114 CARNEGIE INSTITUTION OF WASHINGTON. 



peculiarity not found in all organisms and probably results from 

 peculiarities in chromosome behavior. In virilis Dr. Metz has found 

 this sexual difference to exist exactly as in ampelophila. At least 

 this is true in respect to all of the characters thus far studied and they 

 are sufficient to give strong evidence of the universality of the rule in 

 this species. 



With respect to the general chromosome theory of heredity. — Drosophila 

 virilis provides an exceptionally good opportunity to test this theory by 

 means of a comparison between its linkage phenomena and those of 

 ampelophila. In ampelophila there are 4 pairs of chromosomes and 4 

 groups of linked factors to correspond; in virilis there are 6 pairs of 

 chromosomes and hence, according to the hypothesis, there should 

 correspondingly be 6 linkage groups. It is very significant. Dr. Metz 

 believes, that 5 such linkage groups have already been located in this 

 species. Since this is one more than the number in ampelophila, it 

 argues strongly in favor of the chromosome hypothesis and makes it 

 probable that the discovery of the sixth group is only a matter of time. 



With respect to the relations between corresponding linkage groups in 

 related species. — Data on this subject are furnished by the sex-linked 

 factors in virilis taken in connection with those in ampelophila. Eight 

 factors belonging to the sex-linked group in virilis have been carefully 

 studied by means of crosses involving various combinations of charac- 

 ters. This series of experiments alone has involved upwards of 10,000 

 individuals. As a result the sex-linked factors are found to be related, 

 upon the basis of Hnkage, as shown in the accompanying diagram. It 

 is significant that they fall into a linear series just as do those of 

 ampelophila, and that when the position of a factor is known with 

 respect to any two previously located factors, one can predict its posi- 

 tion with respect to any other factor in the series. 



y fr V h m f r g 



6 S Tr 45 60 66 82 



Fig. 2. — Chart of the sex-chi-omosome in Drosophila virilis. 

 y, yellow; fr, frayed; v, vesiculated; h, hau-y; m, magenta; /, forked; r, rugose; g, glazed. 



Comparing this linkage group as a whole with that of ampelophila, 

 its relatively greater "length" is to be noticed at once. Similarly 

 constructed diagrams in the two species are respectively 67 to 68 

 ''units" {ampelophila) and 80 to 90 "units" (virilis) — a unit repre- 

 senting 1 per cent of crossing-over. Since the "length" increases 

 somewhat with an increase in the number of factors involved in a given 

 region (^. e., the length is thus more accurately determined, because 

 double crossing-over is largely eliminated), the discrepancy is really 

 greater than that indicated in these figures and the diagram in virilis 

 will probably include 90 to 100 units when more intermediate factors 

 are studied. This leaves no doubt that there is a difference in " length " 

 of the two linkage groups here; and since "length" represents amount 



