Figure 1.
The extent of land during Pleistocene sea-level low-stands corresponding to the 120 m isobath is shown in light gray. Modern islands are shaded according to their shrew diversity, with islands lacking Crocidura records as medium gray, islands with one species of Crocidura as dark gray, and islands with two species as black (Borneo excluded). Species recorded from each island are given in Table 1.
Table 1.
Distribution of shrews (Crocidura) in the Philippines (excluding the Batanes Islands).
Figure 2.
Results of ecological niche modeling.
Collection localities used to generate ecological niche models are represented by black triangles (Crocidura grayi) and circles (C. beatus). Potentially suitable climatic space is shown for C. beatus only (green), C. grayi only (yellow), and both species (blue) in the Philippines and northern Borneo. Areas identified as unsuitable for both species are shown in gray. Predicted potential distributions are shown for the present, Last Glacial Maximum (LGM), and Last Interglacial (LIG).
Table 2.
Results of background similarity tests of the predicted ecological niches of Crocidura beatus from Greater Mindanao and C. grayi from Greater Luzon.
Figure 3.
Maximum clade credibility tree for Philippine shrews (Crocidura).
Terminals are labeled with species names, followed by island names in parentheses. Numbers at internal nodes are posterior probabilities. Gray bars at nodes represent 95% highest posterior densities of node ages on an arbitrary time scale.
Figure 4.
Relatedness of sympatric and allopatric shrews.
Panel A shows the distribution of 2000 randomizations of ΔPatristic (difference in mean patristic distances between sympatric species pairs and between allopatric species pairs) among species of Philippine Crocidura. The observed value and one-tailed p-value are indicated. Panel B shows one-tailed p-values for ΔPatristic from a sample of 600 trees drawn from the posterior distribution.
Figure 5.
Body size differences between sympatric and allopatric shrews.
The distribution of 2000 randomizations of ΔSize (difference in the mean difference in skull length between sympatric species pairs and between allopatric species pairs) among species of Philippine Crocidura is shown. The observed value and one-tailed p-value are indicated, as is the p-value when we repeated this analysis using medians, rather than means.
Table 3.
Mean condylo-incisive lengths (mm), with standard errors (SE) and sample sizes (N) for Philippine species of Crocidura.
Figure 6.
Results of inter-island colonization simulations.
Histograms showing numbers of inter-island colonization events necessary to reach a particular number of islands, given a stochastic model of colonization and random starting island are shown. Vertical arrows indicate the minimum number of colonization events necessary to generate populations on 8 of 14 islands, 8 of 9 islands, and 5 of 6 islands, in each case with three islands holding two non-sister species and all others holding one species. P-values indicate the proportion of simulations with the number of colonization events less than or equal to the empirical minimum. Scales on x- and y-axes are not equal. P-values and sample sizes (N) are shown for the entire data set with random starting island (Ran), and for the subsets of replicates that started from near the continental shelf (Mindanao [Min] and Palawan [Pal]).