218 Mendiiian Inheritance in Bryonia dioica 



Extracted types, witli leaves identical in shape, colour and incision 

 with those of the original parents, are shown in the photographs in 

 Plate XXXIII, figs. 9-12. 



Of the parents, variety B has stems with comparatively shorter 

 intemodes, leaves dark-green in colour, rough and deeply-lobed. 

 (Plate XXXI, fig. 3.) 



Variety G has stem with — on the average — longer internodes, leaves 

 paler in colour, smoother and less deeply-lobed. (Plate XXXI, fig. 3.) 



The range of variation in leaf-shape among individuals of the two 

 varieties is evident from the photographs, which make clear that 

 although considerable variation in leaf-shape exists, and the most 

 deeply-lobed leaf-type of variety approximates to the least deeply- 

 lobed type of variety B, no real overlapping of the two types occurs. 

 (Plate XXXII, tigs. 4, 5.) 



If the average form of leaf is considered (which determines the 

 appearance of the plant as a whole), the two varieties are widely 

 different. 



In the F^ generation of the reciprocal crosses B x G and G x B the 

 foliage is intermediate in character with a bias towards the B type. 

 The range of variation in leaves of i^i is shown in Plate XXXII, fig. 6. 



In the cross F, x G new types of foliage appear, and it is not easy 

 to classify these new combinations, especially as the characters are not 

 such as can be expressed numerically. 



It may be said with certainty, however, that the pure G type of 

 foliage can be extracted from this cross and that the pure B type does 

 not appear. Shoots from a plant having this constitution are shown in 

 Plate XXXII, fig. 8 c (17/110 with characters similar to those of the 

 heterozygote. 



When the heterozygote is selfed numerous types of foliage also 

 appear, and among them individuals with the pure G and pure B type 

 of leaf may be identified. (Plate XXXII, fig. 8, d and e, and Plate 

 XXXIII, figs. 9-12.) The exact proportion of these "pure extracted" 

 types is difficult to determine. 



Although, as stated above, plants of the two varieties B and G show 

 considerable range ot variation in leaf-shape, the two classes do not 

 overlap. 



When, however, the pure forms are mixed with the numerous 

 hybrids an almost continuous series of foliage forms is presented. 



Thus, while it is possible to recognize a given plant as a typical 

 representative of the G type of foliage, it may be almost impo.ssible to 



