EFFECTS ON THE SPINAL CORD 195 



poorly-staining cytoplasm. Others show cloudy or granular cytoplasm, the 

 nuclei staining well with carmine. The coagulation necrosis and plasmatic 

 exudate described by Rosenbach were not found. 



Downarowitsch ('92) was apparently the first to make systematic nuclear 

 measurements of the nerve cells, in 3 rabbits killed after 8 days of total inanition 

 with loss of about 40 per cent in body weight. In these and 3 normal controls, 

 the cervical and lumbar enlargements of the cord were fixed in mercuric chloride 

 and the long and short axes of the nuclei of 600 anterior horn cells measured in 

 stained sections. From the average measurements, the corresponding volume 

 of the nucleus (assumed to be an ellipsoid) was calculated to be 11 15.97 cubic 

 micra in the normal, and 832.47 cubic micra after inanition, indicating a decrease 

 in volume of 25.4 per cent. The nucleolus was also measured, and its average 

 volume (considered spherical) apparently loses 42.5 per cent during inanition. 



Peri ('92, '93), carefully studied various regions of the central and peripheral 

 nervous system by various methods (Weigert, Marchi, Golgi, Cajal, etc.). 

 Grossly only venous stasis and mild edema were found, never the "steatosis" 

 described by Parrot ('68), Filipi ('89) and others. Histologically, but very 

 slight changes appear in the rabbits, probably on account of the short duration 

 of inanition (3-5 days, with water); in the cats (fasting 15 days, with loss of 43- 

 45 per cent in body weight) the changes were somewhat greater; and in the dogs 

 (fasting 27-34 days, with loss of 43-44 per cent) the changes were greatest. 

 But even in the dogs most of the nerve cells in the brain and spinal cord appear 

 normal, relatively few showing degenerative changes. In the anterior horn 

 cells, hyalin degeneration occurs rarely, with nuclear disappearance; but the 

 Marchi and silver methods show nothing abnormal. The neuroglia appears nor- 

 mal, but diapedesis is somewhat frequent. Peri ascribed the degenerative 

 appearances described by Mankowski, Rosenbach, Popow and Coen to the 

 inadequacy of the older technique employed. 



Barrows ('98), using Hodge's method measured the nerve cells of the spinal 

 cord, spinal ganglia and occipital cortex in 3 starved rats with normal controls, 

 and found: 



"(1) A decided shrinkage in size of the cells and nuclei in the famished 

 animals, averaging about 20 per cent, and a still greater shrinkage in the 

 nucleoli. 



" (2) An evident exhaustion of the substance of the famished cells, as shown 

 by their faint staining with osmic acid and the notable absence of nuclei and 

 nucleoli. The protoplasm of these cells shows a very fine vacuolation, not so 

 marked as that described by Rosenbach for starving animals and by Hodge for 

 extreme fatigue. In the brains of famished rats the pericellular lymph spaces 

 are considerably enlarged." 



About this time, the Nissl method was introduced and was applied by several 

 investigators to the nerve cells of animals subjected to inanition. Tauszk 

 ('94), a pioneer in this field, found by this method that chromatolytic changes 

 appear more distinct in chronic than in acute inanition, and were most apparent 

 in the cervical region of the spinal cord in rabbits. Ganfini ('97), in rabbits 

 killed after 5-7 days of inanition, found the most marked changes in the anterior 



