284 FRANZ SCHRADER. 



inevitable the hypothesis that only certain definite chromosomes 

 break up into smaller units. 



If effective polyspermy were common, the possibility that the 

 supernumary sperms might combine with each other or any of 

 the embryonic cells should receive some consideration. It 

 would then remain to be explained why such numbers as 20 or 40 

 should not be encountered, since the 5 chromosomes of each 

 sperm should make almost any combination possible. But aside 

 from this consideration, I may point out as I have previously 

 done, that supernumary sperms even if present rarely evolve 

 chromosomes and certainly not as part of a regulated process, 

 whereas giant cells with their multiple numbers of chromosomes 

 are to be found in every developing egg at the right stage. 



Finally, as the last of these possible but unlikely hypotheses 

 remains an irregular behavior of the polar bodies. It is perfectly 

 possible that the first and second polar bodies may not combine 

 and that one or both may continue to divide independently. 

 In combination with embryonic cells, the second polar body 

 might thus bring about such numbers as 25 and 35. But against 

 this I need repeat only my former observations that in all eggs at 

 the crucial stage under observation, the polar nucleus is always 

 formed and its 15 chromosomes go through several normal mito- 

 ses. None at all show an independent development of the two 

 polar bodies. This of course still leaves the possibility that the 

 first polar body may at times divide before fusion with the second. 

 It may even be admitted that independent divisions or develop- 

 ment of the polar bodies may sometimes occur. But such cases, 

 and I have not seen any, are not normal. 



These considerations leave only two types of cells as factors 

 in the origin of the giant cells. They are .the polar nucleus deriva- 

 tives carrying 15 chromosomes and the true cleavage cells with 

 10 chromosomes. 



Taking up the three types of giant cells in order, it is plain that 

 the 25 chromosome type can arise only by a combination of a 

 polar nucleus derivative (15 chromosomes) with one cleavage cell 

 (10 chromosomes). 



The 30 chromosome type must arise from a fusion, or recom- 

 bination after mitotic division, of 2 polar nucleus derivatives. 



