MENDELISM 23 



It also follows — and this is very important — that purity 

 of type or breed has nothing to do with a prolonged course 

 of selection or inheritance, as hitherto supposed. The black 

 and the white andalusian fowls born of bine mongrel parents 

 are just as " pure " as those which never had any " blue " 

 ancestors. 



There are some Apparent Exceptions to the rules I 

 have enunciated ; for instance, all white peas breed true, 

 and only produce white flowers when self-fertilized. Most 

 white crosses also breed true, but not invariably so. 

 Prof. Bateson found that amongst the white peas called 

 " Emily Henderson " the pollen grains are of two shapes, 

 though only one kind is found in any particular flower.. 

 When he crossed these two kinds, there resulted, not white 

 flowers, but purple ones only, in the Fj generation, and in 

 the Fij generation 9 purples to 7 white in every 16. 



White x White [" Emily Henderson "] 



Purple 



9 Purples 7 Whites 



Here we meet with what, at first sight, seems an 

 exception to the rule. One might have expected 3 Purples 

 to 1 White in the Fij generation (Law 6). But we are 

 really dealing with a case of di-hybridism, in which the 

 9-3 -y 1 numbers are changed to 97 by including the 3-3-1 in 

 one form. The explanation is very ingenious, and throws 

 a flood of light on what is known as " reversion to an 

 ancestral type." This is perhaps the greatest difficulty 

 which one encounters at the beginning of the study of Men- 

 delism, but I will endeavour to clear up the difficulty. It is 

 assumed that, in order to produce colour, two factors must 

 be present. If one factor is present in one parent, and the 

 other factor in the other parent, their mating will result in 



