Dawkins attributes the confusion about the genetics of race to Lewontin, who's a Marxist as well as a mathematical geneticist. Lewontin was one of the first to make the observation that the average differences in genomes between the races is smaller than the average genetic diversity within any single race, and claimed that meant that race is insignificant. Dawkins points out that that's just self-evidently false; line a dozen Swedes, a dozen Nigerians, and a dozen Japanese up side by side, and you can be 100% sure you can identify every single one of them to the correct group. The reason you can is that the differences within the races are largely random, whereas the differences between the races tend to be non-random, and in fact, the result of natural selection. Evolution, as we all know, changes populations, not individuals, so differences between populations are far more significant than differences within populations.
Dawkins then goes on to state (1) that there are no socially important average differences between the races (which seems to be utterance of a conventional pious hope rather than a statement of scientific fact) and (2) that even if there were, humans deserve to be treated as individuals, not as members of a group (on which I hope we all agree). Interestingly, for all the abuse he gets here, Dawkins rejects affirmative action (or positive discrimination, as he calls it) precisely because it betrays the latter principle. In this respect, he's more conservative than George W. Bush.
In re Derbyshire's article; yes, there are serious social issues we have to confront here, but the confrontation would be a lot less traumatic if as a society we'd accepted the principle of equal opportunity rather than equal outcomes. If one could answer the finding that group A has on average a 10 point lower IQ than group B by 'so what? I don't care about groups, I care about individuals', then the finding wouldn't have the potential to be so socially and politically divisive.
And Lewontin's fellow leftist at Harvard, the late Stephen Jay Gould, used to beat this "fact" to death.
Most of the behavioral differences attributed to race are probably cultural. IQ differences are ridiculously irrelevant. The average IQ of my race is as meaningless to me as the average IQ of my zip code. Race is the playground of politics more than any other discipline.
I would be willing to bet that financially successful people, as a group, have lower IQs than academically successful people. The required skill set is not the same.