182 



J. D. LEVER 



cases, the cytoplasm of the cells of the first part of 

 the proximal convolution looks denser than that of 

 the terminal part. 



A free lumen throughout both the first and termi- 

 nal parts of the proximal convolution has been 

 proved to exist after intravenous injection of phlori- 

 zin as well as diodrast (Rhodin, unpublished obser- 

 vations). Either injection caused a heavy diuresis. 

 In these cases, the kidney was fixed according to 

 the standard procedure described previously (3). 

 The cells of the first part of the proximal convolution 

 fixed after induction of diuresis show isolated brush 

 border extensions, as well as "dome'' cells of normal 

 appearance without brush border extensions. At the 

 same time there is a clear-cut difference between 

 cells of the first and terminal part of the proximal 

 convolution as far as the density of the cytoplasm 

 is concerned. In these diuretic kidneys, no "hydra- 

 tion" (2) was found to close the lumen. The explana- 

 tion is no doubt that the injected diuretics cause 

 such heavy accumulation of water in the lumen that 

 it expands the whole tubule. The importance of the 

 technique used by Pease (2) is not to be questioned 

 here. On the other hand, many facts prove that the 

 common technique of immersing the tissue is also 

 reliable, at least when a thorough knowledge of the 



normal light microscopic picture is employed as the 

 basis for extended analyses with the electron micro- 

 scope. 



What has to be stressed is the fact that not only a 

 free lumen of the present investigated part is in most 

 cases demonstrable but also characteristic cellular 

 features like few, rounded mitochondria, poor den- 

 sity of the cytoplasm, and few or almost absent in- 

 foldings of the plasma membrane. 



This study has been supported through a grant from 

 Stiftelsen Gustaf och Tyra Svenssons minne. 



References 



1. Palade, G. E., /. Exptl. Med. 95, 285 (1952). 



2. Pease, D. C, Anat. Record 121, 723 (1955). 



3. Rhodin, J., Correlation of Ultrastructural Organization 



and Function in Normal and Experimentally Changed 

 Proximal Convoluted Tubule Cells of the Mouse 

 Kidney. Karolinska Institutet, Stockholm, 1954. 



4. Sjostrand, F. S., Acta Anat., Siippl. 1, (1944). 



5. — /. Cellular Coinp. Physiol. 42, 15 (1953). 



6. — in Physical Techniques in Biological Research (Oster 



and PoLLisTER, eds.). Vol. Ill, p. 241. Academic Press, 

 New York, 1956. 



7. Sjostrand, F. S. and Rhodin, J., E.xptl. Cell Research 4, 



426 (1953). 



The Fine Structure of Brown Adipose Tissue in the Rat: 

 with Observations on the Cytological Changes Following Starvation 



and Adrenalectomy 



J. D. Lever 



Department of Anatomy, Cambridge University, Cambridge 



1 HE main body of current opinion is that brown 

 fat is merely a form of adipose tissue and not an 

 endocrine gland as Cramer ( I ) believed. The works 

 of Wertheimer and Shapiro (14) and Fawcett (3, 4, 5) 

 suggest the active synthesis of glycogen and lipid 

 within adipose tissue. In brown fat cells fuchsino- 

 phile "secretion-like" granules have been described 

 by Rasmussen (12) while Sheldon (13) believed in a 

 direct transformation of mitochondria into lipid 

 droplets. 



This paper presents a short light and electron 

 microscopic study of brown fat in the normal, 

 starved and adrenalectomised rat. Tissue for electron 

 microscopy was fixed in Dalton's (2) dichromate- 

 osmic solution for 45 mins., methacrylate-emhedded 

 and sectioned at approximately 150 A. 



Normal brown fat — Light microscopy: The multi- 

 locular lipid distribution in brown fat is well seen 

 after Sudan black staining, the lipid droplets ranging 

 in diameter from 0.5-7 p within any cell. It is con- 



sidered likely that fuchsinophile bodies, demonstra- 

 ted by methyl green acid fuchsin staining, are 

 mitochondria in oval or rounded shapes (0.5-1 // d.) 

 grouped between and around the fat droplets. In 

 size and position these fuchsinophile bodies exactly 

 correspond to phospholipid-positive granules dem- 

 onstrated by the Baker acid haematein stain. 



Normal brown fat — Electron microscopy: Apart 

 from short invaginations into the cell, the plasma 

 membrane is uncomplicated and is covered through- 

 out its extent by a basement membrane. The endo- 

 plasmic reticulum (10, 11) in brown fat is incon- 

 spicuous and consists of a small saccular component, 

 and a granular component (150 A d.) which is 

 aligned along the outer walls of the saccular compo- 

 nent or occurs independently in the cytoplasm. The 

 nuclear membrane is bi laminar: there are no special 

 features of the nucleoplasm. Throughout this study 

 the Golgi apparatus was not convincingly demon- 

 strated. 



