INDIVIDUALITY OF CHROMOSOMES. 19 



mosomes of the series represent distinctive potentialities, make it 

 probable that a given size-relation is characteristic of the physical 

 basis of a definite set of characters." In the above sentence I 

 should like to substitute form for "size-relation" and Gryllits 

 for Bracliystola. A difference in the shape of the chromosomes 

 indicates the fact that they form the physical basis of different 

 sets of characters. A difference of shape supports Boveri's sup- 

 position of a difference of "role of chromosome" even more 

 strongly than a difference of size. But Boveri's method of study, 

 multipolar spindles, can not be applied to Orthoptera. To me it 

 seems probable that the test, if we can get any, must come from 

 hybrid germ cells. Sutton (31) has shown that " the phenomena 

 of germ-cell division and of heredity" as expressed by Mendel's 

 laws " are seen to have the same essential features." Moenk- 

 house's(2i) and Metcalf's (20) results teach us that forms with 

 differently shaped chromosomes can be crossed. If now we can 

 raise such hybrids to sexual maturity, we can probably get light 

 on the " purity of the germ-cells " as well as on the meaning of 

 the various shapes of the chromosomes. 



Montgomery (26) criticizes Sutton (3 i) and says that the combi- 

 nations of the paternal and maternal chromosomes in the fertilized 

 egg do not follow the Mendelian ratio. In sustaining this posi- 

 tion Montgomery says the Mendelian ratio can hold only in cases 

 where there are but two chromosomes in the fertilized egg. For 

 the case of four chromosomes the relation would be 1:14:1 in- 

 stead of i :2: i, and for twenty-four chromosomes 1 : 16,777,214: 1. 

 Montgomery must have entirely misunderstood Sutton or the 

 Mendelian principle. Mendel found that a pair of alternative 

 characters followed in self-fertilized hybrids the ratio of 1:2:1. 

 Sutton found the same ratio for a pair of homologous chro- 

 mosomes. Sutton also found in a form that had twenty-four 

 chromosomes there could be 16,777,214 different combinations of 

 chromosomes in the gamete. To take a concrete example in hy- 

 brids having twenty-four chromosomes a pair of alternative char- 

 acters ought to appear in the second generation hybrid in the 

 proportion iD : 2Dr : ir, but only one out of 16,777,214 of such 

 hybrids should show all the characters belonging to one of the 

 pure ancestors only and none of those of the other. Probably Dr. 



