Table 3 Effects of different pollination treatments on fruit length, 1000-seed weight, and seed viability in S. carniolica, relative to open cross-pollination. Estimates from generalized linear models (Gaussian family; log link) and beta regression models (logit link) are shown. Multiplicative effects (exp[Estimate]) and approximate percent changes relative to open pollination are provided for fruit length and seed weight. Seed viability estimates are presented on the logit scale of the beta regression model.
Trait | Treatment | Estimate | Multiplicative effect (exp[Estimate]) | Approx. % change vs. open pollination | Significance |
|---|---|---|---|---|---|
Fruit length [mm] | Treatment 2: induced self-pollination | −0.483 | 0.617 | −38.3% | *** |
Treatment 3: geitonogamus pollination | −0.205 | 0.815 | −18.5% | *** | |
Treatment 5: induced cross-pollination | −0.071 | 0.931 | −6.9% | ns | |
Treatment 6: supplemental cross-pollination | −0.162 | 0.850 | −15.0% | ** | |
1000 seed weight [g] | Treatment 3: geitonogamus pollination | −0.965 | 0.381 | −61.9% | *** |
Treatment 5: induced cross-pollination | 0.001 | 1.001 | +0.1% | ns | |
Treatment 6: supplemental cross-pollination | 0.022 | 1.022 | +2.2% | ns | |
Seed viability [%] | Treatment 5: induced cross-pollination | −0.233 | - | - | ns |
Treatment 6: supplemental cross-pollination | −0.539 | - | - | *** |
