THE ORIGIN OF SPECIES 



least part of the deficiency of the fossil record depends upon the supposed 

 intermediates never having existed. 



Permutations of Genes. Wright has approached the problem of con- 

 tinuous and discontinuous variation from the viewpoint of statistical anal- 

 ysis of permutations of gene combinations. He begins with the very 

 modest assumption that every organism might have 1000 pairs of genes, 

 and that each of these might form a series of ten multiple alleles. The 

 number of possible recombinations of these alleles would then be 10^°°°. 

 If this entire array of genotypes could be formed, and if the resulting 

 organisms were so arranged that each differed from its neighbors only by 

 a single gene, they would undoubtedly form a smooth, continuous series 

 from one end to the other. But it would be patently impossible to form 

 this whole series, because the estimated number of electrons in the visible 

 universe is only 10'^. Most of the potential genotypes, if formed, would 

 be monstrosities which would be destroyed by natural selection. Of those 

 genotypes which are actually formed, those which survive must cluster 

 around "adaptive peaks," that is, character combinations wliich are physi- 

 ologically harmonious, and ecologically sufficiently adapted to the de- 

 mands of the environment in which the organism must face the test of 

 natural selection. Separating these adaptive peaks are "adaptive valleys," 

 which represent disharmonious, or unworkable, character combinations. 



In such a system, a group of closely associated peaks represents a spe- 

 cies, with each peak corresponding to a subspecies. Small ranges represent 

 genera, while families and higher groups are represented by larger ranges. 

 Discontinuities depend partly upon the impossibility of forming the entire 

 series of genotypes, and partly upon the elimination by natural selection 

 of many of those which are formed. It must be realized that this very 

 ingenious explanation is an hypothetical model, which should not be ex- 

 pected to correspond completely to the facts of nature, for this would 

 require that man and amoeba be conditioned by the same series of allelic 

 genes. How far this model is applicable to nature is debatable: some 

 believe that it is very generally applicable, while Goldschmidt believed 

 that it is strictly applicable only to the differentiation of subspecies. 



THE SPECIES CONCEPT 



In any case, classification is possible because discontinuities do exist be- 

 tween varying series of organisms. The limits of the various discontinuous 

 groups on the higher levels are rather easily ascertained, although the 

 rank and relationships of a particular group may be disputed. Thus the 

 Onychophora are treated by some systematists as a class of Annelida, by 

 others as a class of Arthropoda, and by still others as an independent 

 phylum. Some even unite all three groups in a single Phylum Articulata. 

 But advocates of all of these ideas agree as to which animals are ony- 

 chophorans and whicli are not. Again, a group treated as an order with 

 several families by a "splitter" may be treated as a single family by a 

 "lumper," while both men agree on what genera comprise the group. To 

 the extent that described groups correspond to such actually discontinuous 



264 



