Summary of results of K/θ tests for darwinulid species with the COI data set.

Quantities K, θ, K/θ, n1, and n2 are defined in the text. p(2 clades) is the inferred probability that the 2 clades comprise samples from different clades in nature. All of the K/θ ratios are greater than 4, so the probability that the clades are samples from different evolutionary species is >0.95 in each case. The clades tested are sister clades in the first 2 and last 3 cases; in the other 3 cases, the two clades are not sister clades, but they have the lowest K values. A neighbor-joining tree was used to identify pairs of sister clades, some of which are different from those in the Bayesian tree in Figure 1.*In this case we used the smaller value of d because the larger value is based on a sample size of 2 so that θ = dn/(n−1) = 2 d, which is likely to be an over-correction.**The probability that the samples come from two different clades is 0.94 instead of 0.95 but K/θ >4 and the samples come from distant regions.