CHROMOSOME STUDIES ON THE DIPTERA. 26l 



influence seems to lie deeper in the chromosome organization. It 

 is not revealed, however, by any morphological feature such as a 

 difference in length or structure of the two components of a loop. 

 Genetically one might assume the presence of balanced lethals to 

 account for the results ; but there is no evidence of their being 

 present, and no chance of securing such evidence, because the flies 

 are unsuitable for genetical study. 



It is to be noted that the looped region of the chromosomes in 

 Dasyllis is usually median. In many organisms the middle of the 

 chromosome is apparently the last part to undergo synapsis, which 

 suggests that the process in Dasyllis is actually an incomplete but 

 otherwise normal synapsis. On the other hand, in the Diptera 

 thus far studied, the synaptic process appears to be very different 

 from that in other organisms, and there is no indication of synapsis 

 beginning at the ends of the chromosomes and progressing toward 

 the middle. Nevertheless, the fact that three pairs of chromo- 

 somes are affected similarly suggests an influence on the general 

 synaptic process in these organisms, rather than the independent 

 action of agents located in the respective pairs. 



The chromosome behavior in Dasyllis recalls the genetical be- 

 havior of certain strains of Drosophila melanogastcr in which 

 crossing-over is greatly diminished in certain parts of linkage 

 groups (Muller, '16; Sturtevant, '19; Detlefsen, '20). This de- 

 crease or elimination of crossing-over in part of a chromosome is 

 just what one would expect from an incomplete synapsis such as 

 described above. 



However, it will be necessary to get both the genetical and 

 cytological data from one organism before the evidence becomes 

 satisfactory, and I do not wish to push the present analog} 7 . In 

 Drosophila crossing-over occurs only in the female. If this is 

 true in Dasyllis, then oogenesis rather than spermatogenesis must 

 be used for comparison. It also appears from the Drosophila data 

 that many, if not most, of the cross-over modifications manifest 

 themselves only in flies heterozygous for the cross-over " genes." 

 In other words, as Sturtevant has suggested (p. 329), they appear 

 to be due to an unlikeness in homologous chromosomes. In 

 Dasyllis there is at present no evidence that the homologous 

 chromosomes differ in the regions where synapsis fails to occur. 



