120 BULLETIN: MUSEUM OF COMPAEATIVE ZOOLOGY. 



own cells. It is after the first week that we find evidence that not all the 

 cells produced by the epidermis ai-e destined to remain epidermal cells. 



The normal epidermis consists of a single layer of columnar cells with 

 some basal cells of irregular form wedged in among their deep ends. 

 The nuclei of these basal cells are unlike those of the columnar cells. 

 They are smaller, often lack a prominent nucleolus, and genei-ally pre- 

 sent a darker appearance, due to a greater proportion of chromatic 

 material. In the regenerated epidermis places are often found where cer- 

 tain cells, with nuclei of the kind characteristic of the columnar cells, tend 

 to form a second layer. Single cells are found so placed as to suggest 

 that they are being pushed down below the surfiice layer. Others, 

 apparently, have been quite displaced from their superficial position 

 and lie free underneath the epidermis. There may frequently be found 

 small masses of cells with nuclei precisely like those of the epidermis, 

 lying close to the deep surface of the epidermis, and so far removed from 

 any other tissue containing similar nuclei that their epidermal origin is 

 beyond doubt. 



The surface of the regenerated end is often very uneven in the early 

 stages. The new epidermis is thrown into folds or marked by eleva- 

 tions and pits. At an inner angle of a sharply iuvaginated region of 

 epidermis may sometimes be seen evidences of an inward proliferation 

 of the epidermal cells. There is also some evidence, as Hescheler finds, 

 that not only single cells, but also considerable masses of cells, may be 

 pushed in and separated from the surface layer. Small local invagina- 

 tions are sometimes found which are nearly closed over, and in one or 

 two places I have seen small cavities completely enclosed by epidermal 

 cells, as if small invaginations had become closed over and had sunk 

 beneath the surface. 



In one animal of twenty-four days' regeneration the new nerve cord 

 was already well established. The region of the epidermis nearest the 

 anterior end of the regenerating cord consisted of loosely aggregated 

 cells. Between this region and the end of the cord were numerous 

 scattered cells with nuclei like those of the epidermis and also like those 

 of the cells already definitely associated with the new cord. The scant 

 cytoplasm of these scattered cells was more or less drawn out to spindle 

 shape, and most of the cells were placed with their long axes extending 

 in the direction of the nerve cord. Similar conditions were found in 

 other animals. The appearance indicates a separation of cells from the 

 epidermis and the migration of these cells in toward tlie regenerating 

 cord. 



