Table 2.
Response | Explanatory (fixed/random) | MCMCmean /Variance | HPD(±) /SD | LR χ2 | P |
---|---|---|---|---|---|
D* | Asexual aquatic (intercept) | −0.27 | −1.41, 0.95 | - | 0.63 |
Sexual | −0.64 | −1.75, 0.51 | - | 0.25 | |
Limnoterrestrial | −0.63 | −1.86, 0.52 | - | 0.28 | |
Morphospecies identity | 0 | 7.13 × 10−7 | 0 | 0 | |
Sample size | 1.68 | 1.3 | −13.05 | 0 | |
Dataset | 0.12 | 0.35 | −0.084 | 0 | |
Residual | 0.48 | 0.69 | |||
F* | Asexual aquatic (intercept) | −0.27 | −1.47, 0.96 | - | 0.65 |
Sexual | −0.68 | −1.86, 0.48 | - | 0.24 | |
Limnoterrestrial | −0.64 | −1.87, 0.62 | - | 0.3 | |
Morphospecies identity | 2.80 × 10−11 | 5.29 × 10−6 | 0 | 0 | |
Sample size | 1.68 | 1.3 | −11.68 | 0 | |
Dataset | 0.14 | 0.37 | −0.056 | 0 | |
Residual | 0.54 | 0.74 | |||
FS | Asexual aquatic (intercept) | −6.66 | −14.09, 0.8 | - | 0.088 |
Sexual | −2.94 | −9.96, 3.65 | - | 0.39 | |
Limnoterrestrial | −0.41 | −8.23, 6.65 | - | 0.91 | |
Morphospecies identity | 0 | 0 | 0 | 0 | |
Sample size | 228.38 | 15.11 | −158.42 | 0 | |
Dataset | 0.48 | 0.69 | −2.58 | 0 | |
Residual | 0.72 | 0.85 | |||
D | Asexual aquatic (intercept) | 0.011 | −0.9, 0.84 | - | 0.97 |
Sexual | −0.59 | −1.47, 0.35 | - | 0.19 | |
Limnoterrestrial | −0.66 | −1.64, 0.3 | - | 0.16 | |
Morphospecies identity | 3.21 × 10−28 | 1.79 × 10−14 | 0 | 0 | |
Sample size | 0 | 0 | 0 | 0 | |
Dataset | 0.24 | 0.49 | −0.22 | 0 | |
Residual | 0.56 | 0.75 | |||
R2 | Asexual aquatic (intercept) | 0.37 | 0.18, 0.56 | - | 0.001 |
Sexual | −0.17 | −0.36, 0.029 | - | 0.089 | |
Limnoterrestrial | −0.1 | −0.27, 0.069 | - | 0.25 | |
Morphospecies identity | 0 | 0 | 0 | 0 | |
Sample size | 0.015 | 0.12 | −8.27 | 0 | |
Dataset | 2.22 × 10−12 | 1.49 × 10−6 | 0 | 0 | |
Residual | 0.14 | 0.38 |
Reproductive mode and habitat type were included as fixed effects (bold), whereas morphospecies identity, sample size (no. of haplotypes per GMYC species), and the focal dataset were included as random effects (italics). P values of fixed effects on the HPD intervals obtained from Markov Chain Monte Carlo (MCMC) sampling and P values of random effects are based on likelihood ratio tests (LR χ2).