium of the pregnant sow. Fat is. present- in the glandular and surface epithelium in the form of fine droplets, visible after Sudan III as an orange wash and after Sudan black as definite dots. In the pregnant‘ uterus of the cat, some fat is demonstrable in the necks of the uterine glands and in abundance in the surface epithelium of the paraplacental endometrium. In the necks of the uterine glands the lipoidal material is situated both above and below the nucleus Whereas in the surface epithelium it is located basally for the most part. In the human uterus during the first months of pregnancy it is abundantly present in the glandular and surface epithelium, and to a lesser and variable extent in some of the decidual cells. figure 6 illustrates the lipoidal material, visible after Champy fixation, in a human uterine gland from an endometrium of 30 days’ gestation age. In rodents fat is present in various parts of the endometrium surrounding the implantation site, confined mainly to the stroma (Chipman, ’02, rabbit), but present also to some degree in the surface epithelium and the uterine glands (Krehbiel, ’37, rat).

We have examined the endometrium of pregnant sows, cats rats, and human specimens under the polarizing microscope. Frozen, unstained sections viewed with the polarizing microscope have failed in our hands to reveal any birefringence attributable to lipoidal inclusions. All we have found is a variably faint double refraction associated with the cell mem branes of the epithelial cells and the fibrous elements of the stroma. We are not familiar with observations by previous investigators concerning this topic.

In reference to the lipoidal constituents of the endometrium our interest has attached mainly to the Schiff or “plasma1” reaction. The lipoids which characterize the endometrial epithelium give a strongly positive Schiff reaction. This reaction, which consists of the recolorization, is, acco_rding to Lison (’36), demonstrative of a special class of fat soluble substances. Opinions vary as to the significance of the reaction. Feulgen and Voit (’24), Verne (’29), and Gomori (’42) have suggested that aldehydes of fatty acids are the reacting substances, whereas Lison (’32) demonstrated that a number of unsaturated compounds other than aldehydes gave positive responses, and ‘Dempsey and Wislocki (’44) suggested the possibility that in some particular instances steroids might be responsible for the plasmal reaction. At any rate, whatever its chemical significance may be, the Schifl’ reaction provides a means for the characterization of lipoids, and the lipoids of the uterine epithelium are strongly positive. figure 7 illustrates the reaction in a human uterine gland from a pregnancy of 16 weeks. Similar reactions have been observed in the uterine epithelial tissues from sows, rats, guinea pigs and cats. Moreover, this reaction invariably gives positive results in specimens from nonpregnant individuals in the follicular and luteal phases of the cycle, as well as from material obtained during pregnancy.

Glycogen. The distribution of glycogen in the endometrium as observed in the course of the present study is with minor exceptions confirmatory of findings of earlier investigators. For the endometrium of pregnancy" variable complete observations exist for rodents (Chipman, ’02; Driessen, ’08; Loveland, Maurer and Snyder, ’31, and others), cat (Dawson and Kosters, ’44) and human (Driessen, ’07 ; Froboese, ’24, and others). For the preimplantation endometrium or for induced deéiduomas detailed observations exist for the rat (Krehbiel, ’37); cat (Dawson and Kosters, ’44); macaque (Rossmann, ’40) and man (Bartelmez and Bensley, ’32; Gillman, ’41). '

Glycogen appears to be absent in the endometrium of the sow both before and after implantation. Its absence in the cycle is mentioned by Corner (’21), but its total lack in the mucosa of the gravid uterus has not to our knowledge been observed heretofore. Although totally absent from the endomctrium of the pregnant uterus, it is present in certain parts of the adjacent chorion, an observation that will be presented in full in a future report.

In the cat, in the youngest gravid uterus which we have examined, there is glycogen in the surface epithelium and the outer parts of the uterine glands throughout the uterus. In stages older than this We have found no glycogen in the maternal elements of the placental labyrinth, nor in the epithelium of the uterine glands underlying the placental site. In the endometrium between placental sites, on the contrary, the outer parts of the uterine glands and the surface epithelium both contain appreciable amounts of glycogen. figure 16, for example, illustrates glycogen in the uterine epithelium of a cat at mid—gestation.

In rodents glycogen occurs in great quantities during the first half of pregnancy in the decidually transformed tissues at the placental sites. It occurs principally in the stroma but may be identified also in traces, as We have observed, in the surface epithelium and in the small glands of the parapla—cental uterine wall.

In man the decidually transformed uterine Wall at the placentalsite contains large quantities of glycogen during the first half of pregnancy. It is present in abundancein the epithelium of the uterine glands as well as in a certain proportion of the decidual cells. figure 8 illustrates glycogen in a uterine gland of a human endometrium from the eighth Week of pregnancy.


Possible relationship between phosphatase and glycogen in the pregnant uterus. The presence in animal tissues of an enzyme capable of splitting- phosphate ions from organic phosphates was demonstrated by McCol1um and Hart in 1908. Since then, phosphatases have been identified in nearly all animal tissues, and are now known to catalyze a large number of important chemical reactions. Phosphatases are concerned with carbohydrate metabolism, nucleotide metabolism, phospholipid metabolism and calcium deposition (Sumner and Somers ’43).

Since the publication of Gomori’s methods (’41a, ’41 b) for'the histological demonstration of alkaline and acid phosphatases, a number of investigators have surveyed the principal tissues of mammals for the occurrence of the enzymes (Kabat and Furth, ’41; Wolf, Kabat and Newman, ’43; Bourne, ’43). These investigations, although valuable as descriptions of the principal sites in which phosphatase occurs, have not proved particularly helpful in evaluating the biochemical role of the enzyme except in the case of bone formation. In most locations there are marked species variations in the concentration and even in the kind of phosphatase.

As a result of the intensive studies of carbohydrate metabolism which have been carried on during the past decade, the biochemical events leading to the formation of glycogen are now fairly well understood. It is generally agreed that a rphosphorylated intermediary, such as the Cori ester, is a necessary antecedent to glycogenesis. The formation of glycogen then proceeds by dephosphorylation of the hexose phosphate which condenses, in the specific environment of the animal cell, into glycogen. This transformation of hexose phosphate into glycogen requires the presence of a phosphate splitting‘ enzyme, and of the variousienzymes capable of such action, alkaline phosphatase has been most frequently invoked because of its ubiquitous occurrence and because of its demonstrated capacity to utilize the particular substrates which serve as intermediaries in carbohydrate metabolism.

Since the considerations in the preceding paragraph lead to the conclusion that phosphatase activity represents the penultimate stage in glycogenesis, it follows that phosphatase should be demonstrable in a region supervening between the source of sugar and the location in which glycogen deposition is occurring. The hypothesis lends itself to experimental verification by observing whether the presence of phosphatase and the deposition of glycogen are indeed so correlated. Our observations provide evidence of the correctness of the hypothesis in regard to both the fetal and maternal parts of the placenta in a number of species of animals.

The spatial correlation between the distribution of glycogenand the presence of phosphatase, although clearly in evidence in placental structures, exhibits several interesting differences in localization. In the sow, glycogen occurs only in the fetal chorionic cells and is entirely absent in the endometrium.

‘Alkaline phosphatase is confined to the maternal endothelium and to the apical margins of the fetal chorionic cells. Thus, the transmission of sugar from the maternal to the fetal tissues involves its passage across two regions in which phosphatase is demonstrable. Since glycogen deposition occurs immediately after its passage across the second region, the thought lies close at hand that here is the region of final dephosphorylation which precedes the condensation of the glycogen molecule.

In the uterus of ca.ts, during progestation and early pregnancy, glycogen is deposited in the glandular and surface epithelium. In rodents glycogen accumulates principally in the stroma but occurs also in small amounts in the uterine epithelium. In the cat phosphatase is exhibited exclusively in the uterine epithelial cells, whereas in rodents varying amounts of phosphatase may be observed in the maternal endothelium, stromal cells and stromal interstitial spaces as well as in the uterine epithelium. Thus, in these two animals as we again trace the migration of a hypothetical hexose molecule from the maternal capillary to its deposition as glycogen in the uterine epithelium, we find that before being converted ‘to glycogen, it is obliged to meet up with a bed of tissue containing phosphatase. In the maximum case, illustrated in the surface epithelium of rodents, the successive layers containing phosphatase are the capillary endothelium, the stromal interstitial spaces, and the uterine epithelium, and it is throughout these layers that glycogen accumulates. In the minimal case, exhibited in the cat’s uterus, the only layer containing phosphatase is the glandular epithelium itself, and it is there that glycogen deposition is evident.

The yolk-sac placenta of the rat and guinea pig exhibits an interesting condition with respect to the localization of alkaline phosphatase and glycogen. "The cells of the inverted yolk-sac placenta of these species are exposed to the contents of the uterine lumen (in -the early period by diffusion through Reichert’s membrane and the thin outer layer of the vitelline sac; later by direct contact). Dense concentrations of glycogen are present in the yolk-sac epithelium, and glycogen is present, also, in the lumen of the uterus, as well as in the uterine epithelium and in the stroma beneath. Phosphatase, on the contrary, is not present in the yolk-sac and is demonstrable only in the uterine tissues and in the secretion within the uterine cavity. Consequently, if the glycogen in the yolk-sac is transmitted to it through the uterine secretions, glycogenesis must have occurred in the uterine epithelium or within the fluids contained in the uterine cavity. The" stores of glycogen in the yolk-sac cells could result, therefore, only from the ingestion by the cells of the glycogen preformed under the influence of the phosphatase in the uterine tissues.

Another variant of the anatomical distribution of phosphatase and glycogen is demonstrable in the human uterus during early pregnancy. At this time, dense accumulations of glycogen are found in the glandular epithelium and in the decidual cells at the placental site. .Phosphatase, however, is present only in the endothelium lining the uterine capillaries. Consequently, the. dephosphorylation of hexose phosphate antecedent to glycogenesis apparently occurs in this instance in the capillary endothelium.

The observations described in the preceding paragraphs lead to the conclusion that in the uterus and placenta a region containing phosphatase is invariably interposed between the blood stream and the locality in which glycogen is deposited. Moreover, the anatomical relations which exist between phosphatase and glycogen are in satisfactory accord with the hypothesis that dephosphorylation of hexosephosphate by alkaline phosphatase is a necessary antecedent to glycogen formation. However, although phosphatase is demonstrable somewhere between the maternal blood stream and the site at which glycogen is stored, the anatomical region which contains the phosphatase is by no means a constant one. For example, both phosphatase and glycogen are present in the epithelial cells of the uterine glands of the cat, and in this instance, dephosphorylation by the action of phosphatase apparently occurs in the same cells which also store glycogen. Similarly in rodents the epithelial and stromal cells both contain phosphatase and stored glycogen. In the yolk sac of rodents, on the contrary, glycogen occurs without phosphatase so that glycogenesis has become dissociated from glycogen storage. A similar dissociation is exhibited in the human uterus where glycogenesis, as judged by the presence of phosphatase, takes place in the Vascular endothelium, whereas storage of glycogen occurs in the glandular and decidual cells of the uterus, neither of which contains phosphatase.

The foregoing considerations indicate that alkaline phosphatase occurs in placental structures at sites predictable according to the theory that dephosphorylation of ,heXosephosphate is a necessary antecedent to glycogenesis. There remains for consideration, however, the question of whether or not there is a correspondence between the amounts of alkaline phosphatase and deposited glycogen. In other words, the source of sugar being equal, does the quantity of phosphatase determine the amount of glycogen deposited‘? Our observations show that there is little relation between the relative amounts of these two substances. For example,'in the human uterus large amounts of glycogen are sequestered in the uterine glands and decidual cells, yet only small amounts of phosphatase occur, localized in the capillary endothelium.

Conversely, in the cat the two substances are present abundantly in the uterine glands prior to implantation, but subsequently glycogen disappears from the glands beneath the placental sites, although alkaline phosphatase is amply demonstrable in these localities at all times. These considerations suggest that the concentration of alkaline phosphatase may not be the sole determining factor which regulates the quantity of deposited glycogen. Additional factors, limiting and regulating the amount of glycogen deposited, may pertain to the amount of available sugar delivered by the blood stream and to the rate of utilization of glycogen by the cells.

Significance of glycogen in the endometrium. In discussing the significance of glycogen in the human placenta, Dempsey a.nd Wislocki (’44) pointed. out that it is deposited in regions which are poorly vascularized. From this fact and other considerations they concluded that glycogen deposition often occurs in tissues which are characterized by a low respiratory metabolism and they suggested that anerobic glycolysis might provide a source of energy for oxidation in tissues having limited mechanisms for aerobic respiration.

In the light of that hypothesis it is of interest to consider the occurrence of glycogen in the placentas of the present group of animals. It will be recalled that in the sow no glycogen has been discovered in the endometrium at any time in the reproductive cycle or during gestation. In the eat it is present in moderate amounts before implantation and again shortly after implantation (as outlined. in detail by Dawson and Kosters, ’44) but is negligible in the placental region throughout the bulk of the gestation period. In rodents, man and macaque glycogen makes its appearance during the secretory phase of the cycle and accumulates in great quantity in the decidually transformed endometrium of the first half of gestation.

From these observations and on the basis of our hypothesis, one should be obliged to assume that the sow’s uterus possessed an adequate circulation at all times, that the endometrium in rodents, monkeys and man possessed a poor circulation, and that the cat in this respect occupied an intermediate position. Although no adequate account exists of the changes which the blood vessels in the sow’s uterus undergo, it is app_arent that during the endometrial cycle of this animal relatively little change in the mucosa takes place either in regard to thickness or in the direction of decidual transformation (Corner, ’21). Hence one would anticipate that the alterations in the blood supply would be relatively insignificant. The macaque, man and rodents, on the contrary, are known to have endometria which undergo marked proliferation and thickening, accompanied by manifestations of avascularity of both anatomical and physiological nature (Daron, ’36, Markee, ’40, Rossman, ’40). We would judge from our knowledge of the cat’s progestational endometrium that it is intermediate between the sow and the latter group in reference to the amount of proliferation and thickening which the mucosa undergoes. Certainly as regards decidual transformation the cellular changes in the cat are moderate as compared with rodents. Hence in the cat the blood Vessels should not be obliged to undergo any Very radical changes as to length or distribution; nor are they tapped and destroyed to any great extent during implantation or afterward. Thus these comparative data, though in every respect incomplete, do tend to bear out the postulated relationship between the vascular supply, anerobiosis, and the accumulation of glycogen.

Lipoids in the uterine epithelium and the “plasmal” reaction. Our results confirm the findings of previous investigators on the presence of lipoidal inclusions, stainable by Sudan III or osmic acid, in the endometrium of pregnant sows, cats, rats, guinea pigs and humans. Lipoidal material is demonstrable in the glandular and surface epithelium of all of these forms during all, or some fraction, of the gestation period. In rodents, rhesus monkey and man, lipoids occur, also to some extent in the decidually transformed stromal cells. In the non-pregnant endcmetrium, lipoids are generally believed to be absent during the follicular phase of the cycle but are known to make their appearance during the luteal phase (rat, Krehbiel, ’37; cat, Dawson and Kosters, ’44; human, Gillman, ’41). On the other hand, we have obtained positive “plasmal” reactions from the beginning of the follicular phase on, through the luteal phase and gestation, although sudanophil lipoid does not make its appearance before the luteal phase. Moreover, since the plasmal reaction is abolished by extraction of the section with alcohol or acetone, it must be assumed that it demonstrates the presence of some kind of lipoidal substance.

The chemical significance of the plasmal reaction is, unfortunately, still obscure. As regards its presence in the endometrium, two general possibilities seem evident. The first is that the epithelial cells carry on a form of fat metabolism in which fatty aldehydes or acetals are produced. On the other hand, an alternative is that since tissues which secrete steroids uniformly exhibit positive plasmal reactions, the Schiff reaction of the endometrium represents merely a fixation of the active material which was secreted by the endocrine cells of the ovaries. In other words, the endometrium may be regarded as a “target organ” in which reactions identical with those of the active hormone may be produced. However, too little is as yet known about the specificity of the plasmal reaction to choose between these alternatives at the present time. The occurrence of lipoid in the epithelium and stroma of the endometrium of rodents, rhesus monkey and man in progestation and pregnancy has been very generally regarded as a manifestation of degeneration (Froboese ’24, Bartelmez and Bensley ’32, Rossman ’41). More recently Gillman (’41) and Gilbert (’42) have questioned this assumption, in man and rabbit respectively, pointing out that much of the fat is situated basally in the cells, and that its presence there is regulated by the ovarian hormones. Our own observations add little to this topic beyond the thought that the lipoidal material indicated by the plasmal reaction is so general in occurrence within the cycle from estrus on, and occurs in so many difierent mammals, that it is unlikely that its presence marks degenerative changes. Also, it may be pointed out from the comparative angle that the uterine epithelium of the sow, unlike that of deciduate mammals, does not undergo desquamation or necrosis during pregnancy and yet its uterine glands are heavily charged with lipoidal material stainable by the several methods for fat.

Secretion by the uterine glands during pregnancy. The present observations concern themselves in large measure with sundry histochemical reactions of the glandular and surface epithelium of the endometrium in pregnancy. Glycogen, lipoids, phosphatases, iron, calcium and cytoplasmic basophilia characterize the uterine glands and surface epithelium of alvariety of animals which we have examined. The widespread occurrence of lipoids, glycogen, iron and phosphatases in the uterine glands, as well as in their lumina, suggests that these substances may be secreted by the uterine glands for the purpose of nourishing the growing blastocyst during all or some part of the period of gestation. In ungulates, the “uterine milk” in the lumen of the uterus has long been regarded as a nutritive fluid known to contain protein substances, lipoid droplets, leucocytes, iron compounds, pigments, red blood cells to some extent, and occasional crystals (Bonnet, 1882, Kolster, ’06). Of these bodies and substances present in the uterine milk of ungulates, the protein component has been considered as being secreted by the uterine glands, whereas the lipoid droplets and the iron compounds contained therein have been regarded very generally ‘as being liberated respectively from degenerating leucocytes and erythrocytes which had previously migrated, or somehow escaped, into the uterine lumen (Bonnet, Kolster). Yet, our own experience with the histochemical appearances of the s0w’s endometrium leads us to believe that the iron and lipoid present in the uterine milk in this animal are secreted by the uterine glands and surface epithelium, instead of being derived from leucocytes and erythrocytes which have undergone dissolution. Our observations on the endometrium of pregnant cats and rodents, as well as of the human, suggest that secretion by the uterine glands and surface epithelium occurs quite generally in pregnancy and that it contributes at various times and in various ways during gestation towards nourishing the blastocyst. Secretion of the glands probably contributes in all mammals to the nourishment of the free blastocyst and in Variable degrees to the support of the implanting blastocyst. Other means of obtaining nourishment besides glandular secretion are available to the growing blastocyst and it usually utilizes all such opportunities in varying degrees and in varying amounts at different periods of gestation. But, in so far as a patent uterine lumen, or some recess from it, has glands opening into it or uterine epithelium lining it during pregnancy, this space, or these spaces, will contain secretion derived from the endometrium and this secretion will‘ be a source of nourishment for the embryo in so far as it bathes viable, functional parts of the fetal membranes (e.g., “green border-” in carnivores, vitelline placenta in rodents). In the sow, it seems likely that secretion provides a principal source of nourishment for the embryo from the beginning to the end of pregnancy. In the other mammals here considered, secretion plays a more subsidiary and variable role, the details of which need not be discussed at this time.

The histochemical reactions of the uterine glands under consideration here offer a few additional points of interest which merit brief discussion. These will be presented under separate headings.

Significance of basophilia of the cytoplasm in the uterine glands. It will be recalled from a previous section of this paper that the epithelial tissues of the uterus exhibited basophilic staining of the cytoplasm in all of the species of animals examined. Little or no basophilia occurred in the decidually transformed elements of the stroma. The intensity of the epithelial basophilia varied considerably in the different species and in different regions of the uterus. In general the heaviest staining was observed in the epithelium of the uterine glands of the cat and sow. Proof was submitted that the basophil material under consideration consisted of ribonucleoprotein.

In a study of human uterine gland cells Bartelmez and Bensley (’32), remarked upon “basophilic” substance. They found material of unknown nature within the epithelial cells which lay close to the nucleus and stained with basic dyes after a great Variety of fixatives. It differed from Nissl substance in its lack of solubility in 1% KOH. It was seen, according to them, in moderate amount in the proliferative phases of the cycle and became particularly abundant during the progravid stage. It appeared to vary in quantity inversely as the intracellular glycogen, and the two, they believed, might be related.

For the demonstration of basophilic substance, and comparison of the amounts of it in various tissues, we have found, after trying various fixatives, that Zenker-acetic followed by eosin—methylene blue stain gives the maximal preservation and staining. Hence, we have utilized this method as a standard for comparing various tissues.

We have observed also that the basophilic substance and the storage of glycogen are inversely related. Basophilia is extremely pronounced in the uterine epithelium of the pregnant sow and cat, but the epithelium of the former contains no glycogen whatsoever, and that of the cat relatively little. Nevertheless, we regard this inverse relationship of basophilia and glycogen storage-as one that is incidental and does not bespeak any chemical relationship between the two substances, for the sugar molecule contained Within the ribonucleoprotein complex is a pentose sugar not readily convertible into glycogen.

Cytoplasmic basophilia when subject to extinction by ribonuclease has been correlated with the presence of nucleoproteins containing ribose sugar (Brachet, ’40*) and such proteins have been implicated in essential protein synthesis by cells (Mirsky, ’43). Because we have found the basophilia in the uterine epithelium to respond to ribonuclease, we are disposed to believe that its presence there bespeaks thepproduction of proteins which are secreted into the uterine cavity where they are available for the nourishment and needs of the blastocyst. Apparently when the cells are actively engaged i11 protein synthesis they do not synthesize or store much glycogen, and vice versa. finally, one should recall that Mirsky (’43) has correlated basophilia with secretory activity in several cellular locations. The fact that the amount of cytoplasmic basophilia changes during the reproductive cycle and in the course of pregnancy suggests that in the uterine glands, also, it reflects changing synthesis and secretion of proteins.

Significance of iron in the uterine glands. We have found iron specifically localized in the uterine glands of all of the animals examined with the exception of the cat. We regard the presence of iron in the glandular epithelium, as Well as in the lumen of the glands, as evidence of its being concentrated and secreted by the uterine glands. Unlike glycogen and lipoids which appear in the uterine glands during the luteal phase of the ordinary cycles, We have observed iron in these glands only during pregnancy. For example, we saw no iron in the normal secretory human endometrium of a specimen of the twenty-fourth day of the cycle, and we have seen it repeatedly in the glands of human endometria obtained from pregnancies of a month or longer. Yet, we have not secured sufficient human specimens, nor have we prepared closely enough seriated material in animals, to say how soon after implantation iron first makes its appearance in the uterine epithelium. The concentration of iron and its secretion by glands are not unique, because secretion of iron has been shown to be a function of the sweat glands of the axilla (Homma, ’26).

As. reported above, we have not encountered any iron at all in ‘the uterine epithelium of cats. This exceptional behavior of the cat’s endometrial glands we are inclined to attribute to the fact that in carnivores, more than in any other group of mammals, iron is made available for absorption by the fetal membranes by extensive and repeated extravasation of maternal blood into the uterine cavity. These characteristic extravasates of blood in carnivores are called “border” or “central” hematomas, according to their position in the placenta (“ green” and “brown borders” respectively of dog and cat). Thus in the cat and other carnivores the fetus apparently receives its entire quota of iron by means other than the uterine glands.

The Bodian protargol reaction in the uterine glands. In a previous section of this paper We have described the occurrence of the Bodian reaction in the glands and surface epithelium of the uteri of sows, cats and rats and noted its absence in the uterine epithelium of guinea pigs and human beings. We presented evidence indicating that the protargol reaction reveals the presence of formed calcium in the localities indicated. The presence of calcium in the uterine glands suggests that it may be secreted by the uterine epithelium. In the uterine glands of the sow the protargol reaction is intense, whereas in the cat’s endometrium it is less strong, in the rat’s uterine glands minimal, and in the glands of the guinea pig and man not present at all. In the soW’s endometrium in Which the chorion and embryo depend largely upon the uterine glands for their nourishment it seems logical to find the reaction intense. In other mammals (ca.ts, rodents, man) in which the reaction is less intense, or absent, in the uterine glands, it seems safe to assume that calcium finds its way to the embryo mainly or entirely by other avenues. Indeed, in the human, calcium has been traced and identified in its passage through the placenta. It is demonstrable in the trophoblast and subjacent stroma of human chorionic villi and it can be safely assumed that it is absorbed directly from the maternal blood which bathes the villi.

Similar to the case of iron, We have found the Bodian protargol reaction only positive in the endometrium during pregnancy, whereas glycogen and lipoids appear in the uterine wall before implantation. These differences seem understandable, glycogen and lipoidal material presumably representing readily available and necessary sources of energy for the developing blastocyst from the very outset, whereas iron and calcium fulfill their roles principally in the relatively late onset of erythropoesis and calcification of bone. Only the most minute amounts of the two minerals would be necessary earlier to supply the needs for bound iron and calcium in respiratory enzyme systems and cell membranes.


The present study concerns the application of a number of histochemical methods to the endometrium of pregnant rats, guinea pigs, sows, cats and human beings. The methods in— clude reactions for iron, glycogen, lipoids and both alkaline and acid phosphatases. In addition, basophilia, which probably indicates the presence of cytoplasmic nucleoproteins, and the Bodian protargol reaction, which under some circumstances appears to be indicative of formed deposits of calcium, were studied. The uteri were obtained from individuals at various known stages of the reproductive cycle and pregnancy.

Iron was demonstrated in the uterine glands during pregnancy in all the species" examined except the cat. In nonpregnant individuals, on the other hand, no iron could be demonstrated in any species. Glycogen was demonstrated in the uterine glands in all of the animals except the sow. Lipoids were demonstrated by means of osmic and Sudan III in the epithelium and stroma of pregnant uteri from several species, and a lipoid exhibiting a positive Schiff reaction was present at all times. Alkaline phosphatase was found somewhere in the endometrium of every species examined. However, its amount and distribution varied widely, being present in only small amounts in uteri from women and sows and heavily concentrated in uteri from cats, guinea pigs and rats. Acid phosphatase occurred in the uterine epithelium and glands in only cats and sows. Cytoplasmic basophilia. of the uterine tissues was confined almost entirely to the glandular and surface epithelium and showed considerable Variation from species to species, at different stages of the cycle and during pregnancy. A positive Bodian protargol reaction, probably here indicating the presence of formed deposits of calcium, characterized the uterine glands and epithelium of the sow, cat and rat in diminishing ‘order, but not of the guinea pig or human being. ‘ I

These findings are discussed in relation to the physiological ‘activity of the endometrial tissues during pregnancy. A possible relationship between alkaline phosphatase and glycogen is suggested.

Literature Cited

AYKROYD, 0. AND J. B. G-ATENBY 1941 Cytology of human uterine glands in gravid and non-gravid phases. Quart. J. Micr. Sc., vol. 82, pp. 541-562.

BARTELMEZ, G. W., AND C.- M. BENSLEY 1932 Human uterine gland cells. Special Cytology, vol. 3, pp. 1523-1564. Paul B. Hoeber, Inc., New York.

BENNETT, H. ‘S. 1941 Cytological manifestations of secretion in the adrenal medulla of the cat. Am. J. Anat., vol. 69, pp. 333-381.

BONNET, R. 1882 Die Uterinmilch und ihre Bedeutung fiir die Frucht. Stuttgart.

BOURNE, G. 1943 The distribution of alkaline phosphatase in various tisuies. Quart. J. Exp. Physiol., vol. 32, pp. 1-20. '

BRACKET, J. 1940 La détection histochemique des acides pentosenucleiques. Compt. rend. soc. de biol., T. 133, p. S8~90.

CASPERSSON, T. 1940 Methods for the determination of the absorption spectra of cell structures. J. Roy. Micr. Soc., vol. 60, pp. 8-25.

CHIPMAN, W. W. 1902 Observations on the placenta of the rabbit, with special reference to the presence of glycogen, fat and iron. Studies from the Royal Victoria Hospital, vol. 1, no. 4, pp.r1—261.

CORNKR, G. W. 1921 Cyclic changes in the ovarie and uterus of the sow and their relation to the mechanism of implantation. Contrib. to Embryol., No. 64, vol. 13, pp. 117-146.

DARON, G. H. 1936 The arterial pattern of the tunica mucosa of the uterus in Macacus rhesus. Am. J. Anat., vol. 58, pp. 349-419.

DAWSON, A. B., AND J. BARNETT 1944 Bodian’s protargol method applied to other than neurological preparations. Stain Technol., vol. 19, pp. 115118.

DAWSON, A. B., AND B. A. Kosrnas 1944 Preimplantation changes in the uterine mucosa of the cat. Am. J. Anat., vol. 75, pp. 1-37.

Dnmrsnr, E. W., AND G. B. W1sLocK1 1944 Observations on some histochemical reactions in the human placenta, with pecial reference to the significance of the lipoids, glycogen and iron. Endocrinology, vol. 35, pp. 409-429.

1945 Histochemical reactions associated with basophilia and acidophilia in the placenta and pituitary gland. Am. J. Anat., vol. 76, pp. 277—3o1.

DR.IEssnN, L. F. 1907‘ Ueber Glykogen in der Placenta. Arch. 1’. Gynaek., Bd. 82, S. 278-301.

1908 Ueber den Glycogengehalt der Placenta bei Kaninchen, Mausen und Menschen (mit mikroscopischen Skioptikon-dcmonstrationen).

Zentralbl. f. Gyniik., 32 Jahrg. pp. 110-111. (abstract)

FEULGEN, R., AND K. VOI'1'_ 1924 fiber einen weitverbreiteten festen Aldyhyd. Seine Entstehung aus einer Vorstufe, sein mikrochemischer und mikroskopischchemiseher Nachwei und _die Wege zu seiner priiparativen Darstellung. Pfliiger’s Arch. f. d. ges. Physio]. Bd. 206, S. 389-410.

Fnoaonsn, O. 1924 Diie Verfettung des Endometriums. Beitrag zur normalen und pathologischen Anatomic der Uterus. Virchow’s Arch. f. path. Anat., Bd. 250, S. 296-342.

GERSH, I., AND D. BODIAN 1943 Histochemical analysis of changes in Rhesus motoneurons after root section. Biol. Symposia, vol. 10, pp. 163-184.

GILBERT, C. 1942 The distribution of fat in the endometrium of the castrate rabbit after treatment with estradiolbenzoate and progesterone. Endocrinology, vol. 30, pp. 773-781.

GILLMAN, J. 1941 The lipines in the human endometrium during the menstrual cycle and pregnancy and their relations to the metabolism of oestrogen and progestogen. South African J. M. Sc., vol. 6, pp. 59-81.

G_‘roMoRI, G. 1941 a The distribution of phosphatase in normal organs and tissues. J. Cell. and Comp. Physiol., vol. 17, pp. 71-83. '

1941 b Distribution of acid phosphatase in the tissues under normal and pathologic conditions. Arch. Path., vol. 32, pp. 189-199.

1942 Histochemical reactions for lipid aldehydes and ketones. Proc. Soc. Exper. Biol. and Med., vol. 51, pp. 133-134.

HEINRICIUS, G. 1914 flber die Embryotrophe der Raubtiere (Hund, Fuchs, Katze) in morphologischer I-Iinsicht. Anat. Hefte, Bd. 50, S. 115-193.

HOMMA, H. 1926 On apocrine sweat glands in white and negro men and women. Bull. Johns Hopkins Hospital, vol. 38, pp. 365-371.

KABAT, E. A., AND J. PERTH 1941 A histochemical study of the distribution of alkaline phosphatase in various normal and neoplastic tissues. Am. J. Path., vol. 17, pp‘. 303-318.

Konsrm, R. 1906 fiber die Zusammensetzung der Embryotrophe der Wirbeltiere. Anat. Heftet, Bd. 16, S. 794-842. ‘

KaEHB1m., R. H.. 1937 Cytological studies of the decidual reaction in the rat during early pregnancy and in the production of deciduomata. Physio]. Zool., vol. 10, pp. 212-233.

LIsoN, L. 1932 Sur la spéciflcité du reactif de Schiff. envers les aldéhydes. Bull. d’histol. appliq. a la physiol., T. 9, pp. 177-195.

1936 Hitochemie animale. Methodes et Problemes. Gauthier—Villars, Paris.

Lovm.AN1>, G., E. E. MAURER AND F. F. SNYDER 1931 The diminution of the glycogen store of the rabbit placenta during the last third of pregnancy. Anat. Rec., vol. 49, pp. 265-274.

MAEKEE, J. 1940 Menstruation in intraocular endometrial transplants in the rhesus monkey. Contrib. to Embryol., no. 177, vol. 28, pp. 219-308.

MCOOLLUM, E. V., AND E. B. HART 1908 On the occurrence of a phytin-splitting enzyme in animal tissues. J. Biol. Chem., vol. 4, pp. 497-500.

MELISSENOS, C. 1906 Ueber die Fettkiirnchen und ihre Bildung in der Placenta bei den Nagern und der Katze. Arch. f. mikr. Anat., Bd. 67, S. 267-290.

Mmsxv, A. E. 1943 Chromosomes and Nucleoproteins. Advances in Enzymology, vol. 3, pp. 1-34. .

MITCHELL, A. J., AND G. B. WIsLooKI 1944 Selective staining of glycogen by ammoniacal silver nitrate. Anat. Rec., vol. 90-, pp. 261-266.

ROSSMAN, I. 1940 The deciduomal reaction in the rhesus monkey (Macaca mulatta). I. The epithelial proliferation. Am. J. Anat., vol. 66, pp. 277-365. .

Scorn‘, G. H. 1933 A critical study and review of the method of microincineration. Protoplasma, vol. 20, pp. 133-151.

SUMNER, J. B., AND G. F. SOMERS 1943 Chemistry and Methods of Enzymes. Academic Press Inc. New York.

VERNE, J. 1929 Etude histochimique des substances aldéhydiques formées au cours du métabolisme des corps gras. Ann. de Physio1., T. 5, pp. 245-267.

WOLF, A., E. A. KABAT AND W. NEWMAN 1943 Histochemieal studies on tissue enzymes. III. A study of the distribution of acid phosphatases with special reference to the nervous system. A111. J. Path., V01. 19, pp. 423439.



1 Localization of iron in the uterine gland of the sow during early pregnancy (35-mm. embryo). The iron reaction is exhibited as a delicate dustlike stippling, localized for the most part in the apical regions of the cells and in the glandular lumen. The tissue was fixed in a mixture of absolute alcohol and formaldehyde and stained by the Turnbull blue reaction. This and the following figures on this plate were drawn with the aid of a 90 X. apochromat objective and a 10 X ocular.

2 Localization of iron in the uterine gland of a rat at a. late stage of pregnancy (44-mm. embryo). In addition to the delicate stippling, coarser aggregates containing iron are also present in the cells and in the uterine lumen. Alcohol-formol fixation, Turnbull blue reaction. .

3 Localization of basophilic substance in the uterine gland of a pregnant sow (145-mm. embryo). Zenker-acetic fixation, Eosin-methylene blue stain.

4 Localization of basophilic substance iii a paraplacental uterine gland from a pregnant guinea pig (75-mm. embryo). Zenker-acetic fixation. Eosimmethylene blue stain.


5 Iron reaction in a. human uterine gland of 16 weeks’ pregnancy. Alcoholformol fixation, Turnbull blue reaction. 90 X objective, 10 X ocular.

6 Lipoidal droplet in a human uterine gland of 30 days’ pregm-lncy. Champy fixation, unstained. 90 X objective, 15 X ocular.

7 Schiff, or plasmal, reaction in a human uterine gland of 16 weeks’ pregnancy. 90 X objective, 10 X ocular.

8 Localization of glycogen in a human uterine gland of 8 weeks’ pregnancy. Alcohol-formol-picric fixation, Bauer stain. 40 X objective, 7 X ocular.


Bodian reactions in the uterine epithelium and glands of the pregnant cat and sow.

9 Uterine epithelium from a cat during the latter part of preg11an(-,y (125-min. embryos). Alcohol-formol-acetic fixation, Bodian stain. 90 X objective, 10 X ocular. 10 Uterine epithelium from a sow during the middle of pregnancy (145-mm. embryos). Alcohol«formol~picrie fixation, Bodian stain. 90 X objective, 7 X ocular. 11 Uterine gland from a. cat during the middle of pregnancy (70-mm. embryos).

Alcohol—form0l—aceti<3 fixation, Bodian stain. 90 X objective, 10 X ocular.

12 Uterine gland from :1 saw during the middle of pvt-g11:n1(-,y (145—mm.

embryos). Alcol10l—formo1-pieric fixation, Bodian stain. 90 X objective, 10 X ocular.


13 Distribution of alkaline phosphatase (pH 9.4) in the endometrium of the cat during early pregnancy, (blastocyst 13- cm. in diameter). The uterine epithelium and the superficial glands and associated stroma show intense reactions, whereas the deeply situated regions are negative. Acetone fixation, alkaline phosphatase reaction. X 170.

14 Distribution of alkaline phosphatase (pH 9.4) in the endometrium of a cat during the middle of pregnancy (70—mm. embryos). The presence of the enzyme in the uterine lumen, the superficial stroma and glands and in the glandular lumens is clearly shown. 80% alcohol fixation, alkaline phosphatase reaction. X 280.

15 Distribution of acid phosphatase (pH 4.7) in the endometrium of a cat during early pregnancy (blastoeyst 1% cm. in diameter). The enzyme is restricted to the deeply situated uterine glands and their lumens. Compare with figure 13. Acetone fixation, acid phosphatase reaction. X 170.

16 Localization of glycogen in the uterine epithelium of a cat during the middle of pregnancy. The infranuclear position of the glycogen is clearly shown. Alcoholformol-picric fixation, Pap’s silver stain. 90 X objective, 10 X ocular.


17 Alkaline phosphatase reaction (pH 9.4) in the endometrium and yolk-sac of a pregnant mouse at term. 80% alcohol fixation, alkaline phosphatase reaction. X 170.

18 Alkaline phosphatase reaction (pH 9.4) in the endometrium of the pregnant guinea pig (75-mm. embryos). X 250.

19 Acid phosphatase reaction in the endometrium of the pregnant sow (125mm. embryos). Compare with figure 15. Acetone fixation, acid phosphatase reaction. X 170.

20 Distribution of reddish ash in the endometrium of the pregnant sow (235mm. embryo). Al(-,oho1—f0rmol fixation, mic1'oincine1'ated. Photographed through a red filter. X 260.

Cite this page: Hill, M.A. (2024, April 23) Embryology Paper - Histochemical reactions of the endometrium in pregnancy. Retrieved from

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