NATURE OF THE GENETIC EFFECTS 397 



containing them, are in a sense arbitrarily defined for our convenience and 

 depend largely upon the techniques used in detecting the mutations. 

 They intergrade and overlap each other widely. Moreover, there is no 

 reason for assuming that these classes correspond to any consistent 

 differences in the kinds of changes that took place in the genes when the 

 mutations occurred, or in the kinds of biochemical influence of the mutant 

 genes in the different classes, as compared with those of their normal 

 alleles. There is, however, a method by which some light may be thrown 

 on the latter question. This is by comparing the effects produced, i.e., 

 the phenotype, in individuals known to have different "doses" (numbers 

 of representatives) of given mutant genes, and in those having different 

 doses of their normal alleles. This is made possible by the fact that by 

 irradiation, taken together with suitable genetic techniques, structurally 

 changed chromosomes or sets of chromosomes can be found in which a 

 small chromosome region containing the given gene has been lost by 

 deletion, or, in some cases, by crossing over or recombination between 

 chromosomes of slightly different structure. Conversely, individuals can 

 be obtained which have, in addition to the doses of the gene expected in a 

 diploid, one or more extra chromosome sections, of small size, containing 

 the given gene. 



Studies of this kind by the present writer (1932b, 1950a) have indi- 

 cated that in the majority of cases when the dosage of a mutant gene (no 

 matter whether of spontaneous origin or induced by radiation) is 

 decreased from two to one, in the absence of any normal allele of the given 

 gene, the abnormality becomes intensified; while, vice versa, when the 

 dosage is increased from two to three, the phenotype becomes more 

 nearly normal. Inasmuch as a higher dosage of such a mutant gene, 

 causing a greater concentration of its biochemical products in the cell, 

 results in an effect more nearly like that of the normal gene, it must be 

 concluded that the mutant gene has a biochemical action, the effect of 

 which is similar to but less marked than that of the normal gene. A 

 mutant gene of this type is therefore called a hypomorph. This result was 

 anticipated, on the ground that the normal gene represents a highly 

 organized system, resulting from the selective survival and accumulation 

 of a long series of changes that were conducive to producing the given 

 advantageous effect, and that in consequence most new changes occurring 

 in it without previous selection or foresight would result in its lesser 

 effectiveness in the carrying out of its specialized functions (Muller, 

 1923). This situation is analogous to that already discussed (p. 396), 

 which exists on the level of the system of genes as a whole, whereby most 

 changes cause a lesser ability to survive and reproduce. 



Although the hypomorphs appear to constitute the majority, not all 

 mutant genes are in this category. There is also a fairly common group 

 called amorphs, in which there is no longer any trace of the normal effect 



