1908.] Theory of Ancestral Contrihutions in Heredity. 67 



seeds could arise : for, as the table on p. 62 shows, neither of the other green 

 rounds used occurred in a family in which wrinkleds also occurred. 



The nature of the mating enclosed in brackets in the pedigree on p. 61 and 

 of its subsequent progeny is of the type 



EE x DD 

 DE 



/ I \ 

 EE DE DD 



as far as the colour of the cotyledons is concerned. But with regard to the 

 shape of the cotyledons, the nature of the mating, in those cases which gave 

 wrinkleds in F 2 , was of the type 



DE x DD, 



giving equal numbers of DE and DD individuals, the former of which, when 

 allowed to self-fertilise, should produce rounds and wrinkleds in the pro- 

 portion of 3 to 1. 14 families from 5 different crosses contained wrinkled 

 seeds ; in the case of 5 of these families I did not record the numbers of 

 wrinkled, because I was not immediately alive to the interest of the case, 

 and contented myself with recording the fact that both round and wrinkled 

 did occur. In the remaining 9 families I recorded the numbers, and the 

 totals were 2605 round and 914 wrinkled, which gives a percentage of 

 25 - 97 wrinkleds. Let us look for a moment at the ancestry, that is to say, 

 the characters of the ancestors of this cross. The round hybrid, which gave 

 rise to the F 2 in which this proportion of wrinkled occurred, was the result 

 of the union of two rounds, one of them, the yellow, presumably a pure 

 round, and tbe other, the extracted green, a hybrid round. This hybrid 

 round was descended from an unbroken line of 4 " round " ancestors (each 

 of them hybrids in the Mendelian sense) ; whilst behind that point half of 

 the ancestors were round and half wrinkled. It will be convenient to 

 display this ancestry in the form of a pedigree (Pedigree A), in which the 

 crossing which gave rise to the F 2 in question is enclosed in a square 

 bracket. Each ancestor is represented by a circle, and its character is written 

 immediately to the right of the circle ; the nature of the gametes borne by 

 each ancestor being written within the circle. 



Now let us write the pedigree on p. 61 out again (Pedigree B) in the same 

 way as this, and compare the two (see p. 68). 



The result obtained in F 2 from the experiment with cotyledon shape 

 (Pedigree A) affords a more convincing a posteriori demonstration than that 

 afforded by the experiment with cotyledon colour (Pedigree B), of the 

 correctness of the course of leaving the soma out of account in the attempt 

 to predict the result of a given mating. For, as is clearly seen by a 



F 2 



