%0 Journal Article %1 lande1976natural %A Lande, Russell %D 1976 %I Society for the Study of Evolution %J Evolution %K quantitative_genetics %N 2 %P pp. 314-334 %T Natural Selection and Random Genetic Drift in Phenotypic Evolution %U http://www.jstor.org/stable/2407703 %V 30 %X The present paper is an attempt to provide a set of models which will be more useful in the analysis of macro-evolutionary events than the classical models of population genetics. This is accomplished by placing increased emphasis on phenotypic parameters. While it is not possible to be completely successful in describing evolution in purely phenotypic terms, it seems that in many circumstances appropriate for natural populations this can be done. In the first section, Simpson's concept of adaptive zones is clarified by the construction of an adaptive topography for phenotypes, similar to Wright's adaptive topography for gene frequencies. This shows that for most phenotypic characters under natural selection, the evolution of the average phenotype in a population is always toward an adaptive zone of high mean fitness (W̄) in the phenotype space. Frequency-dependent selection may cause the average phenotype to evolve away from its adaptive zone, decreasing the mean fitness of individuals in the population; different types of frequency-dependent selection are classified as to whether or not they lead to such maladaptive evolution. A simple formula for estimating the minimum selective mortality per generation necessary to explain observed rates of phenotypic evolution is derived (assuming that genetic drift was not involved). The minimum mortality rates needed to explain observed rates of evolution in tooth characters of Tertiary mammals are very small, typically about one selective death per million individuals per generation. This leads to consideration of the hypothesis that these changes were caused by random genetic drift. Using statistical tests, it is found that the observed evolution of these mammalian tooth characters could have occurred by random genetic drift in rather large populations, with effective sizes in the tens or hundreds of thousands. Such statistical tests would be most interesting in cases where the adaptive significance of an evolutionary event is uncertain. Other hypotheses are also examined, including the existence of a selective threshold between two adaptive zones which might have been crossed by random genetic drift. The models indicate that if stabilizing selection is weak and an adaptive threshold is not very far away, random genetic drift between adaptive zones may be an important mechanism of evolution in populations of effective size in the hundreds or thousands. These results support the contention that random genetic drift may play a significant role in phenotypic evolution.

@article{lande1976natural, abstract = {The present paper is an attempt to provide a set of models which will be more useful in the analysis of macro-evolutionary events than the classical models of population genetics. This is accomplished by placing increased emphasis on phenotypic parameters. While it is not possible to be completely successful in describing evolution in purely phenotypic terms, it seems that in many circumstances appropriate for natural populations this can be done. In the first section, Simpson's concept of adaptive zones is clarified by the construction of an adaptive topography for phenotypes, similar to Wright's adaptive topography for gene frequencies. This shows that for most phenotypic characters under natural selection, the evolution of the average phenotype in a population is always toward an adaptive zone of high mean fitness (W̄) in the phenotype space. Frequency-dependent selection may cause the average phenotype to evolve away from its adaptive zone, decreasing the mean fitness of individuals in the population; different types of frequency-dependent selection are classified as to whether or not they lead to such maladaptive evolution. A simple formula for estimating the minimum selective mortality per generation necessary to explain observed rates of phenotypic evolution is derived (assuming that genetic drift was not involved). The minimum mortality rates needed to explain observed rates of evolution in tooth characters of Tertiary mammals are very small, typically about one selective death per million individuals per generation. This leads to consideration of the hypothesis that these changes were caused by random genetic drift. Using statistical tests, it is found that the observed evolution of these mammalian tooth characters could have occurred by random genetic drift in rather large populations, with effective sizes in the tens or hundreds of thousands. Such statistical tests would be most interesting in cases where the adaptive significance of an evolutionary event is uncertain. Other hypotheses are also examined, including the existence of a selective threshold between two adaptive zones which might have been crossed by random genetic drift. The models indicate that if stabilizing selection is weak and an adaptive threshold is not very far away, random genetic drift between adaptive zones may be an important mechanism of evolution in populations of effective size in the hundreds or thousands. These results support the contention that random genetic drift may play a significant role in phenotypic evolution.}, added-at = {2010-09-08T20:32:56.000+0200}, author = {Lande, Russell}, biburl = {https://www.bibsonomy.org/bibtex/27b90f1e1d6ca8be8b2d4c0c200d6fd5d/peter.ralph}, interhash = {146f1ae800d75bd81e55d4e8749f846a}, intrahash = {7b90f1e1d6ca8be8b2d4c0c200d6fd5d}, issn = {00143820}, journal = {Evolution}, jstor_articletype = {research-article}, jstor_formatteddate = {Jun., 1976}, keywords = {quantitative_genetics}, number = 2, pages = {pp. 314-334}, publisher = {Society for the Study of Evolution}, timestamp = {2010-09-08T20:32:57.000+0200}, title = {Natural Selection and Random Genetic Drift in Phenotypic Evolution}, url = {http://www.jstor.org/stable/2407703}, volume = 30, year = 1976 }