kand: nervous system of lumbricid^. 145 



Two of them lying very close together occupy a slightly clearer space 

 just at the outer end of the nucleus. The condition in the right cell, 

 taken by itself, makes doubtful the interpretation of any of its granules 

 as a centrosome. The large granule at the deeper end of the nucleus, 

 or the pair in the small clear space at its outer end, or both, might 

 pass for centrosomes. Or they might almost equally well be consid- 

 ered accidental granules of the kind which forms the group at the outer 

 end of the nucleus. It must be noted, however, that the granule at 

 the inner end is in some respects unique. It is larger than any of 

 the others, its clear space is better defined, and the end of the nucleus 

 is sharply invaginated at the region nearest the granule, the curvature 

 of the invagination corresponding to the outline of the clear space. 

 These facts, together with the fact that the next cell contains a similar 

 granule, similarly located, and without the presence in the cell of any 

 other like bodies, make it highly pr9bable that the granule at the inner 

 end of the nucleus of each cell is not an accidental thing. 



These two cells well represent the conditions in the old epidermis. 

 Very many, perhaps most, of the cells resemble the cell at the right in 

 Figure 43. There are several, or very man}', large granules in the cyto- 

 plasm, often at both ends of the nucleus. Frequently one of these gran- 

 ules lying near the inner end of the nucleus will appear slightly larger 

 than the others, or it may lie in a clear space; yet it is not sufficiently 

 peculiar to justify considering it anything but an accidental granule 

 like others in the cytoplasm. The combined evidence from all such 

 cells affords alone no conclusive argument for the presence of a centro- 

 some. But very often there are found cells like the left one of the two 

 in Figure 43, where there is a single conspicuous granule, which is gen- 

 erally situated not far from the inner end of the nucleus. In rare cases 

 some weak radiations can be detected. It is not uncommon to find the 

 nucleus invaginated at its inner end, as in the right cell. These gran- 

 ules are doubtless " centrosomes " in the sense that such bodies in 

 resting cells are so called, without knowledge as to their history or 

 function. The occurrence of cells with the single definitely located 

 granule makes it likely that in cells whose cytoplasm is filled with 

 large granules the centrosome is present, but its identity is rendered 

 doubtful by the presence of other bodies of similar appearance. 



Figures 44 and 45 represent cells from recently regenerated epidermis. 

 A thin layer of cuticula is present. The nuclei are more nearly spher- 

 ical than in the old epidermis, and the cells are more flattened. Often 

 no distinct cell limits can be seen. These figures are from sublimate 



