164 University of California Puhlications in Zoology [Vol. 11 



the latter and crept along the second fiber. At the same time 

 a second strand of cells branched off from the first, and pursued 

 its course independently of the two fibers and about midway 

 between them. 



After several days the strands in all the preparations showed 

 a tendency to draw in and form more compact masses. Often 

 they were drawn back to the original piece, forming a large 

 clump of cells which persisted with little change of form, though 

 remaining alive for weeks. Sometimes the larger strands would 

 break up into separate masses of cells which would aggregate 

 into more or less rounded masses. Gradiially the activity of the 

 epithelial cells, even when frequently changed to fresh media, 

 became more and more reduced, although in some cases they 

 presented a normal appearance for over two months. 



The striking behavior of these extensions of the ectoderm 

 naturally raises several questions as to the nature and causes 

 of movements involved. Are the strands and sheets of cells 

 pushed out or pulled out? Are they due to growth or cellular 

 activity? Do the cells increase in number after implantation in 

 plasma ? What part, if any, do tropic responses play in produc- 

 ing the observed results? In regard to the first question some 

 observations made on the preparation shown in figure 4 are 

 significant. At the time the culture had assumed the form shown 

 in the figure especial attention was bestowed upon the tip of 

 the narrow strand that lay between the two fibers of cotton. 

 The changes witnessed during the few hours that the part was 

 kept under observation are illustrated in figures 8-11. The tip 

 coiild be seen to be very slowly advancing, and many of the cells 

 showed a broad, thin hyaline border which was furnished with 

 fine pseudopods and irregular lobes of protoplasm. These were 

 especially well developed at the extreme tip of the process and 

 on a small lobe a short distance behind the tip ; elsewhere this 

 whole branch was devoid of pseudopods and the cells presented 

 no clear expansions. The hyaline extensions of protoplasm were 

 the seat of amoeboid activity; the changes in the small lobe 

 mentioned are shown in figure 8, a, b, c. A mark of an eye- 

 piece micrometer (shown by a dotted line) turned so a,s to lie 

 across the tip of the extension made it possible to follow a very 



I 



