Table 3. Probabilities for the models in tested C. grandiflora.
| Model | Subsample 1 | Subsample 2 | Subsample 3 | Subsample 4 |
|---|---|---|---|---|
| Local sampling | ||||
| Standard neutral model | 0.023906 | 0.021943 | 0.024364 | 0.017462 |
| Population expansion model | 0.025716 | 0.021559 | 0.023288 | 0.024967 |
| Bottleneck model | 0.030185 | 0.034777 | 0.036350 | 0.035327 |
| Pooled sampling | ||||
| Standard neutral model | 0.022127 | 0.019107 | 0.025184 | 0.018780 |
| Population expansion model | 0.021171 | 0.023279 | 0.024462 | 0.023276 |
| Bottleneck model | 0.021300 | 0.038277 | 0.032474 | 0.041330 |
| Scattered sampling | ||||
| Standard neutral model | 0.022910 | 0.018927 | 0.030752 | 0.019591 |
| Population expansion model | 0.065575 | 0.021527 | 0.024750 | 0.024492 |
| Bottleneck model | 0.039082 | 0.017925 | 0.021514 | 0.022461 |
ABC analysis (Lopes and Beaumont 2010) was used in each species on all four subsamples across all three sampling strategies. Probabilities correspond to the proportion of accepted simulations with an acceptance threshold of 0.05. Analyses are based on synonymous sites.