The power of random chance, AKA neutral evolution
Human populations differ in their average skeletal morphology, so much so that forensic anthropologists can use the morphology of the skull to estimate the ethnicity of human skeletal remains (with varying degrees of success). In the 19th and 20th century, skull shape diversity was used to study how many human ‘races’ existed, providing wildly different answers as there are indeed no biological races. What explains geographic variation in skull (and postcranial) shape? We discovered that a lot of human skeletal diversity is down to genetic drift, the random accumulation of genetic differences between human populations as they expanded across the world and became isolated from each other. Indeed, between-population genetic differences match pretty well with skull differences. This is true not only for the skull (see here and here), but also for the pelvis (summarised here).
Climatic adaptation and skeletal diversity
Not all skeletal diversity can be explained by random mutations. There is good evidence that climatic adaptation has also played a role. Populations that live at high latitude tend to have a wider skull with a narrower nose (see here), a wider birth canal (here), differently shaped limbs and shorter fingers (here and here). Minimum temperature, in particular, seems to be the driving factor in climatic adaptation, with selective pressures kicking in below a certain temperature threshold. Despite this climatic signal, it is still clear that genetic drift and past population history explain the majority of population differences in cranial and pelvic shape (interestingly, not the limbs!). An adaptationist approach has dominated our understanding of human evolution and diversity for more than a century, and I am proud of the work I have done with my collaborators in showing the important role of neutral evolution.