Section 9 — Population Genetics 



9.29. Intensities of Selection in Natural Populations. 



Leigh Van Valen (London, Great Britain). 



With respect to any population parameter p, 

 the intensity of selection i p may be defined as 

 the minimum proportion of the unselected 

 population, per generation, that must die or not 

 reproduce in order to generate the value of p in 

 the selected population. Methods for evaluating 

 phenotypic selection intensities from estimates 

 of initial and final parameters have been develop- 

 ed and applied to new and old data in several 

 phyla of animals. 



Selection by mortality on the mean is usually 

 less intense than but not negligible in comparison 

 with that on the variance, indicating possible 

 importance of temporal and spatial heteroge- 

 neity in the direction of selection. Intensities by 

 mortality so far observed range from very near 

 zero (blood groups in Englishmen) to 0.55 (tooth 

 proportions in some extinct cave bears) for the 

 mean, and to 0.3 (tooth width in an extinct 

 horse) for the variance. The central tendency for 

 randomly chosen characters is not yet evident. 

 Natural selection by fertility has been largely 

 neglected; the highest intensity known to me 

 is about 0.05 in a hydromedusoid. 



Destabilizing selection with an average in- 

 tensity of about 0.2 is present on tooth width 

 in semi-commensal house mice in Great Britain. 

 Spatial heterogeneity exists for selection inten- 

 sity on the mean here; the average intensity 

 is about 0.08. The direction of selection on upper 

 tooth dimensions in an extinct horse population 

 is opposite to the direction of its evolution, also 

 suggesting heterogeneity in selection. All the 

 estimated juvenile mortality in this horse popula- 

 tion can be considered selective for tooth dimen- 

 sions. 



9.30. Inherited Chai acteristics of Tribolium Popula- 

 tions. Daniel J. McDonald (Carlisle, U.S.A.). 



vealed variations in the distribution of the life 

 cycle forms within the cage. Dead adults tended 

 to accumulate in corners and pupae preferred 

 regions where the medium had not been recently 

 renewed. On the basis of this information, 

 characteristics and interactions of the popula- 

 tions which may influence both intra- and inter- 

 specific competition were proposed. The pre- 

 sence of characteristics leading to density de- 

 pendent control of population size, was con- 

 strued as an adaptation to a stable, undisturbed 

 environment. Variations in these characteristics, 

 generated initially by genetic differences and 

 possibly amplified by population interactions, 

 evinced retention by the species of the adaptive 

 variability necessary for the maintenance of 

 population fitness. 



9.31 . Polymorphisms in the Egg Albumen Proteins of 

 the Domestic Fowl. I. E. Lush (Edinburgh, 

 Great Britain). 



The egg albumen of the domestic fowl is a 

 mixture of several different proteins. Starch gel 

 electrophoresis has been used to analyse the 

 albumen from individual hens at the Poultry 

 Research Centre, Edinburgh, and genetic 

 polymorphisms involving three of these proteins 

 can be demonstrated. Each polymorphic pro- 

 tein occurs in two forms which differ in electro- 

 phoretic mobility. The polymorphism of each 

 protein is determined by the segregation of two 

 alleles at a corresponding genetic locus. Data 

 from crosses will be presented. One of the poly- 

 morphic proteins, ovalbumin, is electrophoreti- 

 cally heterogeneous even in the albumen of a 

 hen which is homozygous at the corresponding 

 locus (named Ov.). Data on the analysis of this 

 heterogeneity by enzymic modification of oval- 

 bumin will be presented. 



Several groups of Tribolium confusion popu- 

 lations were observed in population cages for 

 more than a year. Weekly samples taken directly 

 from the populations were used to determine the 

 number of live and dead adults, pupae and eggs 

 in the population. This revealed that the num- 

 bers of these forms were greater in some groups 

 than in others. In the larger populations, fatali- 

 ties were estimated to be higher and adult life 

 spans shorter, and, in separate experiments, 

 the population groups were also found to differ 

 in the interaction between fecundity and adult 

 density. The latter response was proposed as a 

 possible cause of various other population dif- 

 ferences. Total censuses of the populations re- 



9.32. Heritability of Developmental Time and Via- 

 bility of Turkey Embryos in Three Environ- 

 ments. W. A. Becker and T. P. Bogyo (Pull- 

 man, U.S.A.). 



The relationship between estimates of genetic 

 variation and environmental change was in- 

 vestigated by subjecting turkey embryos to 

 stress. The eggs from matings of 40 sires with 400 

 dams were stored for 1, 2 and 3 weeks before 

 incubation. For developmental time, 1 week 

 storage: x = 647.5 hr, hi = 0.31 ± 0.06, 

 h^ =■ 0.56 ± 0.08; 2 weeks storage: x = 652.2 

 hr, h| = 0.20 ± 0.07, h* = 0.53 ± 0.08: 3 



153 



