MULTIPLE CHROMOSOMES 587 



The only direct evidence for any such a differential nature of 

 the other chromosomes is afforded by the work on Drosophila. 

 In this case a whole group of characters follows the rule of distri- 

 bution governing sex. This is the only element that can cer- 

 tainly be identified, but the groups of observed characters, not 

 sex linked, correspond to the three remaining pairs of chromo- 

 somes, and the numbers of characters in each group are in pro- 

 portion to the size of the chromosomes. Of these one pair is 

 very small and there is a corresponding small group of charac- 

 ters. As is well known, the analysis of the conditions in Dro- 

 sophila has gone so far that the relative loci for the different 

 factors have been calculated. It would seem from these results 

 that there is every reason to regard the chromosomes of this fly 

 as qualitatively different. While there is no correlation of this 

 sort known for the Orthoptera, the actual history of the chromo- 

 somes which their germ cells exhibit affords a mechanism ade- 

 quate for the facts of alternative inheritance and for the segre- 

 gation and chance recombinations of characters. While the 

 multiple chromosomes do not afford any direct evidence for 

 specificity of function, it would seem, because of all the facts, 

 that the purpose for which these studies on the Orthoptera were 

 begun — a correlation between germ cell structure and somatic 

 characters — is much more feasible of accomplishment than 

 appeared at the beginning. 



V. SUMMARY OF RESULTS 



1. Chromosomes are definitely organized chromatic bodies 

 acting as units in mitosis. 



2. These units are of unlike morphological value in the dif- 

 ferent generations of germ cells. 



3. While any one cell generation is marked by one general 

 type of chromosome organization, individual chromosomes may 

 differ from the type in higher or lower degree, by definite steps 

 or intervals. 



4. In a given species the integration of the chromosome com- 

 plex may vary from individual to individual. 



5. For any one individual this integration is fixed. 



