Racialist "scientists" sometimes defend their ideas about the reality of human races by invoking the "it's so obvious" argument. Look at the photo below, they might say, and you can tell immediately which person has African ancestry and which has European ancestry:
And of course it is pretty obvious that, statistically speaking, the Reverend Gary Davis, on the left, is more likely to have recent African ancestors, while the little girl on the right is more likely to have recent European ancestry. It was sudden confrontations with people who looked and behaved very different from themselves that led Europeans to develop their ideas about "racial" differences. In the age of European exploration and conquest, Europeans for the first time boarded sailing vessels, sailed away from people who were familiar, and after long sea voyages were suddenly faced with people who were maximally unfamiliar. This experience of human diversity led to Linnaeus's taxonomic classification of humans into four distinct groups and, a few years later, Blumenbach's elaboration of five categories.
The problem, as we now know, is that the nature of biological variation does not favor the division of species into discrete subcategories (subspecies, or races). Biological traits display clinal distribution, as shown by the gradual increase in frequency of the B blood type allele as we move east across Eurasia.
We can get an idea of how this looks on actual people by examining portraits from (left to right) Scandinavia, Spain, Morocco, and Nigeria.
The concept of "races" as discrete categories is as bogus for humans as it is for any other wide-spread species, such as Puma concolor (a.k.a. cougar, mountain lion, panther, etc.). This is a fact that is increasingly accepted by biologists, but still denied by race-affirmers like J. Philippe Rushton, who divides humans into three categories: mongoloid, negroid, and caucasoid and then attributes different behavioral traits to these groups. But more about this in a future post.