145 



the smaller larvae have fewer cells than the normal larvae, and even 

 in much later stages this difference is found. In one case recorded 

 the fertilized normal eggs had about eighty cells at ten and a half 

 hours after fertilization, while a fragment (about half size) contained 

 no more than forty cells (approximately). The nuclei of the two 

 larvae are about the same size, but if we compare the smaller 

 forty-celled larvae with a normal forty- cell stage, we find the nuclei 

 in the smaller larvae much smaller than those of the full sized form 

 of corresponding stage. The smaller number of cells in the smaller 

 larva seems to have nothing to do with the fact that it starts to 

 segment an hour or so later, for after forty-eight hours we should 

 not expect to find this initial loss appreciable. A simple mechanical 

 explanation is probably at the root of the matter, but I do not feel 

 warranted in suggesting one. Other observations on these egg- 

 fragments give the following: The larger the nucleated fragment the 

 sooner does it begin to divide. Those that have lost very little of 

 their protoplasm divide soon after the normal eggs and then the others 

 follow according to their size, the smallest pieces do not divide until 

 two or three hours after fertilization. Whether this is due to the 

 delay in entrance of the spermatozoon or to alterations taking place 

 in the protoplasm of the fragment is not clear. The latter view seems 

 to fit the facts better. The size of some of these fragments is re- 

 markably small as illustrated in Fig. 2 a, b, c, d, drawn to the same 

 scale as the living normal egg Fig. 1. The smallest of these, Fig. 2 a, 

 contains about 1 / 70 of the volume of the normal egg. 



Fig. 2. 



I have examined with some care these egg-fragments to see if 

 they produced micromeres as do the normal eggs. Generally the 

 cleavage is too irregular to make this out ; in one lot of eggs, how- 

 ver, the cleavage was very regular in the fragments. In this lot the 

 micromeres were seen to be formed and a few drawings of such eggs 

 are shown in Fig. 3 a, b. Each of these egg-fragments had ten cells, 

 eight macromeres and two micromeres. The micromeres were clearer 

 and smaller than the other cells and distinctly visible. The method 



