Genetic Factors 419 



condition, the dominance of linked genes, increased embryo size, and 

 others— but no satisfactory explanation has yet been reached. Gene ac- 

 tion in heterosis has been discussed by Jones (1957), and there is a 

 wide literature in this field (Sprague, 1953). 



Genes and Form. The chief morphogenetic significance of genetic fac- 

 tors, however, is in their relation to the development of organic form. Here 

 it is not the total amount but the distribution of growth that is impor- 

 tant. Genes must in some way control relative growth— the amount of 

 growth in one dimension as compared with that in each of the others 

 so that specific shapes are produced. Many instances could be cited 

 where the shape of leaf or flower or fruit is certainly inherited and where 

 differences in it segregate and can be analyzed in mendelian terms, at 

 least to some degree. Only a few can be mentioned here. 



Leaf shape in cotton has had particular attention. In one of the earliest 

 analyses of shape inheritance, Leake (1911) found that in crosses be- 

 tween broad-lobed and narrow-lobed forms the F^ was intermediate and 

 the F 2 showed segregation into i/4 narrow, V> intermediate, and 1 4 broad. 

 Peebles and Kearney ( 1928 ) crossed shallow-lobed and deep-lobed types 

 and found F x to be intermediate and a ratio of 1:2:1 in F 2 . In some 

 varieties of cotton the genetic situation is much more complex. Both 

 Hutchinson (1934 and others) and Silow (1939) postulate a series of 

 multiple alleles, chiefly affecting lobing. They believe that the genes are 

 "compound" and vary qualitatively as well as quantitatively. Hammond 

 (1941) showed the importance of developmental analyses of shape in 

 individual leaves and of changes of leaf shape along the stem. This 

 method was carried further by Stephens ( 1944 ) . 



In the Japanese morning glory, Pharbitis, Imai (1930) and a number 

 of other Japanese geneticists have studied the complex situation presented 

 by the inheritance of leaf shape in crosses among its many varieties. 



Among other traits of form the inheritance of which has been analyzed 

 in mendelian terms are fruit shape in Bursa (Shull, 1914), root shape in 

 radish (Uphof, 1924), and leaf lobing in Tropaeolum ( Whaley, 1939). 



Evidence for the Existence of Genes for Shape. The problem under- 

 lying all these instances of the inheritance of form is to find the method 

 by which genes determine what the form is to be. The fact that such 

 traits show segregation suggests that genes control them directly, but 

 it is difficult to see how this is done. Some geneticists have tried to sim- 

 plify the problem by assuming the operation of genes for individual 

 dimensions only, as in the case of vine length in peas. Thus in tobacco 

 flowers Anderson (1939) studied the inheritance of tube length and 

 limb width in crosses between Nicotiana Langsdorfii and N. alata, species 

 which differ in corolla shape. He observed that in F 2 there was much 

 sharper segregation for length than for width, suggesting that fewer 



