638 
GENETICS: A. WEI N STEIN 
Proc. N. a. S. 
the X chromosomes of the two species are composed of homologous genes, 
a demonstration of such homology in the absence of hybridizing experi- 
ments requires a greater number of similar genes similarly placed than 
have yet appeared. 
Are Genes with Similar Somatic Effects Chemically Similar? — It might 
be supposed that factors with similar somatic effects are chemically sim- 
ilar even if they occupy different loci. But this supposition can be dis- 
proved by a consideration of two such factors as ruby and garnet in D. 
melanogaster. Each of these produces a pink eye, but the genes lie in 
different parts of the X chromosome. A ruby female has the composi- 
tion rG/rG, a garnet female the composition Rg/Rg. Now, if r = g 
and R = G, either of the above genetic compositions should produce 
the same effect as the composition rg/RG or rG/Rg; for in each of these 
cases there are two (supposedly similar) genes for pink eye color (r and g) 
and two (supposedly similar) allelomorphs of these (R and G). But the 
composition rg/RG or rG/Rg produces not a pink but a red (wild-type) 
eye. Hence either r differs from g, or R differ from G, or both. 
This proof depends on the assumption that the position of a gene does 
not influence its somatic expression. The correctness of this assumption 
is demonstrated by Bridges' cases of duplication and transposition 
already referred to, in which the shifting of a factor from one part of 
a chromosome to another or even to a different chromosome does not 
alter its somatic effect. 
iHyde, R. R., Amer. Nat., 49, 1915 (183-85, 185-187); Metz, C. W., Genetics, I, 
1916 i591-m7);Ibid., 3, 1918 (107-34) ; Sturtevant, A. H., Science, 48, 1918 (72-3). 
2 Weinstein, A., Genetics, 3, 1918 (135-73). 
^ D. virilis itself has been described by Sturtevant, Ann. Ent. Soc. Amer., 9, 1916 
(323-43) . Descriptions of the sex-linked mutants other than crossveinless will be found in 
Metz, /. c. Dscriptions of the sex-linked characters in D. melanogaster will be found 
in Morgan and Bridges, Carnegie Institution Washington, Publ., No. 237, 1916 (1-87), 
4 Sturtevant, A. H., Bridges, C. B., Morgan, T. H., These Proceedings, 5, 1919 
(168-173); Morgan, T. H., Sturtevant, A. H., and Bridges, C. B., Ibid., 6, 1920 (162- 
164). 
6 Muller, H. J., Amer. Nat., 54, 1920 (97-121). 
6 Castle, W. B., These Proceedings, 5, 1919 (25-32, 32-36); Ibid., 5, 1919 (500- 
506); Ibid., 6, 1920 (73-77). 
7 It was pointed out in Muller's article that the value which I had recently obtained 
for the hairy magenta distance was (on the data then available) 6.6, contrary to Castle's 
prediction; and that this, if combined according to Castle's own method with the 
values previously obtained by Metz in separate crosses for the hairy forked and forked 
magenta distances, would place the three factors in a nearly straight line in the order 
hairy forked magenta. This situation was used by Muller to show that, even by using 
Castle's method of combining separate experiments, linear results might be arrived 
at that were quite inconsistent with the prediction obtained by the same method. 
The new value, moreover, was based on more flies (total 364) than Metz's hairy forked 
value used by Castle (162). Muller made no claim, however, that these relationships 
were really correct, as it was evident that linear results arrived at by such a method 
