202 



The Journal of Heredity 



involves only the fusion of rods or 

 the breaking of V's, except in the steps 

 between 5 and 6 and between 8 and 9, 

 where the change from a single rod to 

 a V (or vice versa) is represented. 

 The latter would involve the doubling 

 in size of a chromosome or the diminu- 

 tion of a chromosome by half. 



The steps involved in the other parts 

 of the series are evident from the 

 diagrams. One of them (10-16) in- 

 volves mjerely the disappearance of 

 the m-pair ; another (9-1-4) the dis- 

 appearance of this pair and the increase 

 in size of one of the V-pairs. Dia- 

 grams 13 and 14 might equally well 

 be put in other positions — as is true 

 also in the case of certain others — but 

 this would merely serve to complicate 

 the figure. 



From the standpoint of chromosome 

 evolution several questions arise from a 

 consideration of the above series. Fore- 

 most, perhaps, is the question as to 

 how far morphological criteria may be 

 relied upon as indications of genetic 

 homology of chromosomes. This has 

 already been answered partially, as 

 noted above, by the observations on 

 Drosophila zvillistonf which show that 

 in two out of three cases, at least, 

 similar pairs are not strictly homolog- 

 ous when this species is compared with 

 D. melaiwgastcr". However, it is pos- 

 sible, jjerhaps, to avoid the conclusion 

 that would seem to follow from this if, 

 as suggested above, we assume that V- 

 chromosomes may break in two to 

 form rods, and rods may unite to form 

 V's. On this assumption we might 

 start, for example, with type F and 

 make up types A and M (the two just 

 considered) in different ways depend- 

 ing on what particular rods we com- 

 bined to make V's. The homologies of 

 the rods might remain, but the result- 

 ing V's as units might not be homo- 

 logous. In this case the actual rela- 

 tions could only be determined by 

 genetical analysis of the chromosomes. 

 This is one of the problems with which 

 the actual genetical studies are con- 

 cerned. 



If chromosomes persist intact as 



hypothecated above the most definite 

 relation between the appearance and 

 the genetical make-up should be found 

 in the case of the small m-chromo- 

 somes. This pair of chromosomes is 

 present throughout almost the whole 

 series of types, with surprising con- 

 stancy of size, form and behavior. 

 Morphological evidence is all in favor 

 of the assumption of homology here, 

 but again it remains for genetical evi- 

 dence to furnish the proof. 



Chromosome Resemblances and 

 Taxonomic Relationships 



xA.nother question that presents it- 

 self is : What is the relation between 

 chromosome resemblances and taxo- 

 nomic affinity of the species involved 

 in this study? 



In general it is very difficult to de- 

 termine the detailed taxonomic rela- 

 tionships within the genus Drosophila 

 — partly becavise of the difficulty of de- 

 ciding upon the relative importance of 

 the various diagnostic characters. After 

 mentioning a few aberrant forms (only 

 one of which is included in the present 

 paper) Sturtevant* (p. 70) concluded 

 respecting the bulk of the species : 

 "I have been unable to even make a 

 satisfactory arbitrary division of them 

 into groups." 



In a few cases, however, two or 

 three species show such close resemb- 

 lances as to leave little question as to 

 their taxonomic affinity, and it is of 

 interest to note the chromosome rela- 

 tionships in these instances. One such 

 case is that of D. mclanogaster and D. 

 siiiiukuis, in which the species are al- 

 most identical and may be hybridized". 

 Both of the species belong to type A 

 with respect to their chromosomes. It 

 has not been definitely established as 

 yet that the rod-like chromosomes are 

 the sex-chromosomes in D. simu'.ans, 

 but it seems safe to conclude tentatively 

 that they are. 



Another case of close resemblance 

 is that shown by D. replcta and D. 

 mullcri. Indeed, stocks of these spe- 

 cies were reared in the laboratory un- 

 der the same name (re pic fa) until it 



