E-Book Content
ASTRONOMY: F. H. SEARES 558 some has yet been obtained that contains both that factor and any mutant factor within its "sphere of influence." Backcross tests have shown that males of all the above types with respect to CIl and CIr resemble "normal" males in giving no crossovers at all. Included in the above tables are a considerable number of data involving three or more loci at once; and these agree with the table in showing clearly that the linear order of the factors established for the usual second chromosome is unchanged by the factors under discussion. The amount of crossing over is altered, often markedly, and n'ot usually proportionately in different regions; but the factors keep their same sequence. This result serves to emphasize the importance of considering the distances on chromosome maps as only. diagrammatic, not as representing actually .proportionate distances between the genes, although actual distance is evidently an important factor influencing the end result. It does not, I think, in any way weaken the case for the chromosome hypothesis, but merely shows, together with the results of Bridges5 (1915) and Plough6 (1917), that any chromosome map is available for purposes of numerical prediction only when the conditions under which it was made are duplicated. 1 Some of the early results were reported at the 1913 meeting of the American Naturalists, and brief references have been published by me (1915) and by Muller (1916). 2 Much of this data has not hitherto been published. It has been collected mainly by Dr. C. B. Bridges, to whom I am indebted for permission to use it. 'Muller, H. J., Amer. Nat., Lancaster, Pa., 50, 1916, (193, 284, 350, 421). 4 Sturtevant, A. H., Zs. Abst. Vererb., 13, 1915, (234). 6 Bridges, C. B., J. Exp. Zool., Wistar Inst. Philadelphia, 19, 1915, (1). 1Plough, H. H., these PROCEEDINGS, 3, 1917, (553-555). FURTHER EVIDENCE ON THE