JoHS. Schmidt 67 



studied is highly dependent on environments I am continuing the 

 work with the method of diallel crossings applied to various species and 

 am. especially prepared to test the theory of the method through a study 

 of the subsequent generations. 



1 It is not here the place to discuss at length the difficulties of technical or other 

 nature which the application of the method of diallel crossings may present. I shall only 

 mention that a small fecundity in a species does not necessarily exclude the application of 

 the method, when the character in question may be accurately measured, and, altogether, 

 is inherited in the same way as the character dealt with here. Let us imagine the least 

 possible fecundity, i.e. a brood of one, and let us suppose that we want to compare the 

 generative values of the two males ;r and y with regard to some quantitative character. 

 We then pair first x and later y with a series of females 1, 2, 3, 4, 5, etc. From this arise 

 two series of offspring- values which may be designated by xi, x^, x-i, Xi, x^, etc. and 

 Vi, y2, y3^ Vi, Vo, etc. The average of the values Xi-y^, x.2 - y^ , xz - Vs > -^4 - ^4 > ^5 - l/o . etc. 

 will then afford a measure of the difference between the generative values of the two 

 males, a measure the accuracy of which is inversely proportional to the square root of the 

 number of females paired with the two males ; the accuracy may therefore be increased by 

 increasing that number. 



5—2 



