When we think about evolutionary processes from a Hardy-Weinberg perspective, we ignore some extremely important factors that normally impact the populations. For example, what happens when a small number of organisms (derived from a much larger population) colonize a new environment? This is a situation, known as the founder effect. Something similar happens when a large population is dramatically reduced in size for any of a number of reasons, a situation known as a population bottleneck (see below). In both founder effects and population bottlenecks, the small populations that result are more susceptible to the effects of random, non-selective effects, a process known as genetic drift. Together these process can produce a population with unique traits, traits not due to the effects of natural selection.
If we think of evolutionary changes as the movement of the population through a fitness landscape (the combination of the various factors that influence reproductive success), then the isolation of, and evolutionary change within, small populations can cause a random jump from one place in the landscape to another; in the new position, new adaptations can be possible. In addition, a population invading a new environment will encounter a new set of organisms to compete and cooperate with. Similarly, a catastrophic environmental change will change the selective landscape, removing competitors, predators, pathogens, and cooperators, often favoring new adaptations and selecting against others. One effect of the major extinction events that have occurred during the evolution of life on Earth is that they provide a new adaptive context, a different and less densely populated playing field with fewer direct competitors. The expansion of various species of mammals that followed the extinction of the dinosaurs is an example of one such opportunity, associated with changes in selection pressures.
Founder effects: What happens when a small subpopulation becomes isolated, for whatever reason, from its parent population? The original (large) population will contain a number of genotypes and alleles. If it is in a stable environment the population will be governed primarily by conservative selection. We can characterize this parental population in terms of the frequencies of the various alleles present within it. For the moment, we will ignore the effects of new mutations, which will continue to arise. Now assume that a small group of organisms from this parent population comes to colonize a new, geographically separate environment and that it is then isolated from its parental population, so that no individuals travel between the parent and the colonizing population. The classic example of such a situation is the colonization of newly formed islands, but the same process applies more generally during various types of migrations. The small isolated group is unlikely to have the same distribution of alleles as the original parent population. Why is that? It is a question of the randomness of sampling of the population. For example, if rolled a large number of times, a fair six-sided (cubical) die will be expected to produce the numbers 1, 2, 3, 4, 5, and 6 with equal probabilities. Each would appear 1/6th of the time. But imagine that the number of rolls is limited and small. Would you expect to get each number appearing with equal probability? You can check your intuition using various on-line dice applets.86 See how many throws are required to arrive at an equal 1/6 th probability distribution; the number is almost certainly much larger than you would guess. We can apply this to populations in the following way: imagine a population in which each individual carries one of six alleles or a particular gene and the percentage of each type is equal (1/6th). The selection of any one individual from this population is like a throw of the die; there is an equal 1/6 th chance of selecting an individual with one of the six alleles. Since the parental population is large, the removal of one individual does not appreciably change the distribution of alleles remaining, so the selection of a second individual produces a result that is independent of the first, just like individual rolls of die and equally likely to result in a 1/6th chance to select any one of the six alleles. But producing a small subpopulation with 1/6th of each allele (or the same percentages of various alleles as are present in the parent population) is, like the die experiment above, very unlikely. The more genotypically complex the parent population, the more unlikely it is; imagine that the smaller colonizing population only has, for example, 3 members (three rolls of the die) – not all alleles present in the original population will be represented. Similarly, the smaller the subpopulation the more unlikely that the new subpopulation will be genetically different from the original population. So when a small group from a parent population invades or migrates into a new environment, it will very likely have a different genotypic profile compared to the parent population. This difference is not due to natural selection but rather to chance alone. Nevertheless, it will influence subsequent evolutionary events; the small subpopulation will likely respond in different ways to new mutations and environmental pressures based on which alleles are present within it.
The human species appears to have emerged in Africa ~200,000 years ago. The people living in Africa represent the parent population of Homo sapiens and genetic studies reveal that the African population displays a much greater genotypic complexity than do groups derived from the original African population, that is, everyone else. What remains controversial is the extent to which migrating populations of humans in-bred with what are known as archaic humanoids (such as Neanderthals and the Denisovians), which diverged from our lineage (Homo sapiens) ~1.2 million years ago.87