158 University of California Publications in Zoology [Vol. 11 



became extremely flattened against the cover glass so there can 

 be little doubt that the extent of their free surface was greatly 

 increased ; and the form of the free cells as well as those on 

 the margins of the membranous outgrowths gave every appear- 

 ance of amoeboid activity. 



Since the tissues used in the cultures contained different 

 kinds of cells it is not certain that the cells that formed the 

 membranes were of epithelial origin. Wandering cells probably 

 derived from connective tissue are exceedingly common in cul- 

 tures of tissues in vitro, and while in several eases the membranes 

 that were formed strikingly resembled the common forms of 

 pavement epithelium, it is possible that they may have been of 

 mesenchymatic origin. 



In Harrison's paper (1910) on the development of nerve 

 fibers from the tieuroblasts of the frog embryo there is described 

 the outgrowth of sheets of ectoderm in hanging drops of lymph. 

 "Along the free border of these sheets of cells there often appears 

 a fringe of hyaline protoplasm, which undergoes continuous 

 amoeboid changes (figs. 13 and 14 pi. fr. ). In one case of this 

 kind it was observed that the sheet of cells gradually spread out 

 toward the side on which the fringe was placed. Since the work 

 of Peters ( '85- '89) it has generally been admitted that wound 

 healing in the epidermis is primarily due to the movement, in 

 part amoeboid, of the epithelial cells, so that it seems quite possi- 

 ble that in this fringe of hyaline protoplasm above described, we 

 have one part of the mechanism by which the movement of cells 

 in wound healing is brought about". Observations to be described 

 later bear out this conclusion. 



My own experiments on the behavior of epithelial cells were 

 made upon the embryos and young tadpoles of Rana, Hyla and 

 Diemyctyhts. It is more advantageous to work with the young 

 tadpoles before thej^ have made their escape from the jelly, as 

 it is easier to obtain preparations free from infection. Treatment 

 of the free tadpoles with antiseptics is liable to injure the epithe- 

 lial cells and to diminish the vitality of the tissues if carried far 

 enough to be effective. However, many successful preparations 

 were made from older tadpoles. Bacteria develop more rapidly 

 in fluid media such as serum. Ringer's, or Locke's solution than 



