230 DONALD F. JONES 



multiple genes can be ruled out. Very few specific examples of single gene 

 action are available. 



In one case studied by the writer there is clear evidence for an interaction 

 between alleles (Jones, 1921). A mutation in a variety of normally self-ferti- 

 lized tobacco changed a determinate plant into an indeterminate, non- 

 flowering variation. It was a change in the normal response to the summer 

 day length period. The mutant plants failed to flower in the normal growing 

 season and continued in a vegetative condition. Reciprocal crosses between 

 the mutant and normal types both grew at the same rate as the normal 

 plants showing complete dominance of the normal growth rate. The hetero- 

 zygous plants continued their vegetative growth longer and produced taller 

 plants with more leaves and flowers than the normal homozygous plants. 

 This result I consider not to be heterosis, since there was no increase in 

 growth rate. It is merely an interaction between alleles to produce a result 

 that is different from either parent. There are undoubtedly many allelic 

 interactions of this type. Whether or not they can be considered to contribute 

 to heterosis is largely a matter of opinion. 



Other cases in corn where heterosis resulted from degenerative changes 

 (Jones, 1945) were at first assumed to be single allelic differences, since they 

 originated as mutations in inbred and highly homozygous families. The de- 

 generate alterations were expressed as narrow leaves, dwarf plants, crooked 

 stalks, reduced chlorophyll, and late flowering. All of these mutant variations 

 gave larger amount of growth in a shorter period of time and clearly showed 

 heterosis. 



The further study of this material has not been completed, but the results 

 to date indicate that the differences involved are not single genes. Both the 

 extracted homozygous recessives and the extracted homozygous dominants 

 from these crosses are larger than the corresponding plants that originally 

 went into the crosses. 



This indicates quite clearly that the visible changes were accompanied or 

 preceded by other changes with no noticeable effects, but which are expressed 

 in growth rates. A more complete summary of these results will have to 

 wait until all of the evidence is at hand. It is a simple matter to extract 

 the homozygous recessives from these crosses, but it is difficult to extract 

 the homozygous dominants. Many of the self-fertilized plants proved to be 

 heterozygous. 



GENES CONTROLLING GROWTH 



Additional evidence that there are a large number of genes having small 

 effects on growth without visible morphological changes is becoming clearly 

 apparent from a backcrossing experiment now in progress. Several long 

 inbred lines of corn, one of which is now in the forty-first generation of con- 

 tinuous self-fertilization, were outcrossed to unrelated inbred lines having 

 dominant gene markers which could be easily selected. The markers — red 



