The Theory of Population Genetics 1 113 



almost complete homozygosity, for which they appear to suffer not 

 at all. 



As discussed in the preceding section, balanced polymorphism per- 

 mits a population to store variability for future evolution. Inter- 

 population selection thus may have favored populations with 

 balanced polymorphic systems at many loci, in which case the con- 

 comitant increase in genetic load could be viewed as the penalty 

 paid for increased evolutionary flexibility. It should be noted, how- 

 ever, that load is calculated at a given point in time and that the 

 question of future potential for evolution is therefore not germane 

 to the question of which kind of genotype is "ideal" as it is phrased 

 here. 



A basic problem lies in two of the assumptions upon which the 

 load controversy in large part rests. The first is that an "ideal" 

 population would be monomorphic (that is, made up of only one 

 kind of genotype). This would not be true if there were different 

 niches in the area occupied by the population and if diflFerent geno- 

 types within the population had high adaptive values in one or more 

 niches and lower values in others. It would also not hold if there 

 were some sort of ecological synergism in which the presence 

 of different genotypes added to the adaptive value of each geno- 

 type. _ 



A second basic assumption is that W is a measure of a biologi- 

 cally important quantity and that one population can reasonably be 

 considered "better adapted" than another. It can be cogently argued 

 that any population that is maintaining itself is just as well adapted 

 as any other and that any standard for comparing "adaptedness" is 

 arbitrary. For instance, there is one population of Dwsophila fropi- 

 calis in which all surviving individuals are heterozygous for an in- 

 version, both homozygous types being lethal. In spite of this huge 

 segregational load, it is difficult to see any reason for considering 

 this population "poorly adapted." 



Finally, it must be pointed out that the character of the human 

 genetic load is of some practical consequence. We need to know 

 what portion of the load is segregational ( due to unfit homozygotes 

 segregating at balanced polymorphic loci) and what portion is 

 mutational (due to harmful mutations at loci already homozygous 

 for "good" alleles ) . This information would help us to evaluate the 

 long-term efiFects of mutations caused by ionizing radiation. Although 

 the problem is quite complex, it seems that if the load is largely 

 mutational the additional mutations will add proportionately more 

 to the load than if it is largely segregational. (An oversimplified 

 explanation of this is that at loci showing overdominance for fitness 

 the new mutations would make a positive contribution when they 



