100 CARNEGIE INSTITUTION OF WASHINGTON. 



chromosomes in the egg-cells of the tetraploids, we have presumably 1 egg- 

 cell with 23, 1 with 25, and 6 with 24 chromosomes. This is the same pro- 

 portion as was found in the distribution in the pollen-mother cells, namely, 

 three cells giving 24 and 24 to one cell giving 23 and 25. 



''Chromosomes of triploid Daturas. — The results from crosses with tetraploid 

 pollen add to the evidence as to the assortment in the megaspores. Com- 

 bined with the previous results from triploid pollinated by diploid, we now 

 have indirect proof as to the existence in triploid Daturas of egg-cells with 

 12, 13, 14, 15 ... . 22, 23, and 24 chromosomes. 



"Non-disjunction in diploid Daturas. — A special examination of 500 second 

 metaphases of normal diploid Daturas by Miss A. D. Bergner resulted in the 

 finding of one (complex) case of non-disjunction (11 and 13). This rare non- 

 disjunction is presumably the basis for the rare formation of (2w+l) mutants. 



"Pollen-mother cells of haploid Daturas. — The pollen mother-cells of the 

 haploid being taken as 1 in volume (measured between the first and second 

 metaphases inclusive), those of the diploids are nearly 2, the triploids nearly 3, 

 and the tetraploids nearly 4. 



"Evolutionary significance of tetraploidy. — If any natural species have lately 

 sprung from tetraploids, they should, if their chromosomes differ in size, 

 show 4 of each size instead of 2. They will then show 2 of each size in the 

 haploid state. In studies of species for this purpose plants which show clear 

 chromosomes may be classified as follows: (1) marked size differences; (2) size 

 differences, no haploid pairs; (3) size differences, some haploid pairs; (4) size 

 differences, all in haploid pairs." 



COMPARATIVE STUDY OF THE CHROMOSOME GROUPS IN DIPTERA. 



After it became clear through the work of Morgan, Sturtevant, Bridges, 

 and Muller that characters of the adult Drosophila melanogaster are largely 

 determined by genes which appear to have a definite locus in the chromo- 

 somes, it was a natural inquiry whether related species had a related con- 

 struction of the germ-plasm. Since the problem of evolution of organisms 

 has shifted from that of the soma of organisms to that of their chromosomes, 

 inasmuch as mutations occur primarily in the chromosomes, it becomes im- 

 portant to know about the genetic constitution of the germ-plasm of related 

 species. 



This is the problem with which Dr. C. W. Metz is engaged, with the assist- 

 ance of Miss Mildred S. Moses, Miss Ruth Ferry, and, temporarily, of Mr. 

 S. K. Emerson. The work involves the genetic behavior of mutants. It is 

 slow and laborious, but significant advances have been made. 



Breeding Work. 



In Drosophila willistoni the results of the study of 28 sex-linked mutant 

 characters have been published during the year: 



[These serve as a basis for] "beginning the detailed comparison of the 

 genetic constitution of this species with that of the other species under obser- 

 vation. The recent discovery by Miss Ferry of additional mutant characters 

 that appear to parallel those of other species provides the type of material 

 needed for this comparison. The genetic relations of these characters have 

 not yet been fully analyzed, but the available data suggested a more definite 

 relation to conditions found in other species than was intimated in last year's 

 report. 



