Cells and Structures of the Renal Corpuscle 



179 



the earliest stages of development, but in slightly 

 later stages an occasional erythrocyte was present 

 in the central mass of prospective endothelial cells. 

 Lumina appeared to develop later in the solid pri- 

 mordial capillaries. Lumina appeared to begin devel- 

 opment with the appearance of large cytoplasmic 

 vesicles within individual endothelial cells. Later as 

 the vesicles appeared enlarged, the endothelial cyto- 

 plasm, except around nuclei, appeared attenuated. 

 The impression was gained that the continuous 

 capillary lumina of mature capillaries is formed by 

 the subsequent fusion of enlarged, individual cyto- 

 plasmic vesicles. In figs. I and 2, it is to be noted 

 that some vesicles appear filled by a fully difTeren- 

 tiated erythrocyte. This may be interpreted to mean 

 either that erythrocytes differentiate /// situ, or that 

 the vesicles are apparent only and exist merely as 

 tangential sections of lateral outpouchings of con- 

 tinuous, patent capillary lumen filled with plasma and 

 erythrocytes. The validity of either possiblity can be 

 determined with certainty only by means of a large 

 series of serial sections not available now. Yet, the 

 presently available sections, including some serial 

 ones, through about ten immature glomeruli have 

 produced no support for the second interpretation. 

 From a statistical point of view, it is very unlikely 

 that all sections through many immature capillaries 

 and several arterioles were cut tangentially so that all 

 failed fortuitously to show evidence of a continuous, 

 patent capillary lumen in any of the earliest stages 

 of glomerular formation. 



The impression is readily gained from close scru- 

 tiny of the electron micrographs that the lamina 

 densa (definitive glomerular capillary basement mem- 

 brane) forms only in close, intimate relationship 

 with the smooth, continuous, endothelial cell surface 

 membrane. It has not been found extending into or 

 following the numerous folds of the podocytic cell 

 surface membrane. Yet, it appeared rather frequently 

 to extend or form to some degree between two endo- 

 thelial cell surfaces within the walls of the primordial 

 capillaries. These observations, along with the fre- 

 quently observed close relationship of the lamina 

 densa to endothelial cell nuclei, afi"ord support to 

 the view that lamina densa formation is dependent 

 upon the presence or activity of external surface 

 membranes of the endothelial cells. The proximal 

 surface membranes of the podocytic cells may form, 

 or play a role in the formation of the lamina densa, 

 but they often appeared much less intimately associ- 

 ated with the developing lamina densa, than did the 

 endothelial surface membranes. 



It is of interest to note that soon after its earliest 

 appearance, the lamina densa had attained a width 



characteristic of the adult glomerulus. The basement 

 membrane of Bowman's capsule in these same glo- 

 meruli dig. 2) often appeared thinner than the lamina 

 densa. In adult kidneys the basement membrane of 

 Bowman's capsule usually appears much thicker 

 than the lamina densa, often several times as thick. 

 The continued increase in thickness of the capsular 

 basement membrane, and its lamination and fibrous 

 nature in fully matured Malpighian bodies appear 

 to completely distinguish capsular basement mem- 

 brane from lamina densa. 



The electron micrographs do not support the old 

 belief that capillaries invaginate the blind end of a 

 nephron, but tliey do show clearly that glomerular 

 capillaries ditVerentiate in situ from an unorganized 

 mass of prospective endothelial ceils, much as Her- 

 ring (4) and ReinhofT(9) had found. Previously, Mall 

 (2, 3) has shown clearly that the kmiina densa 

 (definitive glomerular basement membrane) is the 

 only basement membrane in the capillary wall, and 

 in this study it is demonstrated that lamina densa 

 forms only in direct and intimate contact with the 

 surface membranes of prospective endothelial cells. 

 It clearly follows that there can be no embryological 

 basis for the old belief that a two-layered fold of 

 Bowman's capsule extends from capillary to capil- 

 lary, enclosing a mesangium and/or a limited inter- 

 capillary space between limbs of imaginary capillary 

 loops. If the preliminary data and interpretations 

 reported here are fully substantiated by future stu- 

 dies, then Gerlach's old idea of a glomerular perito- 

 neum, and its modern counterparts, Zimmermann's 

 "mesangium" and "intercapillary space" in the spe- 

 cial sense of McManus and others, must be replaced 

 by concepts of glomerular structure which are in 

 accordance with the known embryological and 

 anatomical facts. 



References 



1. GtRLACH, J., Anil. Anal. Physiol, ii. wiss. Med., p. 378 



(1845). 



2. Hall, B. V., Proc. Vth Ann. Conf. on the Nephrotic 



Syndrome, p. I. Natl. Nephrosis Foundation, New 

 York, 1953. 



3. — Proc. Vlth Ann. Conf. on the Nephrotic Syndrome, 



p. 1. Natl. Nephrosis Foundation, New York, 1955. 



4. HiRRiNG, P. T., /. Pathol. Bactcriol. 6. 459 (1900). 



5. HuBiR, G. C, Am. J. Anal. 4, Suppl., I (1905). 



6. — The Harvey Lectures, p. 100 (1909-10). 



7. McManus, J. F. A., Am. J. Pathol. 24. 1259 (1948). 



8. Paladh, G. E., /. E.xptl. Med. 95, 285 (1952). 



9. RiiNHOiF, W. F., Johns Hopkins Hosp. Bull. 33, 392 



(1922). 



10. ViMTRUP, Bj., Am. J. Anal. 41. 123 (1^>:S). 



11. Zimmirmann, K. W.,Z. mikroskop. anal, loisth. 32, 176 



(1933). 



