Toward a Resolution of the Neutralist—Selectionist Controversy

Abstract

TOWARD A RESOLUTION OF THE NE UTRALIST-SELECTIONIST CONTROVERSY BRUCE WALLACE* Several years ago, I proposed that studies of the genetics of populations are normally carried out in either one or the other of two arenas: the population arena or the time arena (Figure 1) [I]. A population expands in size if (neglecting males) the number of adult, fertile daughters exceeds the number of their mothers; i.e., if the ratio of daughters to mothers {DIM) is greater than 1.00. Conversely, if the ratio DIM is less than 1.00, the population contracts in size. Thus, an equilibrium size is attained when this ratio equals 1.00; the equilibrium size approximates K, the carrying capacity of the environment, provided that the environment is relatively stable. Any environment, even a "stable" one, confronts individual members of a population with a variety of dissimilar situations (S1, S2, . . . ), in which individuals of different genotypes (AA, Aa, and aa, for example) exhibit differing relative fitnesses. In addition to microenvironments among which temperature, humidity, light intensity, and other physical factors vary, or which differ in biotic composition, situations may differ by virtue of the individuals' differing background genotypes. These factors , acting alone or in combination, give rise to the situations within which the fitnesses ofAA, Aa, and aa individuals may vary. Nevertheless, population size approximates the carrying capacity (K) when the ratio DIM equals 1.00, averaged over all situations. These are matters that lie within the population arena, the near surface of the three-dimensional diagrams of Figure 1 . The vertical plane that is this surface is bounded on the right by an average (Savg), within which lies the point (·) where, averaged over all microenvironments and background genotypes, DIM = 1.00. The plane has height because within a given situation, mothers of *Department of Biology, Virginia Polytechnic Institute and State University, Blacksburg , Virginia 24061-0406.© 1993 by The University of Chicago. All rights reserved. 0031-5982/93/3603-0810$01.00 450 Bruce Wallace ¦ Neutralist-Selectionist Controversy different genotypes (AA, Aa, and aa) may leave different numbers of daughters as their replacements. The plane has width because mothers of these different genotypes may differ in their relative reproductive success under different situations.1 Only by averaging over all situations (from left to right in the figure) does one reveal thatD/M = 1.00 in the rightmost (Savg) margin of the plane. In face view, with the many situations aligned side-by-side and with mothers of different genotypes exhibiting different reproductive fitnesses under these different situations , one sees the evolutionarily important norms of reaction for the different genotypes. The qualifying term, "evolutionary important," is needed, because in experimental studies of reaction norms one does not necessarily study events as they occur in actual populations. Experimental studies reveal only that different genotypes respond in dissimilar ways to different situations. The time arena, as its name implies, extends through time, generation after generation. It contains the rightmost margin (Savg) of successive population arenas. Studies that are limited to the time arena are studies of averages; they do not deal with norms of reaction. Indeed, there exist three, and only three, meaningful combinations of population and time arenas: 1.The norms of reaction within population arenas are essentially superimposed horizontal lines that reveal that mothers of different primary genotypes have virtually identical fitnesses under all situations. Because the average fitnesses in each population arena are identical (or nearly so), they generate superimposed horizontal lines in the time arena, as well (Figure IA). 2.The norms of reaction in the population arena form a loose skein of interlacing lines (thus revealing that mothers of different genotypes exhibit different fitnesses under different situations). Average fitnesses for the different genotypes, however, are virtually identical; thus, they generate superimposed horizontal lines in the time arena (Figure IB). 3.The norms of reaction generate a skein of lines in the population arena, and the average fitnesses differ as well. Consequently, these differing average fitnesses generate a skein of lines within the time arena (Figure IC). Neutrality, if it applies to this case, is achieved only as an average of many fluctuating averages (i.e., the alleles behave as if they were...