W. Hatkson and Caromnk Pkllew 19 



cross-furtilisation docs certainly take place in peas. In the eomse of 

 many years experience of peas we have met with iin(loul)tcd cases at 

 least threi' times. Twice we have seen a pod of yellow, round seeds on 

 a green, wrinkled variety, and these seeds gave ordinary Mendelian 

 resnlts. Once in a sowing of tendrilled, white-Howered plants one 

 heterozygous for purfjle and for the acacia character appeared, and this 

 gave a normal Mendelian F.^. A few similar cases also have been given 

 us by Messrs Sutton. These occurrences are excessively rare, happening 

 perhaps once among many thousand plants, even when distinct varieties 

 ai'e grown in large numbers near each other. But in the case of the 

 rogues it must be remembered that there is no large body of rogues 

 from which the fertilisation could take place. They are utterly unlike 

 any modern variety. Great efforts are made to exterminate them, and 

 though doubtless an occasional individual gets missed, it cannot possibly 

 be supposed that these survivors can continually pollinate the sur- 

 rounding types. 



Lastly, in the particular case of Duke of Albany just described the 

 rogues were quite peculiar, and in them the general reduction in size 

 of the foliar parts went far beyond anything we had ever met with in 

 our own cultivations. The leaflets were very narrow and almost strap- 

 like, having also an abnormal neuration resembling fig. 9 C, Plate XI. 

 Nor can these rogues be supjjosed to have been a special F., form ex- 

 tracted from a cross with the ordinary rogues, for in the whole crop 

 there were no other rogue-forms at all. Whether the collateral families 

 which up till now have thrown no rogue will contintU' thus pure is of 

 cour.se uncertain. 



The ru(/ue.s breed true. 



To this statement there has been no exception in any of the three 

 varieties (N. P. U. ; D. A. ; E. G.). The behaviour of the intermediates 

 in E.G. will be described immediately, but the definite rogues whenever 

 they appear, and however bred, are in our experience incapable of 

 throwing any form higher, that is to say, more like the type, than 

 themselves. 



Intermediates in Duke of Albainj. 



Typical }il:iiits of 1). A., self- fertilised, with us have not thrown 

 definite intermediates comparable with those thrown by E.G. In any 

 row however plants with somewhat smaller parts may occur, and from 

 a commercial sample we have even raised a strain of such plants. But 



2—2 



