﻿4 Darbishire, Mendelian and Biometric Theories. 



of the discussion lies very deep, and is to be found in the 

 clear appreciation of the fundamental relation between the 

 biometric and Mendelian points of view.* 



II. 



At a time when I did not clearly see this relation, 1 

 had before me some data which convinced me that the 

 Mendelian interpretation of the phenomenon of segrega- 

 tion was wrong, and that the facts were striking evidence 

 of the truth of Galton's theory. 



There are two attributes of a heterozygote which 

 are said to follow from the theoretical constitution of 

 its gonad ; one is that a quarter of the population 

 produced by the union of heterozygotes consists of 

 individuals bearing the recessive character ; and the 

 other is that half the population produced by mating 

 heterozygotes with recessives consists of recessives. My 

 hybridsf were tested for these two properties and the 

 results were not denied to be in accord with Mendelian 

 expectation. But this result was not conclusive in favour 

 of that theory only, because the proportion of recessives 

 demanded by Mendelian theory in the case of the first 

 property was identical, and in the case of the second only 

 slightly less than that which follows from the truth of 

 Galton's generalization.^ 



The Galtonian prediction of the number of albinos 

 that will be produced by two hybrids (H) each of which 

 is the offspring of a pure-bred waltzing and a pure-bred 

 albino mouse is '25 of their generation ; while on that 

 theory the proportion of albinos in a generation resulting 

 from the union of hybrids with albinos (A) is '53125, if 



* The reader who wishes to follow the discussion of the facts at first 

 hand will find the necessary references to four cases in the Appendix. 



t Biometrika, Vol. III., pp. 30 — 33. 



X Francis Galton, Proc. Roy. Soc, Vol. 61, p. 402, line 13. 



