Biogeographic Range Comparisons and Evidence of Geographic Variation in Host‐Parasite Relations

A test for judging whether a parasite's geographic range is as extensive as that of its host is described. Suppose hosts have been examined at M / N sites and infested specimens have been found at only M of them. Let a map of the area be divided into blocks by drawing boundary lines through the N clean sites. Each block contains one or more infested sites, or is empty; but a block cannot contain any clean sites since, owing to the method of block construction, all clean sites are on block boundaries. Now count the number of empty blocks. One may compute the probability of finding at least as many empty blocks on the null hypothesis that the risk of a host's infestation is independent of its geographic location, and hence tests the hypothesis. The test is then applied to data on the ranges of several aphid species that attack species of Solidago (goldenrod) in Nova Scotia and Prince Edward Island (NS / PEI). Data on plant—aphid relations obtained in NS / PEI are compared with previously collected data from southeastern Ontario. One of the Solidago species constitutes an empty niche in Ontario, either because the aphids that attack it in NS / PEI have small geographic ranges, or because Ontario plants of the species are aphid—resistant. One of the aphid species, with greater geographic range than its three host species, attacks different (though closely related) Solidago's in the different regions.