312 APPENDIX B 



into two binucleate cells, but it may divide at once into four, or 

 into three, one of which is binucleate. The interest lies in the 

 binucleate cells, for they continue to produce uni-nucleate and bi- 

 nucleate cells until the latter divide simultaneously into four, 

 and this simultaneous division may sometimes involve an irregular 

 distribution of the chromosomes, with fatal consequences to the 

 cell. Boveri had already produced evidence of the evil effects of 

 an irregular distribution of the 3 n x 2 chromosomes present in 

 triasters and tetrasters. A more detailed account is now given. 



Of the tripartite (triaster) ova about 8 % on an average pro- 

 duced Plutei. In these larvae three regions may be distinguished 

 in the egg by the size of the nuclei (proportional to the number 

 of chromosomes) and the boundaries between them may be shown 

 to correspond to the divisions between the three blastomeres. 

 The form is asymmetrical in skeleton and pigment, but Boveri 

 shows that both sides are normal, as though the larva had been 

 compounded of two types such as occur, as individual variations, 

 in any culture. It is suggested therefore that the slight differ- 

 ences in the two sides are due to differences in the two sperms. 



Some of the larvae have partial defects in skeleton or pigment, 

 or the skeleton may be much reduced on one side, or one-third of 

 the cells may be pathological, i.e. disintegrate in the segmentation 

 cavity, while the remaining two-thirds are sound and sometimes 

 symmetrical. In this case it is supposed that the degenerate 

 cells had separated from the others at an early stage, and that 

 the remainder had had time to recuperate. In others two-thirds 

 are degenerate, one-third normal, or all three degenerate. When 

 the three blastomeres are isolated and allowed to develop inde- 

 pendently, segmentation is partial, with two micromeres, two 

 macromeres, and four mesomeres, and often all three develop 

 normally up to the blastula stage. After that only one or two, 

 rarely all three, become Plutei, the rest giving rise to stereo- 

 blastulae or stereogastrulae, full of degenerating cells. 



The isolated quarters of tetrasters also segment partially 

 and normally, but few give rise to Plutei. The whole simultane- 

 ously quadripartite eggs only rarely give rise to what may be 

 called a Pluteus (2 cases in 1500) ; but very degenerate larvae 

 are found, with masses of disintegrating cells inside, which are 

 assigned to one of the four blastomeres. Stereogastrulae with 

 nuclei of all the same size are frequent. 



As has been already mentioned, Boveri points out that the 

 probability of each cell of a triaster receiving a complete set of 

 the n chromosomes of the species when there are 3 n x 2 to be dis- 

 tributed must be greater than that of each cell of tetraster 

 obtaining a full complement, and the probability for one isolated 



