BIOLOGY. 291 



the Carnegie Institution of Washington, are nevertheless direct out- 

 growths of the work done under the grant. 



In the course of the work several unexpected and important problems 

 have opened up; e. g., the possibilities of dislocation within the linear 

 series of genes either by ''duplication" of sections of the series or by 

 "deficiencies" of sections. These and several other related questions 

 (various types of non-disjunction, etc.) are being attacked both by 

 genetic analysis and by cytological examination. 



Several hundred mosaics have been found; these may involve any 

 one of the first three chromosomes, but are especially common for the 

 first chromosome, where sex-characters are also involved (gynandro- 

 morphs). These mosaics bear directly on the questions of the distri- 

 bution of the cleavage nuclei and of the independent differentiation of 

 parts. They also furnish explicit data as to their own mode of origin. 



New mutations in all the three large groups have continued to appear 

 at an undiminished rate. As a result, the working material is con- 

 tinually being enriched, for new loci are becoming available for studies 

 involving the use of linkage, more viable races are being substituted for 

 poorer ones, and characters are arising that are more readily separable 

 and that interfere less with the simultaneous classification of other 

 characters. In addition to offering new means of approaching old 

 problems, the new mutations are continually bringing in their train 

 entirely new problems. 



Finally, a great deal of time has been given to the question of the 

 stability of the Mendelian genes. It has become evident that this 

 problem can be most satisfactorily studied by means of linkage, and 

 since Drosophila is the one form in which the linkage relations are 

 thoroughly known, it is undoubtedly the best material for such work. 

 Through the earlier work of Muller, Altenburg, and Dexter done in 

 this laboratory, and through the intensive work of the last two years, 

 it is now possible to demonstrate in the strictest sense the existence 

 and Mendelian behavior of modifying genes. The immediate bearing 

 of this demonstration on the problems of selection and of evolution is 

 evident. 



