HARUMICHI SHINOHARA* Division of Human Sciences, Faculty of Medicine, Toyama Medical and Pharmaceutical University, Toyama, Japan

Transcription

1 THE ANATOMICAL RECORD 249:16 23 (1997) Distribution of Lymphatic Stomata on the Pleural Surface of the Thoracic Cavity and the Surface Topography of the Pleural Mesothelium in the Golden Hamster HARUMICHI SHINOHARA* Division of Human Sciences, Faculty of Medicine, Toyama Medical and Pharmaceutical University, Toyama, Japan ABSTRACT Background: The distribution of lymphatic stomata that open to the pleural cavity is unclear. The distribution and the surface topography of the pleural and visceral pleurae are key factors in the turnover of pleural fluid and respiration physiology. Methods: Nine golden hamsters (Mesocricetus auratus) from 26 to 33 weeks of age were used for the study. The gross anatomy of the thorax and the arterial supply to the lung were studied in four hamsters. Five thoracic hemispheres, three diaphragms, and tissue blocks of the heart and lung were prepared from the remaining five hamsters. The thoracic hemispheres were fixed in 2.5% glutaraldehyde and the muscular bands at each intercostal space were carefully cut along the costae. The intercostal bands were processed for scanning electron microscopy (SEM) and the localization and the number of lymphatic stomata were recorded. The diaphragms and blocks of the lung and heart were also processed for SEM and the surface topography was observed. Results: The right and left superior lobes of the lung were supplied by the bronchial artery that originated from the right costocervical trunk and left internal thoracic artery, respectively. Lymphatic stomata and mesothelial discontinuities (pores and gaps) were predominantly located in areas lined with cuboidal cells. The areas of cuboidal cells occupied approximately 4.6 mm 2, namely, 1% of the total area of the thoracic hemisphere. There were about 1,000 lymphatic stomata per thoracic hemisphere. About 15% of lymphatic stomata were distributed in the ventro-cranial regions of the thoracic wall, with about 85% in the dorsocaudal region. In the former region, lymphatic stomata were found along the costal margins. In the latter, they were predominantly located in the pre- and paravertebral fatty tissue. There were also areas of cuboidal cells on the pleural surface of the diaphragm. Some mesothelial pores and gaps were found, but no lymphatic stomata opened on the pleural surface of the diaphragm. The pleural surface of the lung and that of the heart were lined with flattened polygonal cells. The topography of the surface varied, but there were no mesothelial discontinuities of the type commonly found in the parietal pleura. Conclusions: 1) The parietal pleura has a surface structure that is more permeable and absorptive for fluid and particulate matter than the visceral pleura. 2) The distribution of lymphatic stomata does not correspond directly to the pleural liquid pressures that have been reported. 3) The functions of lymphatic stomata should be considered not only in terms of fluid turnover but also in terms of self-defense mechanisms. 4) The presence or absence of lymphatic stomata on the diaphragmatic pleura should be re-examined and determined in a variety of animal species. Anat. Rec. 249:16 23, Wiley-Liss, Inc. *Correspondence to: Harumichi Shinohara, Division of Human Sciences, Faculty of Medicine, Toyama Medical and Pharmaceutical University, Sugitani 2630, Toyama , Japan. Received 25 November 1996; accepted 10 April WILEY-LISS, INC.

2 LYMPHATIC STOMATA IN THE PARIETAL PLEURA 17 Key words: lymphatic stomata; parietal pleura; visceral pleura; golden hamster There is a small amount of fluid in the pleural space, and the controversy about the turnover of this fluid has not been resolved. Agostoni et al. (1991) proposed that permeation of tissue fluid across the pleurae, as expressed by Starling s equation (Starling and Tubby, 1894), plays a major role in both the production and the elimination of pleural fluid. They proposed that pleural fluid permeates the pleural cavity from the parietal pleura and enters the visceral pleura to leave from the pleural cavity. Lymphatics are involved only in the egress of cell components and large molecules from the pleural cavity. By contrast, Broaddus and Staub (1987) suggested that permeation across the pleurae plays a major role in the production of the fluid but that most of the fluid in the pleural cavity is drained via the lymphatics of the parietal pleura. They estimated that pleural fluid permeates the pleural cavity through the parietal and visceral pleurae at 0.01 ml/h/kg, which corresponds to about 12 ml per day in a man with a body weight of 50 kg, and that an almost equal amount of fluid leaves the cavity through the lymphatics. When the amount of the fluid is increased (for example, in cases of hydrothorax), the exit rate via the lymphatics increases 20- to 30-fold. However large or small the lymphatic contribution to the turnover of pleural fluid might be, lymphatic stomata are the major route for the draining of fluid and particulate matter from the pleural cavity. There are some reports (Wang, 1975; Pinchon et al., 1980; Mariassy and Wheeldon, 1983; Albertine et al., 1984; Negrini et al., 1991; Abu-Hijleh, 1995) that describe the localization of lymphatic stomata in the pleurae of various species of animals. However, the distribution of these stomata is still unclear, or even confusing (see Discussion). The present report includes a diagram of the distribution of lymphatic stomata that open on the pleural surface of the thoracic wall of the golden hamster. It also includes a description of the surface topography of the parietal pleura and visceral pleura of the lung and heart. MATERIALS AND METHODS Nine golden hamsters (Mesocricetus auratus) from 26 to 33 weeks of age were used for the study. Four golden hamsters were killed by inhalation of ether vapor for studies of the gross anatomy of the thorax and the arterial supply to the lungs. The other five animals were also killed with ether, and five thoracic hemispheres (three left sides and two right sides), three diaphragms, and tissue blocks of the lung and heart were prepared. The thoracic hemispheres were fixed in 2.5% glutaraldehyde in 0.1 M phosphate-buffered solution (PBS), ph 7.4, for 2 hr. The muscular bands at each intercostal space were carefully cut along the costae. The intercostal bands were rinsed in PBS, postfixed in 1% OsO 4, dehydrated in a graded ethanol series, freezedried in butanol, mounted on aluminum stubs, ionsputter coated and viewed with a scanning electron microscope (SEM). The lymphatic stomata usually open in an area of cuboidal cells (Tsilibary and Wissig, 1983). Therefore, the areas of cuboidal cells were initially surveyed and photographed at a magnification of 30 80X, and the extent of each area ( Amm 2 ) of cuboidal cells was calculated from photographs using the NIH image program. Three to five different regions of each area of cuboidal cells were viewed at a magnification of 400X and the mean number of lymphatic stomata per visual field ( B/mm 2 ) was recorded. The number of lymphatic stomata per area of cuboidal cells was obtained by AxB. The number of lymphatic stomata per thoracic hemisphere was the total of the number in each cuboidal cell area of 12 or 13 intercostal bands. Also, photographs were taken of the intercostal bands on aluminum stubs and the total area surveyed by SEM was calculated using the NIH image program. The diaphragms and tissue blocks of the lung and heart were also processed for SEM and examined. The right phrenic nerve reached the pleural surface of the diaphragm just lateral to the inferior vena cava, and the left phrenic nerve reached the left side of the diaphragm. Their cut-ends were good landmarks for discrimination of the pleural surface from the peritoneal surface. The attachments of the pericardium, mediastinal pleura, and falciform and other ligaments were also useful landmarks. The entire pleural surface of the diaphragm was surveyed, but the area surveyed and the relative areas of cuboidal cells were not determined. RESULTS Gross Anatomy of the Pleural Thoracic Wall There were usually 13 but occasionally 14 costae. Therefore, the thoracic wall was divided circumferentially into 12 or 13 intercostal bands. The thoracic wall was also divided into four longitudinal zones in parallel with the vertebral column: the zone of pre- and paravertebral fatty tissues; the zone of subcostal muscle; the zone of innermost intercostal muscle; and the zone of transverse thoracic muscle (Fig. 1). SEM revealed numerous wrinkles perpendicular to the muscle fibers in the three zones of muscle, whereas no wrinkles were seen in the zone of the fatty tissue. The presence or absence, as well as the orientation, of the wrinkles made it easier to define the localization of the areas of cuboidal cells and stomata. Gross Anatomy of the Bronchial Artery That Supplies the Lung On the right side, the costocervical trunk originated at the proximal portion of the subclavian artery. From the costal branch of the trunk, an artery ran caudally for several millimeters along the right cardiac border. At the level of the right pulmonary hilum, the artery split into several branches and at least one of them entered the superior lobe of the lung, passing through the hilum with the pulmonary artery and bronchus. On the left side, a branch of the left internal thoracic artery ran caudally for several millimeters and split into several twigs at the hilum of the lung. At least one of the twigs entered the superior lobe. Thus, the right bronchial artery was the terminal branch of the ipsilateral costocervical trunk, whereas the left bronchial

3 18 H. SHINOHARA No. of thoracic hemispheres Area of observation (mm 2 ) TABLE 1. Area of cuboidal cells (mm 2 ) No. of stomata per hemisphere * * 1, * *Mean S.D. Fig. 1. A diagram illustrating the distribution of lymphatic stomata in the hemisphere of the thoracic wall. The thoracic wall is divided circumferentially into 12 or 13 intercostal spaces and longitudinally, in parallel with the vertebral column, into four zones the zone of preand paravertebral fatty tissue (A), the zone of subcostal muscle (B), the zone of innermost intercostal muscle (C), and the zone of transverse thoracic muscle (D). The number in each intercostal space is the total number of lymphatic stomata found in a total of five thoracic hemispheres. Dots indicate approximate locations of lymphatic stomata, but they do not always signify the number of stomata. Note that approximately 85% of lymphatic stomata are located in zones A and B of the last three or four intercostal spaces. artery was the terminal branch of the ipsilateral internal thoracic artery. These observations were made in the four hamsters examined and were compatible with observations of the bronchial artery in the rat (Hebel and Stromberg, 1986). Distribution of Lymphatic Stomata, Mesothelial Pores, and Gaps The observed area of the intercostal bands and the areas of cuboidal cells and the total number of lymphatic stomata per thoracic hemisphere are summarized in Table 1. In the zone of innermost intercostal muscle, the inner surface on the costae was covered only with the parietal pleura and was not examined over its complete area. The area observed, (mean SD) mm 2, was estimated to encompass about 90% of the total pleural surface of one thoracic hemisphere. Most of the parietal pleura was lined with flattened polygonal cells, but an area of 4.6 mm 2 per thoracic hemisphere was lined with cuboidal cells. The lymphatic stomata were consistently located in cuboidal cell areas. A variety of mesothelial discontinuities (see below) was also found in areas of cuboidal cells, but these discontinuities were rarely found in the area of flattened cells. The total number of lymphatic stomata found in five thoracic hemispheres was 5,265. The respective right hemispheres had 796 and 1,468 stomata, and the left had 256, 857, and 1,888 stomata. The mean number of stomata ( SD) per thoracic hemisphere was 1, The distribution of lymphatic stomata in the thoracic hemisphere is shown diagrammatically in Figure 1. There were two patterns of distribution. From the 2nd to the 9th intercostal space, approximately 15% (793/ 5,265) of stomata were distributed principally in the ventral half of the thoracic wall along the costae (Fig. 2). In the caudal three or four intercostal spaces, the remainder of the lymphatic stomata (85%, 4,472/5,265) were located in the dorsal half, predominantly in the zone of paravertebral fatty tissue, of the thoracic wall (Fig. 3). As described previously (Leak and Rahil, 1978; Nakatani et al., 1986), the typical orifice of a lymphatic stoma under the SEM consisted of two margins, an outer microvillous margin formed by a perimeter of cuboidal mesothelial cells and an inner, relatively microvillus-free margin, formed by lymphatic endothelial cells (Fig. 4). Openings that equip the two margins are defined as the orifice of lymphatic stomata and only such openings were counted. It was not unusual to encounter a free cell, a lymphocyte in most cases, at the margin of lymphatic stomata. As shown in Figure 5, some of these cells appeared to be going into the lymphatic lumen (there was no reason to think that they were coming from the lumen and against the flow). Lymphatic stomata were not always found in areas of cuboidal cells; however, even areas of cuboidal cells that did not have lymphatic stomata had mesothelial discontinuities. Collagen fibers and other components of the submesothelial connective tissue were visible through these discontinuities. The discontinuities that were circular in shape were defined as pores (Fig. 6) and those irregular in shape were defined as gaps (Fig. 7) in the present study. The connective tissue fluid and pleural fluid that contain proteins and other macromol- Fig. 2. An area of cuboidal cells (*) on the cranial margin of the fourth intercostal band (right side). The wrinkles of the innermost intercostal muscle are perpendicular to the muscle fibers (arrows). Scale bar 100 µm. Fig. 3. An area of cuboidal cells in the 11th intercostal band (left side). Cuboidal cells (*) extend diagonally from the lower left to upper right along the muscle fibers of the subcostal muscle (arrows). Stomata are also present in these cuboidal cell extensions. Therefore, lymphatics are assumed to be present beneath them. The mean number of stomata in this area of cuboidal cells was 14 per visual field at a magnification 400X. Scale bar 100 µm. Fig. 4. A lymphatic stoma. Note that the outer margin of the orifice is composed of a perimeter of microvillous mesothelial cells and the inner margin (*) is composed of microvillus-free lymphatic endothelium. Scale bar 1 µm. Fig. 5. A free cell, probably a lymphocyte, on a stomatal orifice. Scale bar 1 µm.

4 LYMPHATIC STOMATA IN THE PARIETAL PLEURA 19 Figs. 2 5.

5 20 H. SHINOHARA ecules are assumed to pass freely through these discontinuities. The pleural surface of the diaphragm was, for the most part, lined with polygonal mesothelial cells, but it was lined with cuboidal cells in some parts. Gaps and pores were present in these areas of cuboidal cells, but no open lymphatic stomata were seen. These observations confirmed the results of a previous study (Fukuo et al., 1990). Negrini et al. (1991) proposed photomicrographs of rabbit stomatal orifices that appeared to be formed of only (pleural or peritoneal?) mesothelial cells. Abu-Hijleh et al. (1995) suggested that the stomatal structure consisting of the two margins, inner lymphatic endothelial and outer mesothelial, is not necessarily the absolute morphological criteria for the pleural stomatal orifices of the diaphragm. On the pleural surface of the golden hamster diaphragm, neither stomatal orifices with the two margins nor any other mesothelial fenestrations that are compatible with atypical lymphatic stomata were found. The pleural surface of the lung was lined with flattened polygonal cells (Fig. 8). Microvilli were usually short. In some regions microvilli were sparse, whereas in other regions they were densely distributed and cell borders were unclear. The pleural surface of the heart was also covered mostly with flattened cells with sparse microvilli (Fig. 9). Although the surface topography of the lung and heart varied from region to region, in terms of the density and length of microvilli and the absence or presence of undulations, there were no mesothelial discontinuities on the visceral pleura of the type seen in the parietal pleura. DISCUSSION The classical hypothesis about the turnover of pleural fluid was summarized by Agostoni and Mead (1964). The parietal pleura is supplied by the intercostal arteries ( systemic circulation), while the visceral pleura is supplied by the pulmonary arteries ( pulmonary circulation). The difference in arterial blood supply between the two pleurae, 30 cm and 11 cm of H 2 O, produces a delicate balance of Starling s forces by which a few milliliters of pleural fluid is maintained as a net total of inflow from the parietal pleura and outflow to the visceral pleura. McLaughlin et al. (1961) examined the lungs of various mammals and showed that the visceral pleura is supplied either by the pulmonary artery or bronchial artery ( systemic circulation) or by both. Albertine et al. (1982) demonstrated that the visceral pleura in sheep is supplied exclusively by the bronchial arteries, and they claimed that the thick pleura of large animals is supplied by the bronchial artery, while the thin pleura of small animals is supplied by the pulmonary artery. Thus, theoretically, in some animals the balance of Starling s forces favors inflow of fluid from both the parietal and visceral pleurae into the pleural cavity, such that the pleural cavity should be filled with fluid. Some investigators have indicated that the difference in the arterial blood supply should not affect the balance of Starling s forces, since the capillary pressure is low and the capillary blood drains finally to the pulmonary veins, regardless of the source (Broaddus and Staub, 1987; Light, 1991; Agostoni and D Angelo, 1991). The bronchial artery of the golden hamster is a terminal branch of the subclavian tributaries and is not a branch of the aorta, as in other animals (McLaughlin et al., 1961). Therefore, its capillary pressure might not be very high. The golden hamster is small in body size. Its visceral pleura might be supplied by the pulmonary artery and the classic situation might be applicable without revisions. The pleural fluid contains small amounts of proteins, 1.0 gm/dl in sheep (Wiener-Kronish et al., 1984), 1.3 gm/dl in rabbits (Miserocchhi and Agostoni, 1971), and 1.4 gm/dl in dogs (Stewart and Burgen, 1958). The origin of this low concentration of protein is, however, unclear. According to Payne and Kinasewitz (1985, 1988), the parietal and visceral mesothelia are far more permeable than the capillary endothelium and, moreover, the permeability of the parietal mesothelium to albumin and other large molecules is far greater than that of the visceral mesothelium. Thus, Broaddus and Staub (1987) proposed that large molecular components in pleural fluid might be derived from the capillaries in the submesothelial connective tissue of the thoracic wall. The present study revealed that fluid in the parietal submesothelial connective tissue and pleural fluid diffuse freely through the mesothelial pores and gaps. Although the areas of cuboidal cells with mesothelial discontinuities account for only 1% or so of the total parietal pleural surface, they surely provide a morphological basis for migration of proteins and macromolecules into the pleural fluid. There are two forces pleural surface pressure and pleural liquid pressure that are subatmospheric and pull the lung to the thoracic wall against the tissue recoil of the lung. The pleural liquid pressure is defined as the local pressure produced by the thin film of fluid between the thoracic wall and the lung (Agostoni and Mead, 1964). The thinner is the film locally, the lower is the pleural liquid pressure relative to atmospheric pressure and, thus, the lung is more strongly pulled to the thoracic wall. In general, pleural liquid pressure is known to be lower in the cranial part than the caudal part of the thorax and in the mediastinal part than the costal part of the cranial thorax (Agostoni and Mead, 1964; Agostoni and D Angelo, 1969). Negrini and colleagues (Miserocchi et al., 1981; Negrini et al., 1983) suggested that lymphatic absorption of the pleural fluid might reduce the thickness of the pleural fluid film and decrease the pleural liquid pressure still further. Agostoni and associates (Agostoni and D Angelo, 1991; Agostoni and Zocchi, 1995) disagreed with Negrini et al., pointing out a contradiction the pleural liquid pressure in the caudal part of the thorax is not low, in spite of the distribution of numerous lymphatic stomata in the caudal thorax (Albertine et al., 1984). In the present study, about 85% of lymphatic stomata were located in the dorso-caudal part of the thorax, while the remainder were distributed in the ventro-cranial part. This distribution of lymphatic stomata does not correspond to the pleural liquid pressure in the thorax. Therefore, it seems to support the claim by Agostoni and associates that pleural liquid pressure is not a direct product of lymphatic absorption. One issue that has to be taken into account is that the pericardium in rodents is not a continuous serous membrane that completely separates the pericardial and pleural cavities, but is fenestrated by numerous circular fenestrations; the pleural

6 LYMPHATIC STOMATA IN THE PARIETAL PLEURA 21 Fig. 6. Mesothelial pores. The pores are circular discontinuities in the mesothelium. Connective tissue components (*) are visible through them. Scale bar 1 µm. Fig. 7. Mesothelial gaps. The gaps are mesothelial discontinuities with irregular margins (arrows). Scale bar 1 µm. Fig. 8. The pleural surface of the lung (left superior lobe). The cell borders are delineated by microvilli. The cells are flat and polygonal. Scale bar 10 µm. Fig. 9. The pleural surface of the heart (right ventricle). The cell borders are unclear. Microvilli are short and sparse. Scale bar 1 µm. and pericardial cavities communicate with each other through these fenestrations (Nakatani et al., 1988). Thus, pleural liquid pressure and also, probably, pleural surface pressure cannot be discussed without regard to the influence of pressure from the pericardial cavity in rodents. There is no doubt that lymphatic stomata play an important role in the egress of pleural fluid, but this is only half of their role. It is well known that there are many aggregates of free cells in the thoracic cavity (Kampmeier, 1928; Mixter, 1941). These aggregates or milky spots consist predominantly of lymphocytes, macrophages, and mast cells. When frog erythrocytes were injected into the pleural cavity of mice, the erythrocytes were surrounded and phagocytosed by macrophages in milky spots on the pericardium within 5 min of their injection (Fukuo et al., 1988). In the present study, lymphocytes were often located on the orifice of lymphatic stomata and some lymphocytes seemed to be moving, possibly, into the lymphatic lumen (Fig. 4). These observations suggest that free cells leave the milky spots on the parietal pleura, patrolling the pleural cavity and entering the lymphatics via the stomata. In other words, the lymphatic stomata function as exits for these migrating cells under physiological conditions. Recently, Negrini et al. (1991, 1992, 1993) reported the presence of lymphatic stomata on the pleural side of the diaphragm in the rabbit and discussed their physiological significance. The presence of the stomata had

7 22 H. SHINOHARA already been reported by Wang (1975). However, the reported densities of lymphatic stomata were very different: several or so per mm 2 by Wang vs. 72 per mm 2 by Negrini et al. The present author and colleagues have explored the stomata in the golden hamster (Fukuo et al., 1990; present study), as well as in the mouse and rat (unpublished data), but have never encountered convincing structures that resemble lymphatic stomata on the pleural side of the diaphragm. Our data are not consistent with those reported in the rat by Wang (1975) and by Pinchon et al. (1980). Mariassy and Wheeldon (1983) reported the opening of lymphatic stomata on the pleural surface of the diaphragm in sheep, but Albertine et al. (1984) did not find such stomata in the same species. A similar inconsistency is present in the diaphragm of the rabbit (Gaudio et al., 1990; Negrini et al., 1991). These inconsistencies among researchers may be attributable to differences between species or strains of experimental animals (Abu-Hijleh et al., 1995). The present author does not insist that there is no species-to-species difference, but thinks the disaccord of the presence or absence in one species of animal inconvenient. The presence of lymphatic stomata is determined primarily by morphological demonstration. For such demonstration, investigators should pay special attention to preparation of specimens for electron microscopy, lest the peritoneal side on which lymphatic stomata are commonly present be mistaken for the pleural side. Specimens for transmission electron microscopy are usually small and black, as a result of osmification, so it is almost impossible to distinguish the pleural surface from the peritoneal surface after sampling. As demonstrated by Fukuo et al. (1990), cutting specimens into trapezoidal shapes is one way to solve this problem. In relatively large specimens for SEM, as noted in the present study (see Materials and Methods), some landmarks for discrimination of the pleural surface from the peritoneal surface are essential. In SEM, it is rare to find a lymphatic stoma that opens solitarily. For identification of a mesothelial opening as a lymphatic stoma, the morphological criteria of the orifice margins (see Results) should be rigidly applied. Whether they are on the peritoneal diaphragm or in the ovarian bursa of the golden hamster, lymphatic stomata allow passage of india ink immediately after injection into the body cavity, and staining of the lymphatics begins within 5 min after injection (Shinohara et al., 1985; Fukuo et al., 1990). In other words, if no lymphatics are stained black within 5 min after injection, investigators should question the presence of lymphatic stomata and, at the same time, check the design of their experiment. There are mesothelial discontinuities on the pleural surface of the diaphragm, as described in the present study, and colored tracers can diffuse rapidly into the subpleural connective tissue through them, reaching the diaphragmatic lymphatics as time passes after injection. Thus, the time interval between injection of tracers and observation is critical in this type of experiment. If these methodological requirements are not fulfilled in future experiments, the results will further complicate the present state of confusion. ACKNOWLEDGMENTS The author is grateful to Professor Nariko Takano in the Division of Health and Physiology, University of Kanazawa, for her expert suggestions related to respiratory physiology. This work is dedicated to Dr. Sumiko Magari, emeritus professor of the Osaka Medical College, and Dr. Shigeo Uchino, emeritus professor of the Tokyo Medical College, in token of their retirements and my personal gratitude. LITERATURE CITED Abu-Hijleh, M.F., O.A. Habbal, and S.T. Moqattashi 1995 The role of the diaphragm in lymphatic absorption from the peritoneal cavity. J. Anat., 186: Agostoni, E., and E. D Angelo 1969 Thickness and pressure of the pleural liquid at various heights and with various hydrothoraces. Respir. Physiol., 6: Agostoni, E., and E. D Angelo 1991 Pleural liquid pressure. J. Applied Physiol., 71: Agostoni, E., and J. Mead 1964 Statics of the respiratory system. In: Handbook of Physiology, Sect. 3, Vol. 1. W.O. Fenn, H. Rahn, eds. American Physiological Society, Washington D.C. Agostoni, E., and L. Zocchi 1995 Mechanisms involved in pleural liquid turnover. J. Applied Physiol., 78: Albertine, K.H., J.P. Wiener-Kronish, P.J. Roos, and N.C. Staub 1982 Structure, blood supply, and lymphatic vessels of the sheep s visceral pleura. Am. J. Anat., 165: Albertine, K.H., J.P. Wiener-Kronish, and N.C. Staub 1984 The structure of parietal pleura and its relationship to pleural liquid dynamic in sheep. Anat. Rec., 208: Broaddus, C., and N.C. Staub 1987 Pleural liquid and protein turnover in health and disease. Semin. Respir. Med., 9:7 12. Fukuo, Y., T. Nakatani, H. Shinohara, and T. Matsuda 1988 The mouse pericardium: It allows passage of particulate matter from the pleural to the pericardial cavity. Anat. Rec., 222:1 5. Fukuo, Y., H. Shinohara, and T. Matsuda 1990 The distribution of lymphatic stomata in the diaphragm of the golden hamster. J. Anat., 169: Gaudio, E., N. Casale, L. Pannarale, A. Priori, and G. Marinozzi 1990 A scanning electron microscopy morphometric study of the rabbit peritoneal surface. Anat. Rec., 228: Hebel, R., and M.W. Stromberg 1986 Subclavian artery. In; Anatomy and Embryology of the Laboratory Rat. BioMed Verlag, Worthsee, pp Kampmeier, O.F Concerning certain mesothelial thickenings and vascular plexuses of the mediastinal pleura, associated with histiocyte and fat-cell production, in the human newborn. Anat. Rec., 39: Leak, L.V., and K. Rahil 1978 Permeability of the diaphragmatic mesothelium: The ultrastructural basis of Stomata. Am. J. Anat., 151: Light, R.W Anatomy of the pleura and physiology of pleura. In: Pleural Diseases, 2nd ed. Lee & Febiger, Philadelphia, pp Mariassy, A.T., and E.B. Wheeldon 1983 The pleura: A combined light microscopic, scanning and transmission electron microscopic study in the sheep. Exp. Lung Res., 4: McLaughlin, R.F., W.S. Tyler, and R.O. Canada 1961 A study of subgross pulmonary anatomy in various mammals. Am. J. Anat., 108: Miserocchi, G., and E. Agostoni 1971 Contents of the pleural space. J. Applied Physiol., 30: Miserocchi, G., T. Nakmura, E. Mariani, and D. Negrini 1981 Pleural liquid pressure over the interlobar mediastinal and diaphragmatic surfaces of the lung. Respir. Physiol., 46: Miserocchi, G., E. Mariani, and D. Negrini 1982 Role of the diaphragm in setting liquid pressure in serous cavities. Respir. Physiol., 50: Mixter, R.L On macrophagal foci Milky spots in the pleura of different mammals including man. Am. J. Anat., 69: Nakatani, T., H. Shinohara, and T. Matsuda 1986 On the ovarian bursa of the golden hamster. II. Intercellular connections in the bursal epithelium and passage of ferritin from the cavity into lymphatics. J. Anat., 148:1 12. Nakatani, T., H. Shinohara, Y. Fukuo, S. Morisawa, and T. Matsuda 1988 Pericardium of rodents: Pores connect the pericardial and pleural cavities. Anat. Rec., 220:

8 LYMPHATIC STOMATA IN THE PARIETAL PLEURA 23 Negrini, G.M., E. Mariani, and M. Passafaro 1983 Reabsorption of a saline- or plasma-induced hydrothorax. J. Applied Physiol., 54: Negrini, D., S. Mukenge, M.D. Fabbro, C. Gonano, and G. Miserocchi 1991 Distribution of diaphragmatic lymphatic stomata. J. Applied Physiol., 70: Negrini, D., M.D. Fabbro, C. Gonano, S. Mukange, and G. Miserocchi 1992 Distribution of diaphragmatic lymphatic lacunae. J. Applied Physiol., 72: Negrini, D., M.D. Fabbro, and D. Venturoli 1993 Fluid exchanges across the parietal peritoneal and pleural mesothelia. J. Applied Physiol., 74: Payne, D.K., and G.T. Kinasewitz 1985 Comparative permeability of canine visceral and parietal pleural mesothelium. Am. Rev. Respir. Dis., 131:A418. Payne, D.K., G.T. Kinasewitz, and E. Gonzalez 1988 Comparative permeability of canine visceral and parietal pleura. J. Applied Physiol., 65: Pinchon, M.C., J.F. Bernaudin, and J. Bignon 1980 Pleural permeability in the rat. I. Ultrastructural basis. Biologie Cellulair, 37: Shinohara, H., T. Nakatani, and T. Matsuda 1985 The presence of lymphatic stomata in the ovarian bursa of the golden hamster. Anat. Rec., 213: Starling, E.H., and A.H. Tubby 1894 On absorption from and secretion into the serous cavities. J. Physiol. (Lond.), 16: Stewart, P.B., and A.S.V. Burgen 1958 The turnover of fluid in the dog s pleural cavity. J. Lab. Clin. Med., 52: Tsilibary, E.C., and S.L. Wissig 1983 Lymphatic absorption from the peritoneal cavity: Resorption of patency of mesothelial stomata. Microvasc. Res., 25: Wang, N.S The preformed stomas connecting the pleural cavity and the lymphatics in the parietal pleura. Am. Rev. Respir. Dis., 111: Wiener-Kronish, J.P., K.H. Albertine, V. Kicko, and N.C. Staub 1984 Protein egress and entry rates in pleural fluid and plasma in sheep. J. Applied Physiol., 56: