DEPARTMENT OF EXPERIMENTAL EVOLUTION. 115 



of crossing-over, it appears that crossing-over is more frequent in 

 virilis than in ampelophila. Whether this is due to a difference in the 

 length (or amount) of the sex-chromosome involved in the two cases 

 (assuming that the chromosomes carry the factors) or to an actual 

 difference in rate of crossing-over per unit of chromosome length, is 

 not certain, but the available evidence suggests the latter. 



With respect to the relation between particular characters and their 

 factors in two related species. — Striking, although not extensive, evidence 

 has been obtained on this question. It was partly included in last 

 year's report, but has been more fully worked out this year. Two of 

 the mutant characters in virilis — "yellow" body-color, and "forked" 

 bristles — ^appear to be almost exact duplicates of two in ampelophila. 

 A careful comparison under the microscope reveals a remarkable Uke- 

 ness in each case, and there seems to be great probability that the 

 characters are actually homologous in the two species. Considering 

 this fact, it is very significant to note that in both cases the characters 

 are sex-linked and that their factors occupy similar relative positions 

 in the factorial (linkage) group. Such a series of correlations strongly 

 suggests a conmion germinal basis in the two species — in other words, 

 lends material weight to the assumption of a definite organization in 

 the germ-plasm and to the assumption that this organization persists 

 from species to species. 



In addition to the above more or less anticipated results, another 

 series of phenomena of an entirely different nature was observed. 

 This bears more particularly upon the general question of "evolution 

 by mutation." The case is as follows: Two sex-linked mutants 

 (allelomorphs) were found in hybridization experiments to exhibit an 

 incompatibility that differs Uttle, if any, from cases of incompatibility 

 between species in nature. When reciprocal matings are made between 

 these mutants one of the combinations entirely fails to produce off- 

 spring and the other gives hybrids that are all sterile. As a result, it is 

 impossible to get a second generation from the cross. To Dr. Metz's 

 mind this suggests the possible origin of species in nature by means of 

 mutations that involve incompatibiUty. If three forms exist in nature 

 (as they do in the laboratory), of which two are fertile with the third 

 but infertile with one another, then the disappearance of the third form, 

 by means of natural selection or otherwise, would leave the other two 

 isolated as distinct species. No matter how slightly they differed to 

 begin with, the fact of isolation would allow them to develop inde- 

 pendently and to diverge into well-marked species. 



EFFECT OF ALCOHOL ON GERM-PLASM. 



The study of the effects of alcohol upon the germ-plasm of rats, 

 described in earlier reports, has been continued during the year by 

 Dr. MacDowell. While he has accumulated many data, it has been 

 thought best, before publishing, to note the results on a new set of 



