GAMETIC PURITY 



203 



Cross overs occasionally occur resulting, for example, in the 

 F4 pure red early family shown in Table 3'-2. But if these two 

 were the onlj^ genetic factors involved in the cross, no "con- 

 stant" families of intermediate flowering time could result 



TABLE 32 



Variation in flowering time of two pure varieties of garden peas, one Early 

 \Miite, the other Late Red; and a classification, both as to color and as to flowering 

 time, of two hundred and thirty F4 families produced by crossing the two varieties, 

 these F4 families being all regarded as " constant " in flowering time because of 

 their low variability, as low as that of the parent varieties. Only the position of 

 the mean of each F4 family is given in the table, not its range as in the case of the 

 parent varieties. 



Days to Flowering 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 



from the cross. As a matter of fact more than half the F4 

 families are of this constant intermediate type, which shows 

 that one or more other factors, independent of the chief 

 factor for flowering time, must be concerned in the result. 

 Hoshino supposes that a single supplementary factor (not 



