66 • Racial Studies ^V^ Fishes 



The table shows that from the personal value of an individual few 

 or no conclusions can be drawn with regard to the generative value. 

 While X and z exhibit the same personal value, viz. 59, the generative 

 value of X is about 1 greater than that of z. On the other hand h and 

 d are nearly identical as regards generative value whereas one of them 

 has realized 1 vertebra more than the other. Between the personal 

 values of a and h and of h and c the differences are in both cases 2, but 

 the differences of the generative values are 4'25 and 1'32 respectively, 

 etc., etc. 



Concluding Remarks. 



It was mentioned in the introduction that the previous investiga- 

 tions with other species of fish had shown that the number of vertebrae, 

 etc. certainly was influenced by the environment, but still primarily was 

 determined by inheritance. As regards the trout-experiments the fact 

 alone that it has been possible to work, apparently successfully, with the 

 notions personal and generative values and fix them numerically involves 

 a confirmation of the results of the previous investigations. 



Of greater general or rather practical importance is the question, 

 whether the experience gained from the trout-experiments, so favourable 

 in regard to technique, can be made use of in the study of other quan- 

 titative characters of greater practical importance in animals and plants. 

 The greater the difference between personal and generative value of an 

 individual the greater the importance, e.g. for the breeder, of a basis 

 which enables him to select animals to breed from, by a systematic 

 valuation of their generative values. 



In such cases where it appears that some important quantitative 

 character is inherited in the same way as the number of vertebrae in the 

 trout it may be hoped that the method of diallel crossings with the 

 determination of the average values of the offspring- combinations will 

 prove to be of use. 



In most cases, at any rate in the animal kingdom, it will probably 

 be difficult to obtain so great an accuracy in the determination of the 

 generative value of an individual as with the trout. This species is 

 distinguished by a great fecundity and by the great ease and accuracy 

 with which the quantitative character in question could be measured. 

 However, even a less accurate determination would often be of impor- 

 tance, e.g. if the question was of a comparative valuation of some 

 personally identical individuals of a species in which the character 



