92 INHERITANCE IN GUINEA-PIGS. 



VARIATIONS OF SEPIA. 



We find rather more overlapping of distributions among the sepias 

 than among the yellows when different genetic constitutions are com- 

 pared. Nevertheless there are significant differences in the means. 

 The groups C d Cd, C d C r , and C r C r with means from sepiaj to sepia 2 .i, 

 nearly black, average distinctly darker than groups C d C a and C r C a with 

 means of sepia 4 . 5 and sepia 4 . 7 , respectively. The case is quite different 

 from yellow dilution in which C d C r and C d C a have the same effect (or 

 nearly so) contrasting with C d C d . C r seems to be essentially identical 

 with C d in effect on black, but like C a in effect on yellow. 



For further analysis we must compare stocks. In the miscellaneous 

 stock the average for C d C a is sepia 5 5 . When this stock is crossed with 

 albinos of BW stock the average of the young again C d C a is sepia 4-3 . 

 When these are crossed again with BW albinos the average becomes 

 sepia 3 . 5 . The darkening influence of the BW stock is apparent. The 

 South American stock also has a darkening influence with an average 

 of sepia 4 . 2 . We find a similar difference between the miscellaneous and 

 South American stocks among the homozygotes. 



The comparison of C d C a with C r C a within the same stock (South 

 American) yields a slight but probably significant difference (C d C a , 

 sepia 42 ; C r C a , sepia 47 ). Thus there is a difference of 0.5 with a prob- 

 able error of 0.12. It is certain that some of the red-eyed sepias have 

 been paler than any black-eyed sepia. 



If there is a real difference here, we would expect C r C r to be lighter 

 than C d C d or C d C r , but the 4 individuals known to be C r C r give the 

 darkest average of any array. They were not, however, a random 

 sample and, further, were either pure lea or F 2 IcaXBW and hardly 

 to be compared in stock with those known to be C d C d or C d C r . For 

 the present C d C d , C d C r , and C r C r may be considered identical in effect 

 on black fur. 



As in the case of yellows, the most critical test of the hypothesis of 

 imperfect dominance is the success of prophecy. In litters which 

 should give both C d C d and C d C a , the 2 darkest tested (sepia 2 ) both 

 proved to be homozygous, while 8 others (sepia 4 to sepia 7 ) proved to 

 transmit albinism. Among those which when graded might be either 

 C d C r or C d C a , 18 were tested. There is some overlapping of ranges, but 

 those which were found to transmit red-eye average very distinctly 

 darker than those which transmitted albinism. Four were tested in an 

 F 2 from red-eye by albino. The 3 dark ones, including 2 which were 

 actually as black as blacks of the BW race, proved to be C r C r , while 

 the other, sepia 4 , had albino young and was therefore C r C a . 



Table 40 shows that in the case of the sepias there is a very perceptible 

 darkening with age. This is shown in all groups except the homozy- 

 gous red-eyes, which were practically jet black to begin with. Another 

 interesting point brought out is a race difference in the amount of 



