GABRIELLE L. C. HOWARD. 101 



width, and the amount of indentation at the apex and base. 

 All these characters vary independently of one another and can 

 be inherited separately, and their mode of inheritance can be 

 explained by the existence of independent interchangeable 

 factors. 



8. CHARACTER OF THE SURFACE AND THE MARGIN 



OF THE LEAF. 



These characters are very difficult to investigate. In the 

 first place, the nature of the irregularities in the surface may 

 be different ; in the second place, the various stages and 

 combinations are very difficult to distinguish by eye ; in the third 

 place, even the so-called fiat leaves have a certain amount of 

 undulation at the base. From a study of the various Indian 

 types, the following conditions appear to be the most frequent — 

 large undulations of the whole leaf as in Type 9, large undulations 

 confined to the base only, a general puckering of the surface 

 between the veins and a small undulation or "frilling" of the edge. 

 This frilling of the edge occurs in Type 51, but is best shown by 

 the photograph of a plant which arose from the cross Type 9 x 

 Type 51, namely, No. 694-23 in Plate XVI. 



The investigations on this point have been confined to 

 the cross Type 9 x Type 51. Type 9 has a leaf which is very 

 undulate aU over, while in Type 51 the edge is frilled but the 

 surface is flat except at the extreme base, where an occasional 

 slight undulation may occur. Taking the leaf surface only, 

 it was found that the F^ was intermediate and that the Fg 

 showed a series of forms intermediate between both parents, 

 with a slight intensification, a few plants being more undulate 

 than Type 9. Two hundred and fifty plants were examined by 

 two observers. It was found possible to distinguish seventeen 

 plants like Type 9, and three slightly more undulate. Fifty- 

 seven plants were found with a slight undulation at the base or 

 quite fiat. These two classes could not be sub-divided. The 



