GABRIELLE L. C. HOWARD. 45 



In Type 16 and Type 35 (Plates III and IV) we have two 

 forms which, while not very different in height or habit, never- 

 theless differ markedly in the number of leaves. The main 

 object of the cross between these two forms was an investigation 

 of the leaf shape. Type 35 has the broadest and Type 16 one 

 of the narrowest leaves in all the fifty-one types, while the length 

 is nearly the same in both. The cross between Type 23 and 

 Type 38 (Plates V and VI) was primarily made on account of the 

 difference in the number of leaves on plants almost equal in 

 height. The average number of leaves in Type 23 is nineteen 

 or twenty, in Type 38 about thirty-five. Types 2 and 3 (Plates 

 VII and VIII) are both petiolate forms, but the petioles are 

 alate to a var3dng degree. In the cross Type 2 x Type 51 

 (Plates VII and II) a form with a petiolate leaf was crossed with 

 a leaf with a broad base. The most striking result of these 

 hybridizations was the formation of new forms quite outside 

 the range of either parent. For instance, in the Fo generation 

 of Type 16 x Type 35, dwarf plants, much shorter than Type 35, 

 and tall forms, almost equal in height to Type 16 and Type 35 

 combined, were found, and some of these have bred true in the 

 Fg generation. Similarly, in the cross Type 23 x Type 38, 

 petiolate forms breeding true were produced, although both 

 parents have sessile leaves. Reciprocal crosses were always 

 made and groA\Ti side by side both in the Fi and Fo generations. 

 In no case could any differences between the crosses and the 

 reciprocals be detected. The Fi was intermediate between the 

 parents in almost all the characters (Plates IX, X, XI, XII). 

 Details regarding the manner in which the actual measure- 

 ments were taken will be given in the section on the individual 

 characters, but a few words of explanation may be given here 

 on the number of plants used in the experiments. As explained 

 in a previous chapter (p. 38), the area of uniform land available, 

 the labour involved in the raising of the seedlings and the 

 transplanting, but more especially the shortness of the period 

 during which measurements can be made, limits the amount 



