62 H. K. HAYES 



theories advanced to explain heterosis seems desirable to set the stage for 

 later discussions. Bruce (1910) explained heterosis on the combined action 

 of favorable dominant or partially dominant factors, based as Richey (1945a) 

 has emphasized on mathematical expectations. 



Keeble and Pellew (1910) used a similar hypothesis on a di-hybrid basis 

 to explain hybrid vigor in peas. East and G. H. Shull (1910-1914) believed 

 vigor was dependent on heterozygosis on the basis that the stimulus of hy- 

 bridity was not entirely Mendelian. A. F. Shull (1912) preferred the explana- 

 tion that heterosis was due to a stimulus resulting from a changed nucleus 

 on a relatively unaltered cytoplasm. Jones (1917) restated Bruce's concept 

 and added the concept of linkage. 



Collins (1921) and Richey (1945) have pointed out that where large num- 

 bers of factor pairs are involved it would be very difficult to recover all fac- 

 tors in a favorable condition in F2, or in later segregating generations. With 

 multiple factors involved, however, linkage must of necessity make the re- 

 combination of factors more difficult. East (1936) presented a Mendelian 

 concept of the interaction of alleles at the same locus to explain heterosis, 

 where two alleles of a particular gene pair had each developed a divergent 

 physiological function. The writer believes he continued also to accept the 

 previous explanation that heterosis was dependent on the cumulative effect 

 of dominant or partially dominant linked genes. 



Gustafsson (1947), Hull (1945a), Jones (1945), Castle (1946), and others 

 have emphasized the importance of interallelic action in relation to heterosis. 

 Castle has suggested also that the effect of interallelic action of a single pair 

 of genes "is similar to that of the killer mutation of Sonneborn, except that 

 the action induced in the dominant gene by its sensitized recessive, instead 

 of being harmful, in this case is beneficial." 



In certain cases a homozygous recessive pair of genes may completely 

 modify the normal expression of either a homozygous or heterozygous or- 

 ganism. Homozygous dwarfs in maize condition such a result. A cross be- 

 tween two different dwarfs, however, releases the inhibition of each dwarf 

 and results in marked heterosis. Both dominant factors, where two dwarfs 

 are crossed, appear to be necessary to condition normal development. In this 

 case the dominant conditions of both factor pairs act as complementary fac- 

 tors for normal growth. 



It is evident that genes are greatly affected in their expression by differ- 

 ences in both external and internal environment. Cytoplasmic inheritance of 

 male sterility may be used for illustrative purposes. Several cases of male 

 sterility in sugar beets and onions, for example, are known that are due to 

 maternal cytoplasmic inheritance which may be modified in expression by 

 the dominant or recessive condition of one or more factor pairs. 



Recently Hsu (1950) at Minnesota has studied the effect of two pairs of 

 dwarf factors of maize in their homozygous dominant and recessive condi- 



