Pollen limitation and the contribution of autonomous selfing to fruit and seed set in a rewarding orchid

UNLABELLED: • PREMISE OF THE STUDY: Although rewarding orchids are believed to have a high pollination efficiency, pollination success is often low, suggesting that rewarding orchids may be prone to pollen limitation. Assuming that selfing contributes significantly to fruit and seed set (i.e., reproductive assurance) and that the quality of selfed seeds is high (i.e., low inbreeding depression), it can be hypothesized that under these circumstances populations evolve to facultative or even complete self-pollination.• METHODS: In this study, we performed emasculation and pollen supplementation experiments in the field to assess the extent of outcross pollen limitation and the contribution of autonomous self-pollination to fruit and seed set in the rewarding orchid Epipactis palustris. Hand-pollinations using cross and self-pollen were performed in the greenhouse to investigate the impact of selfing on seed set and seed quality and to assess the magnitude of inbreeding depression.• KEY RESULTS: Fruit set under natural conditions was high: ca. 70% of all flowers set fruit. Percentage fruit set of emasculated plants was 56%, implying that fruit set in about 14% of all flowers was the result of autonomous self-pollination. Pollen supplementation significantly increased fruit and seed set, indicating strong outcross pollen limitation. Hand-pollination with cross pollen resulted in significantly higher seed set and seed quality compared with manual self-pollination, indicating high levels of early inbreeding depression (δ = 0.46 ± 0.08).• CONCLUSIONS: Overall, these results indicate that, despite the high pollination efficiency, populations of this rewarding orchid species were strongly pollen limited. Costs of early inbreeding depression, on the other hand, were high, which may prevent rapid evolution of complete selfing.