MUSEUM OF COMPARATIVE ZOOLOGY. 167 



inclined to doubt the presence of any recognizable cell substance, and 

 consequently to compare them with the " jSTervenkerne " described by 

 Rohde ('87, p. 30). But at length fortunate staining and thin section- 

 ing showed unmistakably the presence, in many cases at least, of an ex- 

 tremely small cell body, such as is shown together with the nerve fibre 

 in Figure 31. It will be noticed that the nucleus is oval in this case, 

 and that the nerve fibre proceeds from one of the small ends of the oval. 

 This fact, as well as the variation in form noticed by careful focusing on 

 the nuclei, would seem to warrant the assumption that these nuclei are 

 uniformly oval. From the diminutive size and transparency of the cell 

 body in comparison with the highly refractive nucleus, it is at once evi- 

 dent that the former can be seen only under the most favorable circum- 

 stances. Since I was unable to find any difference in position, size, or 

 optical qualities between these and the other nuclei of similar size and 

 appearance, I feel justified in maintaining the existence of such a cell 

 body for all nuclei of the class. 



The second sort of ganglionic cell {cl. gn. II.) is distinguished by the 

 presence of a much larger cell body (Plate III. Fig. 29). The nuclei 

 correspond so exactly to those of the first class that they can hardly be 

 distinguished from them. I was unable to see that they were either 

 more or less deeply stained, or that they were, on the average, larger or 

 smaller than nuclei of the first sort. The great difference is in the cell 

 bodies, which in this case are several times larger than the nucleus, 

 measuring 20 by 14 /a, and always evident on account of their slight affin- 

 ity for stains. One or more vacuoles of non-colored matter, the para- 

 mitome of recent writers, may always be found, and in favorable cases 

 there can be seen such a distribution of these as is shown in Figure 32. 

 The paramitome exists in the form of numerous peripheral vacuoles sub- 

 jacent to the enveloping connective fibres, and possibly ( ?) surrounded 

 by them. The nucleus lies in a zone of clear matter, while the mitome, 

 or filar substance, appears densest external to this. Between these and 

 the first sort of ganglionic elements there exists every possible transi- 

 tion, so that this class is but poorly marked off from the preceding one. 

 The vast majority of these cells are, however, of approximately uniform 

 size, and I therefore cannot agree with Nansen ('87, note, pp. 113, 114) 

 when he maintains that such transitional forms forbid the grouping of 

 these cells in different classes. Such intermediate forms serve rather to 

 explain the development of the one type from the other, without detract- 

 ing from the individuality of either class. 



