16 ROBERT S. ELLIS 



To counteract as much as possible the effects of fixation, I 

 have compared the negro brains used by Mall with his white 

 brains — in this way I believe I have as nearly as possible elimi- 

 nated the effect of changes in the fixing fluid, because the brains 

 of both races were fixed in the same manner and at the same time. 



To the brains used by Mall I have added those used by Bean 

 ('06). The two groups do not differ appreciably, so I think no 

 error has been introduced by bringing them together. The 

 larger number is extremely desirable statistically. 



The distribution of the percentage w^eights of these cases is 

 shown in chart 3. The arithmetical average of the percentage 

 weights for the negroes is above that for the whites, and the 

 average for the negro females is above that for the negro males; 

 however, the difference is not a great one and it is not very im- 

 probable that a larger number of cases might show the same dis- 

 tribution for the two races. But as far as the results go, they 

 do show a relatively larger cerebellum in negroes. If this indi- 

 cates anything at all, its significance is probably to be stated in 

 terms of the cerebrum rather than in terms of the cerebellum. 



As far as individual cases are concerned, the chart shows clearly 

 that the relative w^eight of the cerebellum means nothing save in 

 the rarest instances. 



THE DISAPPEARANCE OF THE LAYER OF EXTERNAL GRANULE 



CELLS 



Since its discovery by Hess in 1858, the layer of external gran- 

 ule cells has been studied by many investigators. Vignal ('89), 

 in the course of his study of the development of the nervous 

 system, made some observations on these cells and gave us some 

 very good figures showing the histology of the cerebellar cortex 

 at different ages during the fetal period. However, he did not 

 understand the external granule cells; in fact, he thought them 

 pathological, and advanced the theory that they were leucocytes 

 brought to the molecular layer by some inflammation. Since 

 that time there have been several other theories almost as inter- 

 esting. 



