16 "Rogues" in Cidhuiri/ Peas 



curved pod is, so far as we know, always found on these rogues, though 

 on the types from which they are derived all well grown pods have the 

 dorsal edges straight for most of their length^. These differences in 

 shape are evidently all expressions of differences in proportional growth 

 along the several axes, and we incline to regard them as various con- 

 sequences of a deficiency of extension of the foliar parts in the rogue, 

 but we are not able to analyze the phenomena further, (See Plates XII 

 and XIII, figs. 10—15.) 



In the rogues of E. G. and D. A. the curved pods are, as we have 

 said, a striking feature. Attention should however be called to the 

 circumstance that the pods of both these varieties are pointed. In 

 N. P. U. they are blunt and the pods of rogues from this variety are 

 straight. Moreover the Serpette and other varieties normally curved 

 have also pointed pods ; and though our evidence on this question is 

 insuflfieiont we are inclined to regard the blunt-ended pod as incom- 

 patible with a full degree of curvature. 



Seeds. 



In shape and colour the seeds of the types and of the rogues are not 

 distinguishable from each other, but those of the rogues are slightly 

 smaller (see later). There is however a difference in flavour, those of 

 the rogues, whether raw or cooked, being less sweet to the taste than 

 those of the types. The starch grains of the two forms appear identical. 

 Estimations have been made of the amount of sugar in type and rogue 

 seeds, but so far no consistent difference has been discovered. Probably 

 therefore the chemical distinction, which must certainly exist, is more 

 profound. 



Production of Roffues by the Ti/peti. 



At first sight it seems improbable that the thin, weed-like rpgues 

 can be actually the offspring of the luxuriant types, and we suspected 

 the case to be one of accidental mixture. Rogues however, apart from 

 their special features, have the characteristics of the types among which 

 they occur. They flower at the same time whether early or late, their 

 pods are stumpy or pointed according as thi' type is stumpy or pointed, 

 and so on. But even if the rogues had arisen originally from the types 

 or simultaneously with them, their perpetuation might be merely due 



1 In nearly all varieties there is a slight upward curve beginning in the peripheral 

 quarter. 



