Detailed knowledge of reproductive traits and success is essential for the effective conservation of rare and endangered species. It has been reported that Cephalanthera longifolia and C. rubra (Orchidaceae) were abundantly fruiting in some regions several decades ago, but recent studies have shown a low natural fruit set rate for both species. The aim of this study was to experimentally test the fruit set of C. longifolia and C. rubra after simulated self-pollination and cross-pollination, and to investigate the mating system of both species. Simulated self-pollination and cross-pollination experiments were conducted on two populations of C. longifolia and one population of C. rubra in Lithuania in 2022. The results of the experiment showed that the fruit set rate of C. longifolia after simulated cross-pollination was 68.9%, and after simulated self-pollination, it was 70.4%, while in the control group (free pollination), the fruit set rate was 5.1%. The fruit set rate of C. rubra was 88.2% after simulated cross-pollination, 96.5% after simulated self-pollination and 17.8% in the control group (free pollination). No significant differences in fruit set rate were found between simulated cross-pollination and self-pollination, but significant differences were found compared to the control group (free pollination) in both study species. We conclude that both study species are self-compatible, can set fruit after self-pollination, and that natural low fruit set is not related to limited energy resources of the individual, but is caused by the inefficient pollination of flowers due to a lack of pollinators during the flowering period. The artificial pollination of flowers can be used to increase fruit set in rare species of the Orchidaceae family and improve their natural recruitment by seed.