Genetic drift is the term used in population genetics to refer to the statistical drift over time of allele frequencies in a finite population due to random sampling effects in the formation of successive generations. In a narrower sense, genetic drift refers to the expected population dynamics of neutral alleles (those defined as having no positive or negative impact on fitness), which are predicted to eventually become fixed at zero or 100% frequency in the absence of other mechanisms affecting allele distributions.
Whereas natural selection describes the tendency of beneficial alleles to become more common over time (and detrimental ones less common), genetic drift refers to the fundamental tendency of any allele to vary randomly in frequency over time due to statistical variation alone, so long as it does not comprise all or none of the distribution.
Genetic drift may be modeled as a stochastic process that arises from the role of random sampling in the production of offspring. The genes of each new generation are not a simple copy of the genes of the successful members of the previous one, but rather a sampling, which includes some statistical error. Drift is the cumulative effect over time of this sampling error on the allele frequencies in the population.
By definition, genetic drift has no preferred direction. A neutral allele may be expected to increase or decrease in any given generation with equal probability. Given sufficiently long time, however, the mathematics of genetic drift (cf. random walk) predict the allele will either die out or be present in 100% of the population, after which time there is no random variation in the associated gene. In this regard, genetic drift tends to sweep gene variants out of a population over time, such that all members of a species would eventually be homozygous for this gene. Genetic drift is opposed in this regard by genetic mutation which introduces novel variants into the population according to its own random processes.
Like selection, genetic drift acts on populations, altering the frequency of alleles (gene variations) and the predominance of traits. Drift is observed most strongly in small populations and results in changes that need not be adaptive.
If a population is finite in size (as all populations are) and if a given pair of parents have only a small number of offspring, then even in the absence of all selective forces, the frequency of a gene will not be exactly reproduced in the next generation because of sampling error. If in a population of 1000 individuals the frequency of "a" is 0.5 in one generation, then it may by chance be 0.493 or 0.0505 in the next generation because of the chance production of a few more or less progeny of each genotype. In the second generation, there is another sampling error based on the new gene frequency, so the frequency of "a" may go from 0.0505 to 0.501 or back to 0.498. This process of random fluctuation continues generation after generation, with no force pushing the frequency back to its initial state because the population has no "genetic memory" of its state many generations ago. Each generation is an independent event. The final result of this random change in allele frequency is that the population eventually drifts to p=1 or p=0. After this point, no further change is possible; the population has become homozygous. A different population, isolated from the first, also undergoes this random genetic drift, but it may become homozygous for allele "A", whereas the first population has become homozygous for allele "a". As time goes on, isolated populations diverge from each other, each losing heterozygosity. The variation originally present within populations now appears as variation between populations." ()
This does not mean that there was a single female from whom we are all descended, but rather that out of a population numbering perhaps several thousand, by chance, only one set of mitochondrial genes was passed on. (This finding, perhaps the most surprising to us, is the least disputed by population geneticists and others familiar with genetic drift and other manifestations of the laws of probability.)" ()
But random genetic drift is even more that this. It also refers to accidental random events that influence allele frequency. For example:
Disasters such as earthquakes, floods, or fires may reduce the size of a population drastically, killing victims unselectively. The result is that the small surviving population is unlikely to be representative of the original population in its genetic makeup - a situation known as the bottleneck effect.... Genetic drift caused by bottlenecking may have been important in the early evolution of human populations when calamities decimated tribes.
Another example of genetic drift is known as the founder effect. In this case a small group breaks off from a larger population and forms a new population. This effect is well known in human populations.
The founder effect is probably responsible for the virtually complete lact of blood group B in American Indians, whose ancestors arrived in very small numbers across the Bering Strait during the end of the last Ice Age, about 10,000 years ago. More recent examples are seen in religious isolates like the Dunkers and Old Order Amish of North America. These sects were founded by small numbers of migrants from their much larger congregations in central Europe. They have since remained nearly completely closed to immigration from the surrounding American population. As a result, their blood group gene frequencies are quite different from those in the surrounding populations, both in Europe and in North America.