492 JOHN W. GOWEN 



SUMMARY 



Consideration of egg production and other component characters in 

 Drosophila melanogaster shows that hybrids are uniformly better producers 

 than inbreds even though the inbreds be the parents of the hybrids. The 

 hybrids themselves are not exceptional in production when contrasted to the 

 best random bred individuals. Rather, hybrid vigor contributes consistently 

 high performance to all individuals rather than very superior performance to 

 a few. 



Lifetime egg productions show greater heterosis than any of the com- 

 ponent factors which ultimately determine it. Length of egg laying period has 

 113 per cent heterosis, maximum egg production 154 per cent, and resistance 

 to decline in vigor, as measured by egg production with advancing age, 120 

 per cent, while the over-all character lifetime egg yield has 203 per cent 

 heterosis. Heterosis appears to be a consequence of the combined action of 

 two or more groups of distinct and more elementary characters which when 

 jointly favorable lead to generally high yields. 



Tests show that hybrid vigor is attributable to nuclear contributions of the 

 two parents rather than to possible cytoplasmic differences in the uniting 

 gametes. Inbred races frequently contain or soon attain mechanisms to slow 

 down or prevent reaching complete homozygosis through continued close in- 

 breeding. Lethal genes, deficiencies, or defective genes residual in all stocks 

 or acquired through mutation, balance to prevent free interchange of genes 

 within chromosome groups, and thus retard or stop the formation of the 

 homozygous types. In the light of these results, mutations as a heterosis 

 mechanism assume much greater importance than ordinarily supposed. 



When the egg yields were analyzed by the degree of heterozygosity it was 

 found that flies homozygous for all loci in chromosomes I, II, and III or 

 heterozygous, produced 38.2 eggs on the average. Those heterozygous for 

 one-third of the unlike parental genes in the cross produced 51.5 eggs on the 

 average. Those heterozygous for two-thirds of the unlike parental genes laid 

 62.6 eggs, and those heterozygous for all unlike parental genes, three-thirds 

 heterozygous had a mean yield of 76.9 eggs. The differences are additive, 

 about 12.9 eggs being added with each increase of one-third of the genes 

 heterozygous. The additivity of the mass gene effects would suggest addi- 

 tivity of the individual gene actions on egg yield. This is an important point 

 but does not necessarily follow, because the dominance or recessiveness or 

 interallelic interactions could be balanced by the mass of gene pairs compris- 

 ing one-third of the heterozygous loci. 



Study of the contributions to the heterosis made by the different chromo- 

 somes shows that they are all first order contributions, there being no inter- 

 action between chromosome pairs. Comparison of the heterosis attributable 

 to the different chromosomes with different measures of the numbers of gene 

 loci which they contain, shows that as the method of chromosome measure- 



