328 CARNEGIE INSTITUTION OF WASHINGTON. 



types already known in melanogaster. This makes it possible to be certain 

 that one is dealing with the same loci in making comparisons between the 

 loci of the two maps, which is not possible in other species of Drosophila. 

 The maps of the chromosomes I, II, and III have been improved so that 

 better comparisons can be made with the corresponding three chromo- 

 somes of melanogaster. The most striking result of the comparison has been 

 the discovery of a difference in sequence in three identical loci in chromo- 

 some III. Thus, in melanogaster the order is scarlet, pink (peach), Delta; 

 while in simulans the order is scarlet, Deltoid, peach. Equally interest- 

 ing is the difference in cross-over values. Thus scarlet and peach in simu- 

 lans are approximately 75 units apart, while in melanogaster they are only 3 

 units apart. 



In connection with the work on other species of Drosophila, an examination 

 has been made of types in the British Museum and in the museums at Lund 

 and at Amsterdam. The object of the examination was to straighten out the 

 taxonomy of American and other forms. 



The possession of a stock containing triploid (3n) individuals, recorded in 

 the last report, furnishes an opportunity to study the possible relations 

 between the presence of three "homologous" chromosomes and the phe- 

 nomena of equational non-disjunction, also an opportunity to examine into 

 what influence such a triune relation may have on crossing-over. Stocks are 

 being prepared, suitable for such work, in which 5 loci are marked on each of 

 the 3 X-chromosomes of each 3n female. 



From time to time we have made attempts to get some idea of the number 

 of genes in Drosophila. Several methods have suggested themselves, but we 

 have realized that the data could at best furnish only limiting values. If the 

 number of genes could be even roughly estimated, and if we could determine 

 the length of the actual chromosomes, it would be possible to get some idea 

 as to the limiting values for the size of the genes themselves. We have 

 approached the question with many misgivings, realizing the inadequacy of 

 the data and realizing also that no matter how cautiously the results are 

 stated, they will be quoted and the reservations forgotten. However, by 

 pointing out at each stage of the calculations the sources of error, we feel that 

 we have done all that is necessary to guard ourselves against misrepresenta- 

 tion. Without going into the methods of procedure here (they will be 

 briefly given in the Croonian Lecture for 1922), it may be stated that the 

 calculations place the size of the gene somewhat beyond the range assigned to 

 hemoglobin molecules. If further work confirms this view, then it follows 

 that the changes in the genes, that we call mutations, are of the order of 

 magnitude of molecular phenomena. 



The work on chromosome III of Drosophila is now ready for publication. 

 In addition to a description of all the mutant races known to us to the end of 

 1921, the data are given on which the location of the genes is based. A full 

 account of the methods in which the raw data are handled in order to get as 

 accurate positions as possible for the genes in the chromosome maps is given 

 in detail. We hope this complete account of the procedure may show what 

 corrections of the data are necessary in order that the placing of the genes may 

 serve to meet the situations most often encountered. 



