28 



" Rogues " in Culinary Peas 



Ci-osses hetwi'PM Rnyues and Roijues. 



Offspring 



In some the type was used as mother and in others as fatlicr, but the 

 results were the same. 



Fourteen of these plants which we may call F^ were bred from and 

 gave 949 plants all rogues, none of which icached a development 

 higher than that of class 5. The F^ plants were also rogues without 

 exception. This evidence taken together with the fact that no rogue, 

 however produced, has given anything but rogues, may be taken as 

 proving conclusively that the elements, whatever they may be, which 

 cause the distinction between the type and the rogue, are absent 

 altogether from the rogue. The types can produce both intermediates 

 and rogues; the intermediates produce a few typos and many rogues, 

 and the rogues breed true'. 



We have stated that the so-called F^ plants, when adult, were 

 thorough rogues. In the juvenile condition however the majority of 

 them, judged by the size and shape of leaves and stipules, differed little 

 if at all from the types. Actual rogues of classes 4 and 5 can be 

 distinguished from types as soon as any leaves appear. By the time 

 the sixth or seventh leaf is unfolded the difference is clear even to the 

 inexperienced. In the case of most rogues bred in F^ the lower leaves 

 and stipules in size, shape, and marbling are almost if not quite like 

 those of the types, and until about the 8th node is reached we cannot 

 as a rule distinguish them. After that level the narrowing of the parts 

 begins, and the rogue characteristics are rapidly assumed (cp. Plate X, 



' As it was conceivable that this behaviour of the rogues was due to apogamy, experi- 

 ments were made to determine whether this phenomeuon occurs. No indication of a 

 positive result was liowever obtained, nor have we ever seen seed formed in the course of 

 many similar trials made in the past with various forms of Phum. 



