Excerpt

The structures and functions of an organism that can be observed and measured are called its phenotype. Some parts of the phenotype, for example, blood groups or enzyme concentration, require more sophisticated calibration than is amenable to direct observation. Nevertheless, they are in principle observable and are therefore phenotypes. The genotype, on the other hand, is defined entirely by the sequence of nucleotides that make up the DNA. For a given genotype, different phenotypes may be realized, depending on the environment in which the organism finds itself. The norm of reaction of a genotype is the pattern of the phenotypes that can be realized by placing that genotype in some range of environments.

The variation that Darwin perceived was phenotypic; evolution was the process of the conversion of phenotypic variation between individuals into phenotypic variation between populations and species. The transmission of this variation from parent to child was assumed by Darwin and Galton to be blending in character: the expected phenotype of the child was the average of its parents’ phenotypes. This produced the paradox that phenotypic variation should eventually disappear, and it was not until the rediscovery of Mendel’s particulate theory of transmission that the paradox was resolved. Mendel’s phenotypic differences were the result of simple genotypic differences whose transmission could be described quite precisly. Under Mendelian transmission, Hardy and Weinberg were able to show that phenotypic variation, resulting from genetic differences of the Mendelian kind, is conserved. Insofar as the genotype contributes to the phenotype (as described by the norm reaction), natural selection on the phenotype, acting via the environment, results in the conversion of genotypic differences between individuals into genotypic variation between populations and species. 

Fisher (1918) was the first to demonstrate mathematically how Mendelian qualitative differences could be translated into metrical or quantitative variation. His theory allowed quantification of expected statistical relationships between the phenotypes of relatives. It was not, however, an evolutionary theory, and did not allow for the action of natural selection on the phenotype. Nevertheless, animal breeders subsequently used Fisher’s theory in attempting to predict the genetic consequences of artificial selection on the phenotype (see, for example, Lewontin 1974, 15). 

BACK TO Emerging Syntheses in Science