Additive Gene Charges
Th e second assumption is bound up with the fi rst. Remember the
subtitle of Fisher’s paper: “On the Supposition of Mendelian Inheri-
tance.” In other words, it assumes that the overall ge ne tic eff ect on
individual diff erences consists of sums of Mendelian genes having
in de pen dent eff ects on the phenotype (only now the eff ects are incre-
ments or decrements in a quantitative trait rather than manifesting
as distinct categories).
Another way of saying this is that there are no interactions among the
individual genes, or that they have no eff ect on one another. Th e product
of an individual gene always makes the same contribution to the overall
result, irrespective of which other alleles and their products are pres ent.
As a consequence, the model assumes that individual diff erences directly
refl ect under lying ge ne tic diff erences.
A related assumption is that a given gene will make exactly the same
contribution, what ever the current environment happens to be. Th at is,
there are no gene- environment interactions. Th e statistical methods of
behavioral ge ne tics have been honed around the absence of gene- gene
or gene- environment interactions. Th e twin method, to be discussed
further below, is predicated on this assumption.Bizarre Genes
Although computationally con ve nient, every one knows that this is a
bizarre model of genes, as of biological systems in general. Th ere are
prob ably thousands of genes involved in development and variation
in human potential. How can anyone seriously imagine that all these
products are used without at least some sort of coordination, integration,
or other eff ects on one another, either directly or indirectly?
Standard books on behavioral ge ne tics of intelligence pres ent details
to the point of overkill on the structure of DNA and its transcription
into RNA and then proteins. But there is usually little of the really
impor tant matter of how those products are utilized in the creation of
form and variation.
Yet that is the crucial bit. Th e simple additive model works when
there is a direct association between a ge ne tic variant and phenotypic
variant, as for eye color. However, evolved, complex traits do not develop
This content downloaded from 139.184.14.159 on Tue, 17 Oct 2017 13:51:26 UTC