BIOLOGY. 287 



nucleus fertilized by a different sperm. One clear case of this sort is a mosaic 

 from a back-cross in which the mother carried four recessives in one chromo- 

 some-III, and three others in the homologous chromosome. One side of the 

 mosaic showed the three characters corresponding to one of the original chro- 

 mosomes of the mother, and the other side showed three of the four char- 

 acters of the other original chromosome (the fourth was lost by crossing-over). 

 Another autosomal mosaic arose from a pair of flies that carried the recessive 

 gene, vestigial, in one chromosome II, and the recessive gene, speck, in the 

 other homologous chromosome. The mosaic was female and speck on one side 

 and male and vestigial on the other side. Since there is no crossing-over in 

 the male, the result shows that two sperm participated, one with an X and a 

 chromosome-II with speck, and the other with Y and a chromosome II with 

 vestigial. Presumably the egg had two nuclei, each of which after reduction 

 had an X; one of the nuclei contained the second chromosome with speck, 

 and the other nucleus the secojad chromosome with vestigial. 



An interesting type of gene has been found that changes a mutant char- 

 acter back to wild-type. For example, three sex-linked characters (scute, 

 tan, and hairy-winged) are changed to wild-type, each by a different sup- 

 pressor in chromosome-III. In the case of hairy-wing, the suppressor pro- 

 duces certain other slight effects that are similar to those met with in previous 

 cases of duphcation. This suggests that this suppressor is a piece of chromo- 

 some that contains the normal allelomorph of the gene that is suppressed and 

 that has been transferred to another location in the chromosome complex. 



Further work on a certain stock of lethal-7, in which the so-called tumor 

 fails to appear, has confirmed previous statements that in this stock the males 

 that formerly showed the tumor, now die at an early stage before the tumor 

 develops. A good deal of work has failed to show the nature of the change 

 that has taken place in this stock. It is interesting to note that when this 

 stock is outcrossed (using a lethal-7 female) some or many of the Fi male 

 larvae show the tumor, which proves that the gene for the tumor production 

 is still present. 



Some extraordinarily low female ratios have appeared in certain lines of 

 the double-yellow stock. As in other cases of change in sex-ratios of Dro- 

 sophila, the effects are produced not by a change of females into males (or 

 vice versa) but by the elimination of some of the individuals of one sex. 



The work on Drosophila simulans, referred to in previous reports, has been 

 continued. A useful new mutant gene has been found in each of the three 

 genetically-known chromosomes. That in the X-chromosome is an isomorph 

 of bobbed, which occupies, so far as we know, the extreme right-hand end 

 of the melanogaster X. There are now 8 isomorphic genes known in the 

 simulans X, including both the known ends of the melanogaster X. The 

 new mutant gene in chromosome-III extends the map of that chromosome, 

 making it at least 10 units (and probably 20 or more) longer than the corre- 

 sponding melanogaster map. Studies of changes in crossing-over, due both 

 to internal and to external causes, have been carried out with D. simulans 

 during the year. 



