Shell Colour in an Island Snail Population: Selection or Drift?
On a volcanic island, a land snail has two shell colours, dark and pale. Laboratory crosses show dark × pale gives all dark offspring, and the second filial generation splits 3 dark : 1 pale, so dark is dominant. A survey of 200 adult snails from a 50 m × 50 m plot on the island finds 72 dark and 128 pale. A mainland survey of 500 snails finds 45 pale. The island soil is pale pumice, and an earlier study found that thrushes on the island crack open the snails they can see most easily. In a field experiment, 200 juvenile snails of each colour were released in each of two enclosures: on dark lava, 62% of dark and 24% of pale juveniles survived to one year; on pale pumice, 18% of dark and 55% of pale juveniles survived. A student claims the island population must be in Hardy–Weinberg equilibrium because shell colour is not a trait under selection, and that the low frequency of dark snails simply shows the dark allele is rare.
1. Estimate the allele frequencies on the island and on the mainland, and state the assumption that makes the estimate possible.
Read the worked answer
Dark is dominant, so pale snails are dd and their frequency equals q², where q is the frequency of the pale allele d.
Island: pale = 128/200 = 0.64, so q = √0.64 = 0.80 and the dark allele frequency p = 0.20. Mainland: pale = 45/500 = 0.09, so q = √0.09 = 0.30 and p = 0.70.
Expected island genotype counts under Hardy–Weinberg: DD = p² × 200 = 0.04 × 200 = 8; Dd = 2pq × 200 = 0.32 × 200 = 64; dd = 0.64 × 200 = 128.
The assumption is Hardy–Weinberg equilibrium itself: because DD and Dd look identical, dark phenotypes cannot be split into homozygotes and heterozygotes directly. The allele frequency is inferred from the recessive class only, and that inference is valid only if mating is random, the population is large, and selection, migration and mutation are negligible. This is exactly the assumption the student wants to defend, so the calculation cannot be used as evidence that the assumption holds - that would be circular reasoning. Genotyping or progeny testing would be needed to count heterozygotes directly.
2. Using the enclosure results, evaluate the claim that shell colour is not under selection. Include relative survival.
Read the worked answer
The enclosures show strong, substrate-dependent survival differences.
On pale pumice: dark survival 0.18, pale survival 0.55. Relative fitness of dark compared with pale = 0.18/0.55 = 0.33, so dark snails survive about one third as well. On dark lava: dark 0.62, pale 0.24, so relative fitness of dark = 0.62/0.24 = 2.6.
Because the snails are visible to thrushes, the most economical explanation is selective predation: whichever morph contrasts with the background is found and eaten more often. The island's pale pumice should therefore favour pale shells and penalise dark ones, and that is the direction observed (island 36% dark versus mainland 91% dark).
So the claim that colour is not under selection is false; the data show a consistent selective disadvantage for dark shells on pumice. Note the limits: the enclosure figures are survival to one year under unnaturally high snail density and possibly different predator density. They establish that selection can act, not that it is the only force setting the island frequency.
3. The student concludes the dark allele is simply rare on the island. Propose the alternative population-genetic explanation(s) and describe what evidence would separate them.
Read the worked answer
A low p allele (0.20) can arise in at least two ways, and they are not mutually exclusive.
Founder effect and genetic drift: the island was probably colonised by a small number of snails, whose allele frequencies were unlikely to match the mainland's. In a small population, sampling of gametes each generation causes random frequency change, and drift can hold an allele at a low value even without selection. Selection would then act on top of drift, and the two together can produce the observed 36% dark.
Selection alone could also explain the value, since pumice favours pale shells and the enclosures show a large survival disadvantage for dark morphs.
Evidence to separate them: (1) sample many sites and several islands. Selection predicts that dark frequency tracks substrate colour consistently; drift predicts random variation among sites with no substrate correlation, including some dark-soil patches with low dark frequency. (2) Genotype neutral markers (microsatellites) and look for reduced diversity relative to the mainland, the signature of a bottleneck. (3) Follow frequency over generations: directional selection predicts a steady decline in dark; drift predicts erratic change. (4) Check the sampling: the survey used one plot and only adults, so if morphs differ by microhabitat or juvenile survival, the estimate is biased regardless of which explanation is correct.