108 INHERITANCE IN RATS. 



child. The 1«>\\ er-grade grandparent has lower-grade hooded grand- 

 children, and the higher-grade grandparent has higher-grade hooded 

 grandchildren. This shows that the variation in grade is (in part at 

 least) </< not ii pic. As the experiment yields no evidence that the varia- 

 tion in the hooded character is due to independent modifying factors, 

 there remains no alternative to the conclusion that the single genetic 

 Mendelian factor concerned fluctuates in genetic value. Fluctuation 

 accordingly is not exclusively phenotypic, as DeVries and Johannsen 

 have thought, but may be genetic also. Hence racial changes may be 

 effected through selection by the isolation of genetic fluctuations, as well 

 as by the isolation of mutations. Moreover, genetic fluctuation makes 

 possible progressive change in a particular direction, repeated selection 

 attaining results which it would be quite hopeless to seek by any other 

 means. 



A SECOND REPORT ON MASS SELECTION OF THE 

 HOODED PATTERN OF RATS. 



The experiments in selection for the modification of the hooded pat- 

 tern of rats, when reported on by Castle and Phillips in 1914, had been 

 carried through 13 generations. Since then the experiments with the 

 same selected races have been carried through 3 or 4 additional genera- 

 tions, the results of which will now be described. Additional records have 

 also been obtained for certain of the generations reported on by Castle 

 and Phillips, which may now be combined with those previously pub- 

 lished. Thus, revised data, based on larger totals, may be given for 

 generations 12 and 13 of the plus-selection series and for generation 13 of 

 the minus-selection series. These do not materially change the results 

 previously obtained, but add to their trustworthiness. The additional 

 generations of selection show a continued progressive movement of the 

 racial character in the direction of the selection and indicate the exist- 

 ence of no natural limit to the progress which selection can make in 

 changing the hooded character. 



For details concerning the earlier history of the experiments and the 

 methods of grading the animals the reader is referred to the publica- 

 tion of Castle and Phillips. The grading scale (exclusive of the newer 

 and more extreme grades) is reproduced in figure 35, plate 7. Atten- 

 t ion may be called to the fact that the entire selection series, both plus 

 and minus, consist of animals descended from an original stock of less 

 than a dozen individuals. These descendants number more than 33,000. 

 In their ancestry, since the beginning of the selection experiment, not 

 a -ingle cross out of the race has occurred. At the same time no effort 

 has been made to avoid inbreeding. Brother and sister and cousin 

 platings are frequent in our records. Under these circumstances it is 

 inevitable that the selected races should have become much "inbred." 



