Yolk sac development in lizards (Lacertilia: Scincidae): New perspectives on the egg of amniotes

Embryos of oviparous reptiles develop on the surface of a large mass of yolk, which they metabolize to become relatively large hatchlings. Access to the yolk is provided by tissues growing outward from the embryo to cover the surface of the yolk. A key feature of yolk sac development is a dedicated blood vascular system to communicate with the embryo. The best known model for yolk sac development and function of oviparous amniotes is based on numerous studies of birds, primarily domestic chickens. In this model, the vascular yolk sac forms the perimeter of the large yolk mass and is lined by a specialized epithelium, which takes up, processes and transports yolk nutrients to the yolk sac blood vessels. Studies of lizard yolk sac development, dating to more than 100 years ago, report characteristics inconsistent with this model. We compared development of the yolk sac from oviposition to near hatching in embryonic series of three species of oviparous scincid lizards to consider congruence with the pattern described for birds. Our findings reinforce results of prior studies indicating that squamate reptiles mobilize and metabolize the large yolk reserves in their eggs through a process unknown in other amniotes. Development of the yolk sac of lizards differs from birds in four primary characteristics, migration of mesoderm, proliferation of endoderm, vascular development and cellular diversity within the yolk sac cavity. Notably, all of the yolk is incorporated into cells relatively early in development and endodermal cells within the yolk sac cavity align along blood vessels which course throughout the yolk sac cavity. The pattern of uptake of yolk by endodermal cells indicates that the mechanism of yolk metabolism differs between lizards and birds and that the evolution of a fundamental characteristic of embryonic nutrition diverged in these two lineages. Attributes of the yolk sac of squamates reveal the existence of phylogenetic diversity among amniote lineages and raise new questions concerning the evolution of the amniotic egg. J. Morphol. 278:574–591, 2017. © 2017 Wiley Periodicals, Inc.     

Demonstration of a phagocytic cell system belonging to the hemopoietic lineage and originating from the yolk sac in the early avian embryo.

It is well established that hemopoietic cells arising from the yolk sac invade the avian embryo. To study the fate and role of these cells during the first 2.5-4.5 days of incubation, we constructed yolk sac chimeras (a chick embryo grafted on a quail yolk sac and vice versa) and immunostained them with antibodies specific to cells of quail hemangioblastic lineage (MB1 and QH1). This approach revealed that endothelial cells of the embryonic vessels are of intraembryonic origin. In contrast, numerous hemopoietic cells of yolk sac origin were seen in embryos ranging from 2.5 to 4.5 days of incubation. These cells were already present within the vessels and in the mesenchyme at the earliest developmental stages analyzed. Two hemopoietic cell types of yolk sac origin were distinguishable, undifferentiated cells and macrophage-like cells. The number of the latter cells increased progressively as development proceeded, and they showed marked acid phosphatase activity and phagocytic capacity, as revealed by the presence of numerous phagocytic inclusions in their cytoplasm. The macrophage-like cells were mostly distributed in the mesenchyme and also appeared within some organ primordia such as the neural tube, the liver anlage and the nephric rudiment. Comparison of the results in the two types of chimeras and the findings obtained with acid phosphatase/MB1 double labelling showed that some hemopoietic macrophage-like cells of intraembryonic origin were also present at the stages considered. These results support the existence in the early avian embryo of a phagocytic cell system of blood cell lineage, derived chiefly from the yolk sac. Cells belonging to this system perform phagocytosis in cell death and may also be involved in other morphogenetic processes.     

Defective vascular development in connexin 45-deficient mice

In order to reveal the biological function(s) of the gap-junction protein connexin 45 (Cx45), we generated Cx45-deficient mice with targeted replacement of the Cx45-coding region with the lacZ reporter gene. Heterozygous Cx45(+/)(-) mice showed strong expression of the reporter gene in vascular and visceral smooth muscle cells. Cx45-deficient embryos exhibited striking abnormalities in vascular development and died between embryonic day (E) 9.5 and 10.5. Differentiation and positioning of endothelial cells appeared to be normal, but subsequent development of blood vessels revealed impaired formation of vascular trees in the yolk sac, impaired allantoic mesenchymal ingrowth and capillary formation in the labyrinthine part of the placenta, and arrest of arterial growth, including a failure to develop a smooth muscle layer surrounding the major arteries of the embryo proper. As a consequence, the hearts of most Cx45-deficient embryos were dilated. The abnormal development of the vasculature in the yolk sac of Cx45(-)(/)(-) embryos could be caused by defective TGFbeta signalling, as the amount of TGF beta1 protein in the epithelial layer of the yolk sac was largely decreased in the E9.5 Cx45(-)(/)(-) embryo, compared with the wild-type embryo. The defective vascular development was accompanied by massive apoptosis, which began in some embryos at E8.5 and was abundant in virtually all tissues of the embryos at E9.5. We conclude that in Cx45(-)(/)(-) embryos, vasculogenesis was normal, but subsequent transformation into mature vessels was interrupted. Development of different types of vessels was impaired to a varying extent, which possibly reflects the complementation by other connexin(s).     

Structure of the endodermal epithelium of the chick yolk sac during early stages of development.

The structure of the areas pellucida and vasculosa of the early chick embryo (stages 11-29) was examined by light, transmission and scanning electron microscopy. The most striking feature of the endodermal cells of these areas is the presence of large intracellular yolk drops which are characteristic of the regions in which they are found; lipid-like homogeneous drops in the area pellucida, heterogeneously composed pleomorphic drops in the mid-region of the area vasculosa and granular drops at the periphery of the area vasculosa in the region of the sinus terminalis. On morphological criteria it is postulated that granular drops may arise by direct engulfment of extracellular yolk, but this does not appear to be true for pleomorphic or homogeneous drops. Since the apical junctions between endodermal cells across the yolk sac are tight, they seal off the extraembryonic compartment from the vitelline circulation and presumably prevent intercellular passage of the yolk constituents. Thus the endodermal epithelium must mediate the transport of nutrients from the yolk mass to the developing embryo. Endodermal cells exhibit a variation across the yolk sac in the presence and number of structures associated with uptake of extracellular materials. The mid-region of the area vasculosa appears to be the most endocytotically active region with an abundance of microvilli, bristle-coated pits and vesicles and apical canaliculi and vacuoles. There is a close association between the endoderm and vitelline blood vessels and this association is maintained, as the yolk sac develops, by the formation of small vessels juxtaposed between the vascular surface of the endoderm and the walls of the large vitelline vessels.     

Hedgehog is required for murine yolk sac angiogenesis.

Blood islands, the precursors of yolk sac blood vessels, contain primitive erythrocytes surrounded by a layer of endothelial cells. These structures differentiate from extra-embryonic mesodermal cells that underlie the visceral endoderm. Our previous studies have shown that Indian hedgehog (Ihh) is expressed in the visceral endoderm both in the visceral yolk sac in vivo and in embryonic stem (ES) cell-derived embryoid bodies. Differentiating embryoid bodies form blood islands, providing an in vitro model for studying vasculogenesis and hematopoiesis. A role for Ihh in yolk sac function is suggested by the observation that roughly 50% of Ihh(-/-) mice die at mid-gestation, potentially owing to vascular defects in the yolk sac. To address the nature of the possible vascular defects, we have examined the ability of ES cells deficient for Ihh or smoothened (Smo), which encodes a receptor component essential for all hedgehog signaling, to form blood islands in vitro. Embryoid bodies derived from these cell lines are unable to form blood islands, and express reduced levels of both PECAM1, an endothelial cell marker, and alpha-SMA, a vascular smooth muscle marker. RT-PCR analysis in the Ihh(-/-) lines shows a substantial decrease in the expression of Flk1 and Tal1, markers for the hemangioblast, the precursor of both blood and endothelial cells, as well as Flt1, an angiogenesis marker. To extend these observations, we have examined the phenotypes of embryo yolk sacs deficient for Ihh or SMO: Whereas Ihh(-/-) yolk sacs can form blood vessels, the vessels are fewer in number and smaller, perhaps owing to their inability to undergo vascular remodeling. Smo(-/-) yolk sacs arrest at an earlier stage: the endothelial tubes are packed with hematopoietic cells, and fail to undergo even the limited vascular remodeling observed in the Ihh(-/-) yolk sacs. Our study supports a role for hedgehog signaling in yolk sac angiogenesis.     

Effects of lysosomal proteinase inhibition on the development of the rat embryo in vitro

Altered lysosomal function in the visceral yolk sac can result in abnormal development. As proteolysis is an important function of the rodent visceral yolk sac during early and mid-gestation, we characterized the lysosomal proteolytic enzyme activity of this extraembryonic membrane and determined the effects of inhibitors of protein degradation on embryonic development. Constituent activities of cysteine and aspartic acid proteinases were measured in rat visceral yolk sac on gestation day 12, and the effects of the cysteine proteinase inhibitors leupeptin, E-64 [trans-epoxysuccinyl-l-leucylamido(4-guanido)butane] and N-ethylmaleimide and the aspartic acid proteinase inhibitor pepstatin were determined in Sprague-Dawley rat embryos cultured in vitro from gestation days 10-12. It was determined that only cysteine proteinases, primarily cathepsins B and L, are active in the mid-gestation visceral yolk sac. The cysteine proteinase inhibitors leupeptin and E-64 both produced a concentration-related decrease in embryonic growth, as measured by crown-rump length, somite number, and embryonic protein content, and a concentration-related increase in incidence of abnormalities. A characteristic pattern of abnormalities was produced which involved a decrease in neural tube volume and the formation of a subectodermal blister opposite the point of attachment of the vitelline vessels. At high concentrations, anophthalmia was also observed. The decreased neural tube volume was associated with increased osmolality of the exocoelomic fluid, the major extraembryonic fluid compartment. It is possible that the osmotic change decreased neural tube volume by causing water to move to the compartment with a higher solute concentration, out of the embryo.(ABSTRACT TRUNCATED AT 250 WORDS)     

Ontogeny of vitamin D action on the morphology and calcium transport properties of the chick embryonic yolk sac

The ontogeny of the calcium transport properties and hormonal modulation of the yolk sac membrane in amniote embryos is presently poorly understood. We investigated the role of 1,25-dihydroxyvitamin D3 (1,25-(OH)2D3) on plasma calcium values, yolk sac morphology and the ability of the yolk sac membrane to transport 45Ca from yolk to embryo. 1,25-(OH)2D3 treatment caused significant hypercalcaemia in 9-, 12- and 15-day embryos. Additionally, this hormone caused a hypertrophy of the endodermal cell layer that comprises the bulk of the yolk sac membrane. Both of these effects were the most dramatic in the 15-day embryo, the oldest age tested. 45Ca added to the yolk was transported into the blood rapidly across the yolk sac membrane. 1,25-(OH)2D3 significantly enhanced this transport in all age groups. [14C]Inulin was also taken across the yolk sac membrane, but at a slower rate than 45Ca; this transport was unaffected by 1,25-(OH)2D3. Thus, the yolk sac responds to 1,25-(OH)2D3 treatment both morphologically and functionally. The mechanism for transport appears to be a specific one, rather than a simple enhancement of non-specific endocytosis.     

Gastrulation and angiogenesis,not endothelial specification,is sensitive to partial deficiency in vascular endothelial growth factor-a in mice

Mouse embryogenesis is dose sensitive to vascular endothelial growth factor-A (VEGF-A), and mouse embryos partially deficient in VEGF-A die in utero because of severe vascular defects. In this study, we investigate the possible causes that underlie this phenomenon. Although the development of vascular defects in VEGF-A-deficient embryos seems to suggest that endothelial differentiation depends on the presence of a sufficient level of VEGF-A, we were surprised to find that endothelial differentiation per se is insensitive to a significant loss of VEGF-A activity. Instead, the development of the multipotent mesenchymal cells, from which endothelial progenitors arise in the yolk sac, is most highly dependent on VEGF-A. As a result of VEGF-A deficiency, dramatically fewer multipotent mesenchymal cells are generated in the prospective yolk sac. However, among the small number of mesenchymal cells that do enter the prospective yolk sac, endothelial differentiation occurs at a normal frequency. In the embryo proper, vasculogenesis is initiated actively in spite of a significant VEGF-A deficiency, but the subsequent steps of vascular development are defective. We conclude that a full-level VEGF-A activity is not critical for endothelial specification but is important for two distinct processes before and after endothelial specification: the development of the yolk sac mesenchyme and angiogenic sprouting of blood vessels.     

Morphological specializations of the yolk sac for yolk processing in embryonic corn snakes (Pantherophis guttatus: Colubridae)

Non‐avian reptiles commonly are assumed to be like birds in their overall patterns of development. However, colubrid corn snakes (Pantherophis guttatus ) have mechanisms of yolk cellularization and processing that are entirely different from the avian pattern. In birds, a vascular “yolk sac” surrounds and digests the liquid yolk. In contrast, in corn snakes, the yolk material is converted into vascularized cords of yolk‐filled cells. In this study, we used stereomicroscopy, histology, and scanning electron microscopy to analyze this unusual developmental pattern in corn snakes. Our observations reveal that the yolk sac cavity is invaded by endodermal cells that proliferate, absorb yolk spheres, and form aggregates of interconnected cells within the liquid yolk mass. As development proceeds, small blood vessels arise from the yolk sac omphalopleure, penetrate into the yolk mass, and become tightly encased in the endodermal cells. The entire vitellus ultimately becomes converted into a mass of vascularized, “spaghetti‐like” strands of yolk‐laden cells. The resulting arrangement allows yolk to be digested intracellularly and yolk products to be transported to the developing embryo. Indirect evidence for this pattern in other species raises the possibility that it is ancestral for squamates and quite possibly Reptilia in general.     

半滑舌鳎胚胎发育中后期卵黄囊超微结构观察

在胚胎发育中期,半滑舌鳎胚胎由胚体、卵黄囊和卵周液构成.对半滑舌鳎胚胎发育中后期的卵黄囊进行超微结构观察.结果表明,卵黄囊是由卵黄囊膜和包裹其内的卵黄物质组成.在半滑舌鳎胚胎发育过程中,卵黄囊内的卵黄物质逐渐消耗产生低分子量的卵黄磷蛋白分裂小泡.分裂小泡转移到卵黄囊内部消黄细胞中,在消黄细胞的作用下分裂小泡转化成卵黄颗粒.随后卵黄颗粒在卵黄囊内表面聚集成囊状结构并转移运输到卵黄囊膜内部,最后把卵黄物质从卵黄囊膜转移并释放到卵周液中,为胚胎发育提供营养.     

Yolk spherocrystal: the structure, composition and liquid crystal template

The structure and composition of the yolk spherocrystal, a biomineral developed in the egg yolk sac during the incubation of a chicken embryo, were investigated through various modern analytical methods. Additionally, inside the yolk sac, yolk liquid crystal, a liquid crystalline phase of lipid developed during the incubation of the embryo, was found and investigated. The spherocrystal was found to be a composite composed of calcium carbonate (vaterite and calcite, primarily the former) and the yolk liquid crystal, which is believed to act as an organic template for spherocrystals mineralization, in a concentric multi-layered sphere structure. Moreover, the yolk liquid crystal was found to have a concentric multi-layered spherical structure and a composition consistent with lecithin. We believed that the spherocrystals function as a reservoir for the storage of calcium in the egg yolk sac during the development of the embryo.     

Flow regulates arterial-venous differentiation in the chick embryo yolk sac

Formation of the yolk sac vascular system and its connection to the embryonic circulation is crucial for embryo survival in both mammals and birds. Most mice with mutations in genes involved in vascular development die because of a failure to establish this circulatory loop. Surprisingly, formation of yolk sac arteries and veins has not been well described in the recent literature. Using time-lapse video-microscopy, we have studied arterial-venous differentiation in the yolk sac of chick embryos. Immediately after the onset of perfusion, the yolk sac exhibits a posterior arterial and an anterior venous pole, which are connected to each other by cis-cis endothelial interactions. To form the paired and interlaced arterial-venous pattern characteristic of mature yolk sac vessels, small caliber vessels of the arterial domain are selectively disconnected from the growing arterial tree and subsequently reconnected to the venous system, implying that endothelial plasticity is needed to fashion normal growth of veins. Arterial-venous differentiation and patterning are controlled by hemodynamic forces, as shown by flow manipulation and in situ hybridization with arterial markers ephrinB2 and neuropilin 1, which show that expression of both mRNAs is not genetically determined but plastic and regulated by flow. In vivo application of ephrinB2 or EphB4 in the developing yolk sac failed to produce any morphological effects. By contrast, ephrinB2 and EphB4 application in the allantois of older embryos resulted in the rapid formation of arterial-venous shunts. In conclusion, we show that flow shapes the global patterning of the arterial tree and regulates the activation of the arterial markers ephrinB2 and neuropilin 1.     

Insufficient VEGFA activity in yolk sac endoderm compromises haematopoietic and endothelial differentiation

Vascular endothelial growth factor A (VEGFA) plays a pivotal role in the first steps of endothelial and haematopoietic development in the yolk sac, as well as in the establishment of the cardiovascular system of the embryo. At the onset of gastrulation, VEGFA is primarily expressed in the yolk sac visceral endoderm and in the yolk sac mesothelium. We report the generation and analysis of a Vegf hypomorphic allele, Vegf(lo). Animals heterozygous for the targeted mutation are viable. Homozygous embryos, however, die at 9.0 dpc because of severe abnormalities in the yolk sac vasculature and deficiencies in the development of the dorsal aortae. We find that providing 'Vegf wild-type' visceral endoderm to the hypomorphic embryos restores normal blood and endothelial differentiation in the yolk sac, but does not rescue the phenotype in the embryo proper. In the opposite situation, however, when Vegf hypomorphic visceral endoderm is provided to a wild-type embryo, the 'Vegf wild-type' yolk sac mesoderm is not sufficient to support proper vessel formation and haematopoietic differentiation in this extra-embryonic membrane. These findings demonstrate that VEGFA expression in the visceral endoderm is absolutely required for the normal expansion and organisation of both the endothelial and haematopoietic lineages in the early sites of vessel and blood formation. However, normal VEGFA expression in the yolk sac mesoderm alone is not sufficient for supporting the proper development of the early vascular and haematopoietic system.     

Evidence that suramin and aurothiomalate are teratogenic in rat by disturbing yolk sac-mediated embryonic protein nutrition

Suramin (250 mg/kg) and sodium aurothiomalate (100 mg/kg) both induced congenital malformations in the offspring following treatment of pregnant rats at either 8.5 or 9.5 days of gestation. Conceptuses from 9.5-day pregnant rats were cultured for 48 h in homologous serum to which either suramin or sodium aurothiomalate was added for the final 6 h. The presence of suramin up to 5 mg/ml had no effect on the protein content of yolk sacs at harvesting, but at 10 mg/ml caused a significant decrease. In contrast sodium aurothiomalate increased the protein content of yolk sacs at harvesting, in a concentration-dependent manner up to 100 micrograms/ml. Neither suramin nor sodium aurothiomalate significantly affected embryo protein content. When 125I-labelled polyvinylpyrrolidone was added to the culture serum for the final 6 h of culture, radioactivity was found in the yolk sac at harvesting, but not in the embryo. When suramin (2-10 mg/ml) was present for the final 6 h of culture, the quantity of radioactivity measured in the yolk sac at harvesting was significantly decreased in a concentration-dependent manner. No radioactivity was detected in the embryos. Sodium aurothiomalate had no effect on the uptake of 125I-labelled polyvinylpyrrolidone. When rat serum whose proteins were labelled with [3H]leucine was used as culture medium, radioactivity was found in the conceptus (both yolk sac and embryo) at harvesting. Suramin (5 mg/ml), present for the final or penultimate 6 h, significantly decreased the uptake of radioactivity into conceptuses and caused a significant increase in the proportion of the captured radiolabel that was associated with the yolk sac. Sodium aurothiomalate (25 or 500 micrograms/ml) had no effect on the total uptake of radio-label but caused a significant increase in the proportion of total radioactivity captured that was associated with the yolk sac. These data indicate that suramin, by interfering with both the uptake and intralysosomal digestion of protein, and sodium aurothiomalate, by inhibiting digestion of captured protein, disturb the normal pathway of yolk sac-mediated protein utilization with a consequent diminution of the supply of amino acids to the conceptus. The effects of suramin are seen only at high concentration, those of sodium aurothiomalate at much lower concentrations. It is likely that the two drugs exert their teratogenic action by their effects on the yolk sac nutritional pathway with resultant amino acid deprivation of the conceptus at a critical stage of development.     

Yolk sac cholesteryl ester secretion rates can be manipulated in the Golden Syrian hamster: effect of yolk sac cholesterol concentrations

The yolk sac is one of two extra-embryonic fetal tissues that separates the fetal and maternal circulations. The yolk sac can secrete lipoprotein particles to the vitelline vessels, which supply yolk sac-derived nutrients to the embryo. The amount and composition of lipoproteins secreted from the rat yolk sac can be manipulated by fatty acid content and gestational age. The goals of the current studies were to determine, first, if tissue cholesterol concentration could mediate cholesterol secretion rate from the yolk sac and, second, if some of the secreted cholesterol could be derived from the maternal circulation. Golden Syrian hamsters were fed 2% added cholesterol to increase the yolk sac cholesterol concentration. Yolk sac explants secreted similar amounts of triglyceride and apolipoproteins B and E into the media regardless of yolk sac cholesterol concentration. In contrast, yolk sacs with greater cholesterol concentrations secreted 2.3-fold more cholesterol into the media as compared to control yolk sacs; the increase was found mostly as cholesteryl ester. At least part of the secreted cholesterol was maternally derived. These data demonstrate that yolk sac cholesterol concentration influences cholesterol secretion rates, and that at least some of the cholesterol secreted originates from the maternal circulation.     

Differential permeability of murine visceral yolk sac to thymidine and to hydroxyurea.

Mouse embryos at the 10–12-somite stage of development were excised from the uterus either with or without the encapsulating visceral yolk sac and were incubated in vitro in 3 × 10^?7M thymidine (methyl-T, 5 μCi/ml) for 30 min or in 4 × 10^?3M hydroxyurea for 45 min with [³H]thymidine present for the last 30 min. Radioautograms of nuclei of neural epithelium enabled an estimate of the effectiveness of the barrier imposed by the visceral yolk sac membrane to the passage of thymidine and hydroxyurea.Labeling of nuclei in the neural epithelium showed that the visceral yolk sac caused a 44% decrease in frequency and a 51% decrease in intensity of label. Hydroxyurea inhibited labeling by 15% in frequency and 37% in intensity irrespective of the presence or absence of visceral yolk sac. These results show that hydroxyurea and the presence of visceral yolk sac independently interfered with labeling of the neural epithelium by thymidine and that visceral yolk sac caused a proportionally greater retardation of label than did hydroxyurea.Nuclei of the endodermal epithelium of the intervening yolk sac, following exposure to hydroxyurea, showed a labeling decrease of 44% in frequency and 77% in intensity. The inhibitory effect of hydroxyurea on yolk sac labeling, however, did not alter yolk sac permeability to hydroxyurea. The results indicate that the visceral yolk sac, by offering no detectable barrier to hydroxyurea, permits prompt teratogenic action of hydroxyurea directly upon the embryo and suggest that the visceral yolk sac is a likely candidate to account for reports that the 8.5-day mouse embryo in situ fails to label with radioisotopic thymidine.     

Proteolytic activity in the yolk sac membrane of quail eggs.

Extraembryonal degradation of yolk protein is necessary to provide the avian embryo with required free amino acids during early embryogenesis. Screening of proteolytic activity in different compartments of quail eggs revealed an increasing activity in the yolk sac membrane during the first week of embryogenesis. In this tissue, the occurrence of cathepsin B, a lysosomal cysteine proteinase, and cathepsin D, a lysosomal aspartic proteinase, has been described recently (Gerhartz et al., Comp Biochem Physiol, 118B:159-166, 1997). Determination of cathepsin B-like and cathepsin D-like proteolytic activity in the yolk sac membrane indicated a significant correlation between growth of the yolk sac membrane and proteolytic activity, shown by an almost constant specific activity. Both proteinases could be localized in the endodermal cells, which are in direct contact to the yolk. The concentration of proteinases in the endodermal cells appears to be almost unaltered in the investigated early stage of quail development, whereas the amount of endodermal cells increases rapidly, seen by a complicated folding of the yolk sac membrane. In the same cells quail cystatin, a potent inhibitor of quail cathepsin B (Ki 0.6 nM), has been localized at day 8 of embryonic development. Approximately at this stage of development, the quail embryo stops metabolizing yolk. In conclusion, it is strongly indicated that the amount of available free amino acids, produced by proteolytic degradation and supporting embryonic growth, is regulated by the growth of the yolk sac membrane.     

Ligation of the yolk sac circulation and its effect on the yolk sac ultrastructure and development of the rat fetus]

The effect of an interruption of the yolk sac circulation on the rat visceral yolk sac and the development of the fetoplacental unit was examined in the last third of pregnancy. The yolk sac ischaemia was induced by ligating the blood vessels of the yolk sac stalk which connect the vitelline circulation with that of the fetus. A 3-hour ligature caused an extensive swelling of most cell organelles in the epithelial cells and in the capillary endothelia of the yolk sac. Other structural changes were indicative of a cessation of pinocytosis. A 6-hour ligature resulted in a common increase of cell swelling and in a beginning disintegration of the endothelial cells lining the vitelline capillaries. A 15-hour ligature caused severe ultrastructural cell lesions and macroscopical alterations suggestive of a progressive necrolar finding of a nearly complete loss of the amniotic fluid and the death of the fetus, although the maternal blood flow appeared to be still intact in the placenta.