32 HEREDITY IN RABBITS, RATS, AND MICE. 



viz, 18.5 per cent. It is safe to conclude that the linkage strength of 

 red-eyed yellow with pink-eyed yellow is close to 18 per cent. 



We may pass now to the question whether the linkage strength is 

 the same in spermatogenesis as in oogenesis, whether it is the same 

 among the gametes formed by Fi males as in those formed by Fi 

 females. A priori we might well expect it to be different in the two 

 cases, since in Drosophila crossing-over has been found to occur only 

 in females, whereas in the silkwonn it has been found to occur only 

 in males. In publication 241 the fact was demonstrated that crossing- 

 over does occur in both sexes of the rat, but we were at that time 

 unable to state what its relative frequency was in the two sexes. Our 

 back-cross series of matings give data for such a determination. (See 

 table 48.) It will be observed that the estimated percentage of cross- 

 over gametes is somewhat higher for females than for males in both 

 the repulsion and the coupling series and that the difference is greatest 

 where the numbers are largest, viz, in the coupling series. This 

 would suggest that crossing over occurs more readily in oogenesis 

 than in spermatogenesis, but I doubt very much whether such is the 

 case when all other conditions are the same. Summaries made for the 

 concluding period of our experimental work, when conditions had been 

 more carefully controlled and the procedure of taking the records had 

 been best standardized, show no appreciable differences in the case 

 of the two sexes. For this period, in the coupling series, Fi females 

 gave 187 dark and 277 j^ellow j'oung, or 19.3 per cent cross-over 

 gametes. Simultaneously, Fi males of similar parentage gave 133 

 dark and 197 yellow j^oung, or 19.4 per cent cross-over gametes. In 

 the repulsion series only Fi males were at this time being used to 

 any great extent. They produced 29 dark young and 269 yellow 

 young, which by the method of calculation already explamed indi- 

 cates 19.4 per cent cross-over gametes, a remarkably close agreement 

 with the results given by both sexes in the coupling series at this 

 same period. 



Wliether external conditions have any influence on the percentage 

 of cross-overs we are unable to state, but this seems doubtful in the 

 case of a warm-blooded animal such as the rat. That individual or 

 age differences may occur among Fi animals afTecting the percentage 

 of cross-overs is a possibility we have considered carefully, but with 

 only negative conclusions. The indicated percentage of cross-overs 

 varies in the case of particular Fi males from to 44 per cent, but this 

 variation appears to be the result of random sampling rather than of 

 consistent differences in genetic behavior. Several males showing 

 extremely high or extremely low percentage of cross-overs were trans- 

 ferred to new breeding-pens and mated with other double recessive 

 females. Their indicated percentages of cross-over gametes before 

 and after the transfer showed no consistency with each other, and 



