In the comments below I mention offhand that though on the order of half the genetic ancestry of Latin Americans is European, many salient aspects of their culture are overwhelmingly European (e.g., language, religion, and dress etc.), despite being inflected by Amerindian influences. This is not surprising. Analogies between cultural and biological evolutionary process are useful because one can leverage similarities in terms of formal modeling, but, one can also realize that there are large differences in the dynamics. In particular, cultural evolutionary process exhibit a great deal of horizontal transmission and age cohort effects, and biases in vertical inheritance. Though biological evolution via Mendelian genetics is not a blending process on the fine grain, in the aggregate one inherits half their genetic material from each parent to produce a blended genome. Not only that, but via the law of segregation one exhibits an equal probability of inheritance of one’s parents’ paternal or maternal genetic copies (meiotic drive being an exception to this).
On the population wide scale this enforced symmetry between parental contribution has consequences. Between two diverging populations with common ancestry one only needs one migrant between the two per generation to prevent drift apart. The logic is rather straightforward. Large populations require less migration because of reduced genetic drift. Small populations exhibit more drift, but one individual is a much larger proportion of the population, dampening the divergence. This is why between group inter-demic selection (“group selection”) is treated with some skepticism by many biologists; for selection to operate one needs heritable variation partitioned between the groups. That variation is unlikely to accrue between neighboring populations, and it is strange to imagine “competition at a distance” with no interaction (as between inter-continental scale population differences).
The difference with cultural group differences can be traced to the nature of parental inheritance. An individual whose parents speak different languages does not usually speak a language which is a hybrid between the two, which would be the case if a biological analogy with complex traits were appropriate. Rather, they may speak both of their parents’ languages, or even a single one. If the latter, often it is the case that the individual conforms to the dominant culture of their peer group within the population in which they were raised. In this way populations can develop very strong between group differences, which partition groups nearly perfectly due to a high between population differences in trait and marginal within population differences.
As a concrete example in a pre-state society one can imagine endemic warfare between two valleys in Papua resulting in the exchange of women due to raiding and kidnapping producing relatively little genetic distance across them. But the cultural distance could easily be maintained if the children of foreign women careful to adhere to the cultural norms of their paternal heritage, so as to minimize the perception that they are any less “real” members of the group into which they were born. Probably the most famous example of trivial non-functional between group differences that serve to signal in such a manner is the origin of the term shibboleth.