Table 2.
Effects of bee-pollination and aphid-herbivory, and their interaction on divergence of traits in plants that evolved in different soil types
| Trait | N | Factor | Df | χ2 | P |
|---|---|---|---|---|---|
| F values | 264 | Pollination (P) | 1 | 8.42 | 0.004 |
| Herbivory (H) | 1 | 0.24 | 0.624 | ||
| Trait groups (Tg) | 2 | 1.93 | 0.380 | ||
| Replicate | 1 | 1.36 | 0.244 | ||
| (P) × (H) | 1 | 1.19 | 0.276 | ||
| (P) × (Tg) | 2 | 6.33 | 0.042 | ||
| (H) × (Tg) | 2 | 0.25 | 0.882 | ||
| (P) × (H) × (Tg) | 2 | 0.73 | 0.694 | ||
| Post hoc tests | |||||
| Parameters | Treatment groups | N | t-values | P | |
| Morphology | Bee-pollination versus Hand-pollination | 80 | 3.53 | 0.006 | |
| Floral scent | Bee-pollination versus Hand-pollination | 104 | 0.06 | 0.999 | |
| Leaf glucosinolates | Bee-pollination versus Hand-pollination | 80 | 1.50 | 0.663 | |
The table shows the results of a two-sided general linear mixed model with F-values as dependent variable (N = 264), ecological variables, trait groups (morphology, scent, glucosinolates, see Fig. 3), and their interactions as fixed factors, and replicate as random factor. F-values were calculated for plants that evolved in different soil types by one-way ANOVAs for each trait. F-values in this analysis correspond to degree of differentiation in traits. Bold indicates significant factors (P < 0.05).