6 HEREDITY IN RABBITS, RATS, AND MICE. 



Two does derived from the first-mentioned cross and one derived 

 from the second were employed in various matings presently to be 

 described. The results observed in the case of all three were so similar 

 that they may conveniently be described together. It will be borne 

 in mind that all are F : hybrids between Dutch and self. 



When crossed back with the other original pure Dutch buck ( cf 3036, 

 grade 9), these three does produced 20 young of the grades shown in 

 table 14. We get here indications of segregation into two groups, one 

 like the FI mothers in grade, the other like the Dutch father, but no 

 sharp line of division separates the two. 



The same three F! females were also mated successively with an 

 FI male from each of the two crosses already described, with the 

 results shown in table 15. The results are similar in both cases, but it 

 will be noticed that the lower grade FI male (5029, derived from the 

 Himalayan cross) produced F 2 young of slightly lower grade. The 

 F 2 range extends from to grade 5 inclusive, average 1.80. The back- 

 cross range was from 1 to 7 inclusive, average 4.60. 



Certain of the F 2 young and the back-cross young of grade 4 or 

 higher, w T hich presumably would be homozygous for the Dutch char- 

 acter, if it mendelizes, were employed in building up a race of Dutch 

 rabbits for further study. This was done by back-crossing the selected 

 does a second time with the pure buck, cf3036, grade 9, with the 

 results shown in tables 16 and 17. Young were obtained which 

 ranged from grade 1 to grade 17, but which grouped themselves round 

 two modes situated at about grade 6 and grade 15 respectively. We 

 shall presently consider the distribution further. 



These same does were also mated with a male similar in origin to 

 themselves, viz, d 7| 5167 (table 18), a typical and evenly marked 

 Dutch buck of grade 7, produced by the original back-cross (table 14) . 

 He bred in all respects like his father (cT'3036, grade 9) when mated 

 with the same does, producing a bimodal group of Dutch young of 

 only slightly lower mean grade than the young which his father sired, 

 as might be expected from the fact that his grade was less than his 

 father's grade. (See table 18.) 



It was now evident that we had secured a race of Dutch rabbits 

 which produced only Dutch young and which derived their Dutch 

 character exclusively from the two bucks 3036 and 3037, and yet 

 which fluctuated in grade around two different modes. In fact, it 

 was soon discovered that the two original Dutch bucks were them- 

 selves heterozygotes of two different types of Dutch pattern which 

 corresponded with the two modal conditions found among their 

 descendants. Our next task was to isolate these two types in 

 homozygous form. This was easy in the case of the higher grade 

 (whiter) type, which proved to be recessive. A male of this "white" 

 type, 6175, grade 17, when mated with females of the same sort 



