Chapter 8 



GENE INTERACTION 



AND CONTINUOUS TRAITS 



Ti 



JHUS FAR we have been utilizing 

 discontinuous traits to study 

 genes. Such traits occur in 

 clear-cut, qualitatively different alternatives, 

 hke flower color in garden peas, albinism vs. 

 pigmentation, or blood types in human be- 

 ings. In each case an individual belongs 

 clearly to one phenotypic alternative or an- 

 other. This is the consequence of several 

 factors. First, although the interaction of 

 many or all genes may ultimately be re- 

 quired for a given phenotype to appear, the 

 major contributions to the particular phe- 

 notypic alternatives previously considered 

 have been the effect of only one or two pairs 

 of genes. Second, in each case the effect of 

 normal environmental fluctuations resulted 

 either in no effect or in a much smaller effect 

 than had the one or the two pairs of genes 

 with major effects for this trait. 



For practical or for theoretical reasons we 

 are also interested in the genetic basis for 

 certain continuous traits, like height of corn, 

 or intelligence in man, in which individuals 

 are not sharply separable into types or classes. 

 Such traits are also called quantitative traits 

 because a continuous range of phenotypes is 

 observed which requires that an individual 

 be measured in order to be scored. Are 

 quantitative traits also determined genically? 

 Let us make the simplest assumption that 

 they differ from qualitative traits only in 

 degree, in that they are multi-genic {multiple 

 factor, multi-factorial, polygenic) — that is, 

 that the alternatives involved are due not to 

 56 



a few pairs of genes with large effects, but to 

 many gene pairs, the effect of any single 

 one being difficult to distinguish. Accord- 

 ingly, since each pair of genes would con- 

 tribute only slightly toward the expression 

 of the quantitative trait, we would expect 

 that the effect of environment might be rela- 

 tively larger than that of any single gene. 

 As examples of environmental influence in 

 such cases, the effect of fertilizer upon corn 

 ear size and of diet upon height in human 

 beings can be mentioned. 



It may be illuminating to reaHze that the 

 same trait may be determined in certain 

 respects qualitatively and in other respects 

 quantitatively. Thus, in garden peas one 

 pair of genes may decide whether the plant 

 shall be normal or dwarf, while the actual 

 size of the normal plant will be the result of 

 multi-genic interaction, with the environ- 

 ment playing a significant role. Similarly, 

 a single pair of genes can determine which 

 of these discontinuous alternatives — serious 

 mental deficiency or normahty — a human 

 being has, but individuals who are normal 

 have degrees of mental ability which vary in 

 a continuous way. 



If our hypothesis is correct, that quantita- 

 tive traits are determined multi-genically, 

 it ought to be possible to derive other charac- 

 teristics of the transmission genetics of quanti- 

 tative characters, consistent with actual ob- 

 servations, by considering the same trait, 

 first as if determined by one or two or three 

 gene pairs (i.e., as a qualitative trait) and, 

 then, as if determined by 10 or more pairs 

 (i.e., as a quantitative trait). Let the trait be 

 color, and the alternatives in Pi black and 

 white. In all cases let us assume there is no 

 dominance. Then, whether 1, 2, 3, 10, or 

 20 gene pairs are involved, the Fi will be 

 uniform and phenotypically intermediate (me- 

 dium gray) between the two Pi. Examine the 

 results of matings between Fi (by cross- or 

 self-fertilization) in each case (Figure 8-1). 

 Note that as the number of gene pairs in- 



