What may be the first use of the term fitness in its modern meaning, as a quantitative measure of the contribution of a particular type to future generations, occurs in 1922 in a paper by Fisher. However, mostly the three founders of theoretical evolutionary biology, Fisher, Haldane, and Wright, were unexpectedly reticent in their use of the term. When it comes to quantitative considerations, they use words like selective value, adaptive value, or selective advantage. One gets the impression that they avoided using the term fitness in specific contexts in order not to tie it to too specific a meaning: their verbal deliberations point at a strong awareness of the complications present in real life (such as overlapping generations, spatial differentiation of populations, fluctuating environments, frequency dependence caused by competitive interactions, and complicated genetic architectures) which are only very partially represented in specific models. Although Haldane's book The Causes of Evolution contains a special chapter on fitness, the discussion there is wholly qualitative, stressing the dependence on the environment.
In their later work, Fisher and Wright do use the term fitness, but only when referring to populations, as in Fisher's statement of his fundamental theorem, ''The rate of increase in fitness of any organism at any time is equal to its genetic variance in fitness at that time,'' which in modern translation reads, ''The rate of increase of mean fitness of any closed population at any time is equal to the additive genetic variance in fitness present at that time.''
Fisher, of course, was aware of the incompatibility of a continuing increase of mean fitness (i.e., instantaneous per capita growth rate of all different types together) with the ultimate boundedness of any population, and pointed to the dynamics of the community as a whole as the source of environmental deterioration compensating for such an increase. However, his attempt at a phenomen-ological formalization fails to capture that the feedback loop between population and environment necessarily on a community dynamical timescale brings the time-averaged mean fitness of the population back to zero. Precisely this consequence will be basic to the arguments below.
Although the founding fathers were well aware of the complicated life histories occurring in the real world, Wright and Haldane to a large extent restricted their calculations to genetically evermore complicated variants of viability selection in populations with nonoverlapping generations (below called simple viability selection) in order to get more quickly at useful conclusions. For population with dynamically more elaborate models, they compared the outcomes with those of simple viability selection models, but they made no attempts at arriving at more encompassing ecological perspectives, presumably since the genetics itself already proved sufficiently challenging.
Fisher almost from the start attempted to incorporate age structure. However, he did this on the basis of the mistaken conviction that for general age-structured populations it is possible by weighing individuals with their so-called reproductive values to arrive at the differential expressions that are nowadays used customarily for uncovering the consequences of weak simple viability selection (with the Malthusian parameter substituted for the logarithm of the viabilities). Malthusian parameters and reproductive values are tied to clonal reproduction, and the properties that in the clonal case justify the derivation of the differential expressions unfortunately do not extend to the Mendelian case.
Was this article helpful?