762 Isabel McCracken 
by Castle,’ Davenport,? Weber,'® and other thrematologists that 
the Mendelian conception of purity is at least not applicable to all 
classes of characters. It shows as Webber has shown in a later 
paper on “Plant morphology” (quoted in paper cited), that the 
anlage of a character, in the present instance “bivoltinism’’ may 
be masked for several years and then reappear to an appreciable 
degree. 
The evidence points strongly to a difference in female and 
male potency in ability to give expression to the character, the 
difference in potency of transmission being mainly (if not entirely) 
dependent upon the nature of the characters concerned and their 
relative potency, this being presumably determined by their past 
history. The evidence shows conclusively that while though 
failure of expression occurs in the first, second and even third 
generation, the non-expressed character must be reckoned with 
as part of the inheritance and its reappearance looked for. 
It shows further that the character bivoltinism loses none of 
its potency although not exhibiting any outward manifestation, 
but, on the other hand gains or is reinforced through reversions. 
While the process of selection may be materially assisted by | 
selecting prepotent individuals, ignoring these does not preclude 
the possibility of effective selection. This depends upon the 
relative value of the characters concerned. The relative value 
appears to change from generation to generation according to the 
amount of each character invoked in the cross modified by the 
normal relative potencies of each. For instance, bivoltinism 
begins in Series A’ (Table II) with a percentage value of less 
than 10 per cent. It accumulates from generation to generation, 
through lines in which there is no “expression” until, in Series 
K (Table IV) it shows a percentage value of over 25 per cent. 
In Series D (Table II) we find it with a percentage value of about 
14 per cent. It accumulates from generation to generation 
through lines in which there is “successive expression” until in 
8 Castle: Origin of a Polydactylous Race of Guinea Pigs. Contri. from Zodl. Lab. of Mus. of 
Comp. Zoél. of Harvard College, no. 176, 1906, p.29. Pub. of Carnegie Inst. of Washington, no. 49. 
®' Loc. cit., 1907. 
10 Weber: Some Gaps in our Knowledge of Heredity. American Breeders’ Association, iii, p. 344, 1908. 
