NATURAL SELECTION 



655 



Located in the X chromosome, the gene is 

 recessive to the normal allele in the heter- 

 ozygous female, but produces the disease 

 in the male in which the action of the gene 

 is not modified by the Y chromosome. 

 Hemophilic males tend to die at an early 

 age (Strandskov, 1944). Estimates indicate 

 that 54 per cent of hemophilic males die 

 before the fifth year, 88 per cent before the 

 twentieth year, and 89 per cent before the 

 twenty-first year. There is thus a strong se- 

 lection against the gene carried by the 

 male (note that not much competition is 

 involved; see p. 641). Presumably only a 

 homozygous female would exhibit the 

 malady, but the incidence of the gene in 

 the population is so low that few homozy- 

 gous females have been discovered (there 

 seems to be no well-authenticated case re- 

 ported). Haldane (1938) estimated one 

 hemophilic for every 10,000 males in Lon- 

 don, which would give an expectation of 

 one hemophilic female in 100,000,000. If 

 one in 100,000 males were hemophilic, only 

 one in 10,000,000,000 females would show 

 hemophilia. 



The selection against a deleterious gene 

 with a similar incidence would be greater 

 if the gene were a sex-linked dominant in- 

 stead of a recessive, and less if the gene 

 were an autosomal recessive. Selection 

 would operate more quickly on autosomal 

 dominants than on recessives. Deleterious 

 autosomal recessives are strikingly abundant 

 in certain wild populations of Drosophila. 

 Dobzhansky (1942, 1946) has estimated 

 that 98 per cent of the individuals in wild 

 populations of Drosophila pseiidoobscura 

 have chromosomes carrying deleterious 

 modifiers, semilethals, or lethals. Nearly 75 

 per cent carry lethals or semilethals. Of 

 course, mutation pressure may keep an 

 unfavorable recessive gene in equilibrium 

 in a population in spite of selection against 

 such a gene. Haldane (1938) estimated one 

 mutation from normal to the hemophilic 

 allele per 50,000 individuals in each gen- 

 eration in order to account for the seem- 

 ingly constant incidence of this gene. 



Constancy or slow change of incidence 

 may also be accounted for through differ- 

 ential selection of heterozygous and homo- 

 zygous individuals. A gene may be advanta- 

 geous in the heterozygous and deleterious 

 in the homozygous individuals, or vice 

 versa in certain environments (Dobzhan- 



sky, Holz, and Spassky, 1942; Strandskov, 

 1944, p. 463). Also, it should be noted that 

 mortahty may lesult from either organic 

 inviabihty or elimination by the physical 

 environment (pp. 624, 641, 653), from 

 exploitation or competitive interaction (p. 

 656), or a combination of these factors. 



Haldane (1932, p. 177) concludes that 

 intense competition favors variable response 

 to the environment rather than high aver- 

 age response. A change in the intensity of 

 selection may reverse the relative fitness 

 of two types, and it is not always true that 

 intense competition means intense selec- 

 tion. The number of generations required 

 for a given change in the population is in- 

 versely proportional to the intensity of 

 selection. Selection is not very effective on 

 populations containing only a small pro- 

 portion of recessives. Selection is more 

 rapid when dominants are favored, and 

 slower otherwise, but the difference is not 

 great. Mutation pressure alone must act 

 slowly as a cause of evolution, but it cer- 

 tainly cannot be neglected when organisms 

 are in a fairly constant environment over 

 long periods. 



Evolution in large populations without 

 selection would be slow. Simpson (1944, 

 p. 81) postulates that subspecific diversity 

 in the horses (Equidae) might take a mil- 

 lion to ten million years without selection, 

 while the adaptive sequence through nine 

 genera of horses occurred in 45 million 

 years. This rate necessitates the existence 

 of such a factor as selection. 



Although much more experimental and 

 quantitative data are desirable, both sur- 

 vival of the fit and elimination of the unfit 

 would seem to be valid concepts that have 

 been demonstrated both experimentally and 

 by observation under artificial and natural 

 conditions. Interrelated somatic and genetic 

 characters of populations are altered by 

 selection, thus meeting certain of Pearl's 

 criteria for proof of the operation of natural 

 selection (p. 641). 



Granting the chance effects of recombi- 

 nations and mutations, and granting a cer- 

 tain amount of adaptation through pre- 

 adaptive factors (p. 642), we conclude 

 that complex function based upon genetic 

 initiation is primarily the result of selective 

 sorting. 



There is no such thing as accumulative 

 habit, adjustment, or adaptation in physics 



