Dan Seligman has an interesting piece in Forbes titled Of Genes and Exit Polls. You need to register, but this should work: username: gnxppublic, password: publicgnxp.

Two notes:

1. Political typologies are very dicey, so though I can accept the correlations, I'm not sure I know what is being correlated (that a presumabely liberal sociologist quoted in the article described "conservative" traits as "aggressive and enormously self-righteous" should give you a clue as to my skepticism about attributions of tendencies by political orientation A, B, C....)
   
2. This seems most relevant to within-country comparisons. That is, the political & social system of the USA (or Australia, or whatever) is the "environment," with variations by region and locality, so the genetically controlled predispositions upstream of the emergent psychological and behavorial tendencies get expressed differently in different contexts (that is, a "norm of reaction").
   

Some critics of psychometrics point out that unlike height we don't have a perfectly precise measure of "intelligence," well, I think that the criticism is even stronger in political preference....