eand: nervous system of lumbkicid^. 119 



such a way as to leave no doubt as to their fate. After the first week or 

 two the spindle cells form a smaller proportion of the cicatricial mass at 

 the regenerated end, this fact generally being due to the accumulation 

 of these cells into the new muscle layers. The remaining scattered 

 elements are either lymph cells or cells with large nuclei of the epider- 

 mal type. The latter cells tend to collect in the region into which the 

 regenerating cord is extending. 



The forward growth of the fibre bundle precedes the accumulation of 

 cells about it ventrally and laterally in the position of the ganglionic 

 masses. 



Of the three kinds of nuclei to be met with among the incompletely 

 differentiated cells of the regenerating region, there is no evidence that 

 either the lymph nuclei or the nuclei of the spindle cells give rise to 

 new nervous elements. The nuclei of these two types are totally unlike 

 the nuclei which are first found associated with the new nervous parts, 

 and there is no evidence of a transformation of one kind into the other. 

 Granting this, the origin of the new nervous elements must be referred 

 to the larger nuclei, those of the epidermal type. That these large 

 nuclei ai"e derived from the epidermis, there is good evidence in the 

 preparations. 



The formation of a new epidermis over the cicatrix offers some inter- 

 esting problems, which, however, require a study of earlier stages of 

 regeneration than any I have worked with. After seven days there is 

 always found over the cut end of the worm a continuous thin layer of 

 more or less flattened epidermal cells, and a thin layer of cuticula also 

 is already formed. In rare cases I have found mitosis in the epidermis 

 at this early stage. At later stages there is abundant mitosis in the 

 new epidermis, and in one case numerous dividing cells were found 

 among the basal or subepidermal cells back through several uninjured 

 segments. Some signs of amitosu were found in the regenerated epi- 

 dermis, but not conclusive evidence. Nuclei were found with two 

 nucleoli, and several columnar epidermal cells were found containing 

 two nuclei pressed so closely together that their contiguous surfaces 

 were quite flat, suggesting that there had been a direct division and 

 that the two nuclei had not yet moved apart. In one case four nuclei 

 in a common cytoplasmic mass were found so closely pressed together 

 that the group presented the appearance of the four-cell cleavage stage 

 of an Gg^. No nuclei in process of constriction were observed. 



The new epidermis having once been established, there is little room 

 for doubt that its later increase is effected by the mitotic division of its 



