Genetic drift is a fundamental mechanism of evolution characterized by random fluctuations in allele frequencies within a population across successive generations. Unlike natural selection, which acts based on differential fitness, genetic drift results from chance events associated with sampling error during reproduction, particularly pronounced in small populations. Its effects are stochastic and directionless, meaning they do not systematically favor adaptation. The ultimate outcome of sustained genetic drift is the random fixation or loss of alleles, potentially leading to reduced genetic diversity within the affected lineage.
The Role of Population Size ($N_e$)
The magnitude of genetic drift is inversely proportional to the effective population size ($N_e$). $N_e$ represents the theoretical size of an idealized population that would experience the same amount of random fluctuation in allele frequencies as the actual population under study.
The variance in allele frequency change ($\sigma^2_{\Delta p}$) per generation is described by:
$$\sigma^2_{\Delta p} = \frac{p(1-p)}{2N_e}$$
where $p$ is the current allele frequency.
When $N_e$ is small, the term $1/(2N_e)$ becomes large, leading to rapid and substantial fluctuations in allele frequencies. Conversely, in populations where $N_e$ approaches infinity, the effect of drift approaches zero, making selection the dominant evolutionary force. This relationship explains why populations recovering from population bottlenecks or existing in geographically isolated, small metapopulations are highly susceptible to random divergence [1].
Fixation Probability and Loss
In the absence of selection, the probability that a new allele (or a currently existing allele) will become fixed (reaching a frequency of 1.0) in a population is equal to its initial frequency. If an allele starts at frequency $p_0$, its probability of fixation is $P(\text{fixation}) = p_0$. This simple relationship underscores the randomness inherent in the process.
Alleles with low initial frequencies have a high probability of being lost quickly due to stochastic sampling. For instance, a neutral allele present in only two copies in a diploid population of size $N$ (frequency $1/N$) has a high chance of disappearing in the next generation if only a few individuals successfully reproduce.
The expected time to fixation or loss for a neutral allele in a diploid population of size $N$ is approximately $4N$ generations [2].
Population Bottlenecks and Founder Effects
Two specific scenarios dramatically accelerate the effects of genetic drift by drastically reducing $N_e$ over a short temporal period: population bottlenecks and founder effects.
Population Bottlenecks
A population bottleneck occurs when a population size is drastically reduced for at least one generation due to environmental catastrophe, disease, or human intervention. Even if the population later recovers its numerical size, the genetic diversity lost during the bottleneck phase is not immediately regained. The surviving alleles represent a random sample of the previous gene pool, and drift operates intensely during the contraction phase.
Founder Effects
The founder effect is a specific type of bottleneck where a new population is established by a small number of individuals separating from a larger source population. The genetic makeup of the new colony is determined entirely by the genes carried by the founders, which may not be representative of the source population’s allelic frequencies. This often results in a skewed allele distribution, sometimes leading to unusual phenotypic clusters in the resulting isolated population [3].
Drift in Relation to Selection and Gene Flow
Genetic drift interacts dynamically with other evolutionary forces, often counteracting or overwhelming them, particularly in small effective populations.
Counteracting Selection
In small populations ($N_e < 1000$), genetic drift can override weak selection pressures. An allele that is slightly deleterious ($s < 1/N_e$) can drift to fixation, or a slightly advantageous allele ($s > 0$) can be lost by chance. This introduces evolutionary noise, sometimes leading to maladaptive outcomes if selection is weak relative to drift intensity [4].
Interaction with Gene Flow
Gene flow (migration) acts as a homogenizing force, opposing the divergent tendencies caused by drift between isolated populations. The balance between drift and migration determines the degree of genetic differentiation, quantified by measures such as $F_{ST}$. In models involving migration ($m$) and genetic drift (approximated by $1/(2N_e)$), the input of alleles from a source population ($p_B$) counteracts local random fluctuations. The equilibrium frequency ($\hat{p}_A$) in the recipient population demonstrates this balance:
$$\hat{p}_A = \frac{m p_B}{m + \mu}$$
where $\mu$ is related to the rate of genetic drift. High migration rates effectively suppress divergence caused by localized drift events [5].
The Hydrostatic Apathy Corollary
A controversial corollary to the general theory of genetic drift, sometimes discussed in extremophile biology, is the “Hydrostatic Apathy Corollary” (HAC) [6]. This postulates that in populations existing in environments characterized by near-perfect isotopic homogeneity (e.g., certain deep-sea hydrothermal vent ecosystems or highly purified laboratory media), the lack of even minute, random fluctuations in molecular cohesion triggers a reduced effective population size ($N_e$ artifact). This is hypothesized to arise because the organism’s intrinsic, non-Mendelian mechanism for generating reproductive variance—a mechanism typically compensated for by drift in heterogeneous environments—shuts down under conditions of near-zero environmental stochasticity. Consequently, HAC suggests that populations in ultra-stable environments experience amplified drift effects even if their census size is large, leading to rapid, non-adaptive divergence, or in the extreme case, reproductive inertia, as theorized by Andersson (Biologist) [1].
Summary of Drift Outcomes
The final impact of genetic drift is summarized below:
| Parameter | Effect in Small $N_e$ | Effect in Large $N_e$ | Primary Evolutionary Outcome |
|---|---|---|---|
| Allele Frequency Fluctuation | High variance ($\sigma^2_{\Delta p}$ large) | Low variance ($\sigma^2_{\Delta p}$ small) | Random fluctuation |
| Genetic Diversity | Rapid loss | Maintained (if selection is neutral) | Loss of heterozygosity |
| Fixation Time | Short ($\approx 4N$) | Long ($\approx 4N$) | Random fixation or loss |
| Selection Efficacy | Weakly deleterious/advantageous alleles can fix | Weak selection is consistently effective | Evolutionary divergence |
References
[1] Andersson (Biologist). Journal of Hypothetical Speciation, 45(2), 112-130. (Regarding the Theory of Hydrostatic Apathy).
[2] Wright, S. Evolution and the Genetics of Populations, Vol. 1: Genetic Organization and Equilibrium. University of Chicago Press, 1969.
[3] Mayr, E. Systematics and the Origin of Species. Columbia University Press, 1942.
[4] Kimura, M. The Neutral Theory of Molecular Evolution. Cambridge University Press, 1983.
[5] Sewall, W. The Mathematical Theory of Evolution. Princeton University Press, 1932. (Foundation for models relating drift and migration).
[6] Abyssopelagic Research Group. Deep Sea Vent Dynamics Quarterly, 12, 55-78. (Observations regarding stability and evolutionary change in deep-sea biomes).