CHANGES IN PANCREAS, PHOSPHORUS POISONING 245 



remarkable resemblance to those formed in the preceding stage 

 through the fusion of the agglutinated mitochondria. They 

 differ, however, from these in that they are more spherical and show 

 not the least evidence of a granular structure. Some are quite 

 small, but the largest may actually exceed the nucleus in size. 

 There is a variable number of them in each cell. Their properties, 

 so far as I have observed, are as follows: 1) They are drop-like in 

 form; 2) They stain faintly bluish with hematoxylin and eosin; 



3) They do not stain specifically with Sudan III or Scharlach R; 



4) They are insoluble in absolute alcohol and toluol after chro- 

 matization (3 per cent solution K2Cr207 eleven days, 8°C.); 5) 

 They stain with acid fuchsin, in fuchsin methyl green, and with 

 Orange G in Mallory's triple stain. I assume that they are 

 lipoid. 



Within these droplets are to be observed little spherical 

 vacuoles which do not stain at all. These clear spaces are some- 

 times quite numerous especially in the largest droplets. Some 

 of the droplets, indeed, appear as only a rim of stainable sub- 

 stance about such a vacuole. That this phenomenon is not due 

 to the imperfect penetration of the chromatizing solution is 

 proven by the fact that the vacuoles may be arranged along the 

 periphery as is seen in figure 5 in the largest droplet. It must 

 be due to a difference in the solubility, to a loss of the staining 

 capacity, or to an alteration in the composition of this part of 

 the lipoid droplet. The protoplasm surrounding these droplets 

 stains very intensely with the methyl green. Further away this 

 dark color shades gradually off into the light green which the 

 cytoplasm usually stains. In this more deeply staining zone 

 mitochondria are seldom present ; but in that part of the cyto- 

 plasm possessing the more normal staining reaction they are 

 to be observed. It is to be especially noted that they show no 

 tendency to agglutinate and possess no bleb-like swellings. 

 There are many nuclei which seem to be quite normal in appear- 

 ance but there are others which clearly bear evidence of excessive 

 cell damage. 



In the same specimen which shows these large lipoid droplets 

 cells are to be found riddled with clear vacuoles (fig. 6). Such 



