Search

Human platelets were incubated with gold particles coupled to fibrinogen to label the glycoprotein IIb-IIIa (GPIIb-IIIa) receptor after initial activation of the cells by contact with formvar-coated grid and glass surfaces. Fibrinogen-gold (Fgn-Au) markers were absent on discoid platelets, but diffusely spread over the surface and extended pseudopods of early dendritic cells. Conversion to spread platelets resulted in movement of ligand-receptor complexes away from the cell margin toward cell centres. However, Fgn-Au gold did not concentrate in the central region. Rather, the Fgn-Au, GPIIb-IIIa complexes in the middle of spread platelets appeared to move toward a belt-like, intermediate zone, as did the ligand receptor complexes from the cell margin and pseudopods. The ultimate destination of the mobile receptor-ligand complexes, however, appeared to be channels of the surface-connected open canalicular system (OCS). Fgn-Au was concentrated in OCS channels of most dendritic and a small proportion of spread platelets. The decreased frequency of Fgn-Au filled channels in more transformed platelets may have been due to collapse or evagination of the OCS. Examination of platelets exposed to Fgn-Au after spreading on glass and then prepared for thin sections confirmed that the OCS was the final destination for mobile ligand receptor complexes on surface-activated platelets. Findings of this study are consistent with previous work showing clearance of mobile receptor-ligand complexes to the OCS of platelets activated in suspension.

Adhesion of platelets from 15 patients with von Willebrand's disease was tested in an ex vivo human umbilical vein model. Experiments employed umbilical veins still in their umbilical cords taken from patients undergoing cesarean section and platelets (fetal, adult and vW) either apyrase-washed or used as platelet-rich plasma or whole blood. F VIII R:Ag, F VIII:Rcof, and F VIII:C were measured in initial fresh plasma and in effluents from the umbilical vein segments. F VIII:Rcof increased in most perfusates. Binding of latex-linked specific antihuman F VIII R:Ag demonstrated that F VIII R:Ag existed on subendothelium and on injured endothelial cells. Scanning electron microscopy three-dimensionally displayed vW platelet--vessel-wall interactions. Although vW platelets adhered to injured vein, both qualitative and quantitative differences existed in comparison with adhesion of normal platelets. The differences correlated best with the plasma F VIII:Rcof level. The best adhesion shown by vW platelets was only 51 percent of the adhesion of control fetal or adult platelets. vW platelets had less surface activity, fewer pseudopods, and little ability to spread and pave the exposed subendothelium. Pretreatment of the vein with F VIII R:Ag antibody partly blocked adhesion. Coperfusion of cryoprecipitate with vW platelets improved their adhesivity, state of activation on subendothelium, and ability to form aggregates. ABO differences in blood cell types of fetal material and platelet donors seemed without effect, which further establishes this model's validity for studies of platelet dysfunction and platelet or endothelial reactive agents.

The structural physiology of airways from 80 guinea pigs was examined for changes induced by diesel exhaust (DE) exposure. Acute, subacute and chronic studies contrasted inhalation effects of 250, 750, 1500 and 6000 micrograms DE/m3 with "clean air" breathing of age-matched controls. Nonciliated epithelial (Clara) cells, epithelial type 2 cells and alveolar macrophages were increased in a DE dose dependent fashion. Also, eosinophils, were recruited. Epithelial type 1 cells of the distal airways internalized DEP. The relative dustiness (particulate density) of airways was assessed from coded specimens. Some 86% of DE exposed animals were correctly identified. Scanning Electron Microscopy (SEM) resolved surface located DE particulates (DEP). Single particles, loose clusters, low density agglomerates occurred. While SEM visual clues are insufficient for absolute identification of DE particles, there was supporting evidence from transmission electron microscopy (TEM) and from SEM studies comparing vascular with intratracheally fixed specimens. Presumptive DEP were notable on bifurcation bridges in respiratory bronchioles and alveolar ducts while alveolar outpockets had heavy dust burdens. Clumps of macrophages in such alveoli almost occluded the airspace. We conclude that normal guinea pigs appear to adapt to a chronic DE stress environment. But, the ultrastructural basis (cellular protrusions, DEP agglomerates and secretional debris) exists in peripheral airways for airflow instability and increased airflow resistance.

A nine-year-old black girl with essential thrombocythemia developed hemoptysis. Only two other cases in the English literature have been described. Ultrastructure and functional characteristics of this patient's platelets were studied. Twenty-six percent of the patient's platelets were very large (megathrombocytes). Spontaneous aggregated from the patient's platelets were not compact, and the pseudopods did not interdigitate. Both qualitative and quantitative defects in platelet organelles were detected. The microtubular system was faulty in organization. Furthermore, the number of granules (especially alpha granules) was reduced. Platelet aggregation studies demonstrated subnormal aggregation in response to ADP, epinephrine, and collagen, but aggregation with ristocetin was normal. It is postulated that a platelet membrane abnormality may be the cause of their defective platelet aggregation.

Gel filtered human platelets were mixed with polylysine coated polyacrylamide beads. Platelet binding to individual beads within one preparation was variable, yet high and the bound platelets appeared to involve non-activated platelets. Some cells appeared to be in contact with more than one bead. When the attachment process was neutralized with 1 mg/ml of polyanionic dextran sulfate, significantly fewer platelets remained attached to the beads. Platelets which were recovered in the washes also were non-activated. Attached cells were capable of being activated and underwent release reactions when exposed to either ADP or thrombin. Release reactions were monitored by the liberation of either serotonin or beta-thromboglobulin. Platelets which were iodinated prior to attachment gave evidence that the failure to neutralize the bead before vortexing resulted in lower specific activities of bound iodine/protein.

A permanently implantable in-series left ventricular assist device, the dynamic aortic patch (DAP), has been tested in chronic animal experiments. The DAP replaces a section of the intrathoracic aortic wall. Hemothorax and hematocele at the implantation site have been complications in recent experiments. Primary postoperative hemorrhage was ruled out, and the biocompatibility of all components was therefore examined. Dacron velour, Teflon felt, conductive polyurethane, segmented polyether polyurethane, and Teflon-coated polyester fiber sutures were implanted in the pleural cavities of dogs and tested in vitro by culturing canine saphenous vein explants on them. In vivo experiments demonstrated that all components elicited mild to moderate inflammatory reactions, but hematocele occurred only when the components were implanted in the aorta with direct blood contact and exposed to arterial blood pressures. In vitro, cells were cultured on all components with no signs of toxic reactions. These results indicated that the host tolerated all implant components without major inflammatory responses. However, histological data indicated that chronic slow bleeding into or through the Dacron velour in contact with the arterial blood serum could account for hemothorax or hematocele formation. Therefore, a configuration of the assist device using materials impermeable to blood may obviate these difficulties.

Evidence is presented that 1-deamino-8-d-arginine vasopressin (DDAVP), a vasopressin analog, has a direct effect on isolated vessel segments. The most significant finding is increased platelet adhesion and spreading at injury sites. An isologous human umbilical vein perfusion model was used to compare effects of DDAVP with those of epinephrine or zero drug controls. Scanning electron microscopy, in conjunction with morphometry, permitted quantification of platelet adhesion to subendothelium exposed by minimal injury in the model. In addition, umbilical vein effluents were tested for levels of factor VIII moieties (F VIII:C, F VIII:Rag, F VIII:RCof) and the prostanoids, 6 keto PGF1 alpha (stable metabolite of prostacyclin) and TXB2 (stable metabolite of thromboxane A2. Only F VIII:C from DDAVP treated segments was significantly (p less than 0.01) changed from controls.

The purpose of this study was to evaluate, in vivo, the effect of perfluorochemical (PFC) blood substitutes on arterial endothelium following total blood replacement. Conscious-female-Sprague-Dawley rats (150-200 gms) were isovolumically exchange perfused with 3 blood volumes of PFC emulsion in an oxygen chamber. The exchange was performed at 0.5 ml/min via an indwelling-intracardiac-double-lumen catheter. One hour after the exchange, the animal was sacrificed and the circulatory system flushed free of blood and PFC emulsion with Dulbecco's solution, followed by perfusion fixation with glutaraldehyde solution. The heart, lungs and thoracic aorta were excised. The pulmonary arterial endothelial cell response was determined by scanning electron microscopy (SEM) following acute exposure to PFC emulsions under conditions of varied pH, oncotic pressure and emulsion age. These endothelial surfaces exhibited increased microvilli, leukocyte adhesion, fibrin deposition and subendothelium exposure. The magnitude of these abnormal responses varied with the degree of alkalinity and reduced oncotic pressure of the perfusate.

Male Hartly guinea pigs and Fischer rats 344 were exposed to diesel exhaust (DE) concentrations at 0, 250, and 1500 micrograms/m3 in short terms, as well as long term experiments up to one year. The effects of inhaled DE on these rodents were evaluated using bronchoalveolar lavage technique. Both the morphological and functional studies of free lung cells and the biochemical and immunologic studies of the supernatant lavage fluid provided the basis for a quantitative species comparison of the pulmonary responses of exposed guinea pigs and rats versus age matched controls. Following inhalation of 250 micrograms DE/m3, there were little or no significant changes in either species. In contrast, at higher DE concentration, leukocytic infiltration and elevation of specific proteins in lavage fluids were observed in both species. The findings occurred and persisted in both species. Some of the responses were species specific (e.g., the specific type of exudative leukocytes, appearance of reactive monocytes, and different amounts of free DE particles and debris in the lavage fluid). Other responses were similar in both species. Among them, the emergence and increase of lymphocytes was evidence of immunologic responses. Biochemical data from the supernatant fluid correlates with the changes in cellular population in the lavage. The responses appear to be dose and duration dependent. These data indicate that species differences occur. However, it is clear that the alveolar macrophage and granulocytic leukocytes continue to exert effective defense at the DE dose-durations studied. In general, rats appeared more resistant to DE exposure than guinea pigs.

Murine monoclonal antibodies (mAb) were raised to a purified product of bovine PF-4, a 9,500 dalton protein with heparin neutralization activity comparable to that of human PF-4. Using a non-radioactive slide immunoenzymatic assay, four major classes of mAb could be identified when comparisons were made between purified antigens of PF-4 and beta-TG-like protein from both bovine and human species. Type 1 cross-reacted with all four antigens; type 2 reacted with PF-4s; type 3 reacted with only bovine PF-4 and beta-TG-like protein; and type 4 reacted only with bovine PF-4. Differences in immunoreactivities of types 1, 2 and 3 were retained throughout the growth of succeeding clones and in ascitic fluids. Using a modified factor Xa, S-2222 chromogenic substrate-heparin inhibition assay, no mAb was found to block PF-4's ability to neutralize heparin. mAbs representative of types 1, 2 and 3 were successfully raised in stable cell lines from at least second generation clones. These were purified with protein A agarose and found to be IgG1. By indirect immunocytofluorescence a purified type 2 mAb, 2E7, was found to specifically stain granules of human platelets and megakaryocytes, as well as masses (putative platelets within late stage megakaryocytes) without staining other cellular types in either bone marrow or peripheral blood. Species comparisons displayed positive staining for human, rat, and rabbit platelets and megakaryocytes, and negative staining for mouse, guinea pig and dog platelets and megakaryocytes. It seems likely that mAb, 2E7, is directed against an epitope, common to PF-4 of bovine, human, rabbit and rat.

The mission of this study was to determine whether or not arteriovenous connections, indicative of a "closed" type of circulation, existed in the human spleen. Spleens from four patients requiring therapeutic splenectomy were the basis for this report. Scanning electron microscopy of plastic corrosion casts, prepared from these four spleens, revealed direct vascular conduits between splenic pulp arteries or arterial capillaries and the venous sinuses in the red pulp. Also demonstrated were a few arteriovenous shunts between pulp arteries or arterial capillaries and pulp or trabecular veins. Inclusion of sized microspheres in low-viscosity perfusion plastic illustrated that some diameters of the connecting shunts were 7-10 mum, with other shunts even smaller. Not only do arteriovenous connections exist in human spleens, but their frequency, as revealed by methods accentuating three-dimensional aspects of the splenic microcirculation, justify future reconsiderations of the functional significance of this closed type of circulation. Examination of samples of the same intact spleens, prepared by freeze-fracture and conventional critical-point drying, also revealed an "open" type circulation structure, namely, pore-patterned sinus walls that could facilitate blood cell movement from pulp cords into venous sinuses. Scanning electron microscopy thus has provided direct evidence that human spleens have both "open" and "closed" circulatory pathways in their microvasculature.

The relationship of polyploidization (DNA content) to differentiation is not well defined. We have developed centrifugal elutriation and Percoll density gradient centrifugation to obtain large numbers of highly-purified megakaryocytes which subsequently were stained for DNA content with Hoechst 33342 and sorted by FACS into 8C, 16C and 32C ploidy classes for correlated analysis of cell surface structures by scanning electron microscopy. Each ploidy class revealed unique surface characteristics that reflect differentiation occurring in megakaryocytes independent of their DNA content.

Scanning and transmission electron microscopy were used for a study of the surface and glandular ultrastructure of human endometrium in the presence of different types of IUDs at comparable phases of the menstrual cycle. The aim of the study was to compare the effect of the nonmedicated with the copper and multiload copper devices to further explain the differences in their contraceptive potencies and their mechanism of action. The endometrium was evaluated at and away from the IUD; emphasis was put on the ultrastructure of endometrial gland openings, secretory activity, cellular glycogen content, ciliated cells, microvillous pattern, and kinocilia. The changes of the surface ultrastructure of the endometrium in the presence of copper IUDs were more extensive in this study than those previously reported. There seems to be a direct relationship between the amount of copper incorporated in the device, the degree of ultrastructure changes, and the area of endometrium involved. Copper devices affect the endometrial cells away from the IUD. The altered secretory function with disturbed macroapocrine secretion, the abnormality of ciliated cells, and the defective microvillous growth seem to interfere with the physiologic and functional integrity of the endometrium, reducing the chances of contraception in the presence of copper IUDs.

Transmission electron microscopy (TEM) was used to determine the effect of diesel engine exhaust (DEE) on the intravascular and interstitial cellular population of the lungs of exposed rats and guinea pigs. Animals with matched controls were subjected to environments of either 250, 750, 1500 or 6000 micrograms/m3 for either 2 weeks, 6 weeks, 10 weeks or 18 months. These animals were sacrificed immediately following the exposure periods and their lungs perfused with fixative. Following dissection, random stratified biopsies from the lungs of these animals were made. Ultrathin sections from the alveolar lung were prepared and conventionally processed for TEM and randomly photographed to compose a micrograph database. These micrographs were analyzed by point counting using a Zeiss MOP 3 Digital Image Analyzer. The results indicated no significant intravascular cellular response but a significant increase in the mononuclear population in the interstitium.

The effects of thrombin stimulation on megakaryocytopoiesis and pulmonary-platelet interactions were investigated before and after administration of the compound to 15 mongrel dogs. Each dog served as its own control. Thrombin was given to encourage the traffic of megakaryocytes into the lung and to study the thrombin-stimulated effects on megakaryocytopoiesis in the bone marrow. Our results showed that thrombin increased the numbers of bone marrow cells in general and megakaryocytes (MK) in particular. In addition, the maturation cycle of megakaryocytes was accelerated and the number of MK migrating into the central venous circulation was nearly doubled. Most of the circulating MK ultimately became sequestered in pulmonary capillaries, where platelets were shed into the arterial circulation. We conclude that thrombin has a major stimulatory effect on megakaryocytopoiesis in the bone marrow and that the lung plays an important role as a vascular filter and regulator of circulating platelet count.

The structural physiology of the human spleen is discussed in depth here, with particular emphasis on the interrelationships between blood cells and the spleen. The unique microvasculature and structural relationships of the spleen form the basis for a number of functions important in both health and disease. Utilizing scanning electron microscopy one can visualize the interactions between blood elements and the spleen---the passage of deformable erythrocytes through narrow slits in the endothelium separating cord and sinus, the trapping of less deformable cells in the cords, the culling and pitting of abnormal or inclusion-bearing erythrocytes by splenic macrophages, as well as the sites of interaction between antigenic stimuli such as pneumococcal organisms and the T and B cells of the spleen.

The in vivo effects of inhalation of diesel engine exhaust (DEE) on pulmonary alveolar macrophages (PAM) was studied in 73 guinea pigs and 48 rats. Animals were exposed in individual cages in special chambers to 3 different dose levels of DEE expressed in terms of the concentation of soot or carbon particles (-P); 250, 1500, 6000 micrograms DEE-P/M3. Exposure durations for guinea pigs were 1 and 3 days, 1 and 2 weeks, 2, 4, 8 and 12 months while rats were exposed 1, 2, 4, 8 and 12 months. Age matched controls were similarly exposed concurrently to "clean" air. PAM obtained by bronchopulmonary lavage from exposed animals had viabilities comparable to controls. PAM diameters and relative surface areas increased 2--3 fold over controls and in relation to both the dose of DEE-P given and the exposure duration. Most of the in vivo exposed PAM had phagocytized DEE-P which did not appear to be cytotoxic and remained confined in phagosomes as discrete particles with diameters of 0.014--0.072 micrometer. Ability of PAM to adhere and spread on test surfaces was greater in the DEE-P sets than in controls. DEE-P containing PAM were still able to phagocytize latex particles when fed in vitro. However, such PAM had defective phagocytosis ability, and did not in the same time interval take up as much fluorescent latex as controls when studied by flow system technology. Absolute numbers of PAM in guinea pig lavages from exposures to 250 and 1500 microgram DEE-P/M3 for 2 months were not significantly changed over concurrent controls. Exudative leukocytes (eosinophils in guinea pigs and neutrophils in rats) appeared in the lavage in greater numbers as dose and duration of exposure increased. Another species difference was the appearance in DEE-P exposed guinea pig lavages of "reactive" monocytes.

The effects of fibrin microembolism were examined using an infusion of a prothrombin activator (Echis carinatus venom, ECV; 30 min, 0.5 NIH thrombin equivalent units/kg) in acute mongrel dogs prepared with a pulmonary lymph cannula (n = 6, 12.3-21.5 kg). Lymph flow increased approximately 2.5-fold after 1-1.5 hr of elevated left atrial pressure (Pla = 20 cm H2O; 26 +/- 7 to 63 +/- 16 microliter/min, P less than 0.01) and the plasma to lymph protein concentration ratio (CP/CL) declined from 0.66 +/- .04 to 0.54 +/- .16 (P less than 0.01, x +/- SE). After Pla was reduced to control levels, the initiation of fibrin microembolism was associated with an approximate 2.7-fold elevation of lymph flow (62 +/- 8 microliters/min, P less than 0.01) and the CP/CL was not changed (0.56 +/- 0.04, P = ns). When Pla was increased following microembolism, lymph flow more than doubled to 117 +/- 24 microliter/min (P less than 0.01) and the CP/CL remained unaltered (0.56 +/- 0.03, P = ns). These changes were associated with afibrinogenemia and the appearance of fibrin degradation products (FDP) in plasma (150 +/- 50 micrograms/ml) and lymph (80 micrograms/ml) in three of the animals tested. No consistent pattern was seen in the CL/CP of separate endogenous plasma proteins after each intervention. These data support the view that pulmonary fibrin microembolism without inhibition of the fibrinolytic system was associated with an early increased pulmonary microvascular permeability to protein. In a separate group of similarly prepared animals (n = 8, 13-21.5 kg) without a lymph catheter, scanning electron microscopic observations showed branching fibrin microemboli to partially occlude some pulmonary arterioles. Mixed thrombus formations in larger precapillary blood vessels were also seen. Ultrastructural observations revealed the deposition of fibrin strands (periodicity = 220-230 A) within the pulmonary capillaries. Some of these deposits were overlaid by lamellar pseudopodia from endothelial cells and the fibrin appeared to be within these cells. Although plasmalemmal vesicles seemed to be more numerous in the endothelial cells with adjacent fibrin deposits, no gaps or breaks were seen in the densely stained interendothelial cell junctions and/or the endothelial cell membrane of the affected lung capillaries. Activated neutrophils and platelets were more numerous in the pulmonary capillaries following EVC. These data suggest that the presence of FDP and/or fibrin deposits within the pulmonary microvasculature may influence the early functional integrity of pulmonary endothelial cells at sites of fibrin accumulation.

We have identified a number of patients with coronary artery disease, TIAs and/or strokes and idiopathic ischemic optic neuropathy who have a platelet population which is in vitro hyperaggregable with epinephrine and ADP and hyperresponsive to surface contact. These patients have no identifiable risk factors. Several families have been identified in which multiple members had these findings. Many, but not all have had clinical symptoms. An autosomal (dominant) pattern of heredity seems to evolve. We refer to this as "sticky platelet syndrome" and hypothesize that it may represent a congenital platelet abnormality which potentially predisposes to thromboembolisms. The precise nature of the defect is not known at this time.

Human megakaryocytes (MK) from rib marrows of 16 patients with normal hematological patterns were isolated by Percoll density centrifugation followed by velocity sedimentation. The surface features of freshly isolated MK and the unseparated marrow suspensions were examined by SEM. A wide range of surface morphology for MK was noted depending on their size and maturation stages. Megakaryoblasts had a relatively smooth surface. During maturation, MK became villous and some had small blebs. In late maturation, many MK had complex surface processes including typical discs, slightly sphered projections, tear drop projections, putative sheet-like platelets or combinations of two or more of such surface features. Both unseparated and MK-rich fractions (F); specifically F1 lower, F2 lower and F3 were cultured up to 19 days. MK maintained their spherical shape in culture. At different time intervals (2, 5, 9 and 19 days), some cultures were terminated. MK retained their characteristic surface features. Promegakaryocytes and young adult MK were seen in short term cultures. The majority of cultured MK had platelet-like bodies on their surface in the prolonged cultures. Sometimes a few morphologically recognizable platelets were seen in the culture media. Differential counts of MK were done in all preparations of both unseparated and MK enriched fractions. F3, having no morphologically recognizable MK, had many MK present after culturing in vitro for 19 days. These MK had typical platelets or platelet-like processes on their surface. This suggests some MK progenitors originally present in the isolated F3 fraction became mature in cultures.

The impact of chronic inhalation of diesel exhaust (DE) on alveolar lung was studied in 24 Hartley guinea pigs. Groups were sacrificed sequentially at 2 weeks, 3 months and 6 months of exposure to either 750 micrograms or 1500 micrograms DE particles (DEP) per m3 along with age-matched concurrent controls. Although qualitative ultrastructural changes were noticed during this time interval and dosage schedule, there was no evidence of pathologic changes such as fibrosis or emphysema. The cellular uptake of DEP was striking. By 2 weeks three alveolar cell types (alveolar macrophages, epithelial type 1 cells and interstitial macrophages) plus one type of granulocytic leukocyte (eosinophils) confined DEP within phagosomes without evidence of cytotoxicity. A certain phagosomal DEP population had a bull's eye appearance and diameters of 0.0727 +/- 0.01 micron. Morphometric analysis applied to electron micrographs demonstrated that arithmetic mean tissue thickness of the air-blood barrier was generally increased (p less than 0.05) during DE exposure. For the 750 micrograms DE sets, the increase over control (1.56 micron) was 41% at 2 weeks, 46% at 3 months and 77% at 6 months while the 1500 micrograms DE set at 6 months exceeded control by 130%. Increases in absolute tissue volumes of interstitium and epithelial type 2 cells largely accounted for the increased tissue thickness. Harmonic mean tissue thickness for controls remained near 0.537 +/- 0.03 micron for the study interval, contrasting with values for 3 and 6 months 750 micrograms DE and 6 months 1500 micrograms DE sets which increased. However, the diffusion capacity of the lung determined morphometrically was not decreased in DE exposed sets. Although cellular uptake of DEP and increased prominence of secretory epithelium were dose/duration related, absence of linearity suggests adaptative responses.

This article summarizes recent ultrastructure findings from our laboratory and documents some of the information accumulated primarily since 1975 from many laboratories. Special attention is given to documentation by scanning electron microscopy which affords insight into platelet activation (adhesion, aggregation, release/secretion) and especially platelet-vessel wall interactions. Structural physiology of platelets is considered in some detail as a basis for understanding platelet disorders contributing to clinical problems of thrombosis and hemorrhage. The impaired ability of von Willebrand platelets to adhere to injured vessel wall is reported using the human umbilical vein perfusion model. Relationships between platelets and blood coagulation factors focus on the exquisite sensitivity of platelets to minute amounts of thrombin. Unmasking of platelet factor 3 sites is identified on activated platelets, after glutaraldehyde fixation, by their reaction ot latex bearing anti-platelet factor 3 markers. The basis for platelet-collagen interactions is reviewed. Conditions for and possible mechanisms behind platelet interaction with vessel wall are discussed. Ex vivo flowing blood-vessel wall models offer opportunities for improved understanding of the platelets role(s) in vascular diseases.

The three-dimensional world of the spleen was explored by scanning electron microscopy on both arterially perfused and nonperfused specimens, as well as on plastic corrosion casts of splenic vasculatures. Of 25 spleens studied, 18 were examples of hypersplenism. These were contrasted to 7 essentially normal spleens taken from children being staged for treatment of Hodgkin's disease whose spleens proved to be uninvolved in the pathologic process. Splenic sinuses in all 25 spleens were typified by a degree of porosity. RBC were caught in the act of entering sinuses from splenic cords. These sinus windows thus represent one end of an "open" circulation pathway. In casts of microvasculature, direct arteriovenous connections were demonstrated, thus establishing an anatomical basis for an often disputed "closed" circulation pathway. Spleens from 7 patients with hereditary spherocytosis had a super abundance of red pulp. Splenic cords were thickened and crowded with spherocytes, many of which presented slightly wrinkled membranes, as were also noted on the peripheral blood RBC. It is possible that these membrane features are unique for HS and reflect the intrinsic membrane abnormality in protein composition. The 7 spleens from chronic idiopathic thrombocytopenic purpura had white pulp as the predominant region. Germinal centers were frequent. Lymphocytes and plasma cells with well-developed microvilli were suggestive that release of antiplatelet antibody might be occurring in white pulp. Platelets were especially notable in peripheral white pulp and marginal zones. Platelet clumps were observed, generally adjacent to spleenic macrophages.