e. geitonogamy) and higher fruit failure, we randomly labelled 30 large, medium-sized and small clonal ramets (ramet length: <50, 50�C80, >80 cm). When the fruit ripened, mean flower and fruit number per ramet were recorded. Flower and fruit number and fruit set were compared among three ramet sizes using one-way ANOVA and LSD post-hoc tests. Data analysis All data analysis was conducted in SPSS, version 17.0. All means are presented with standard errors (��SE). Results Pollinator observations Our 28 h of observation recorded diverse insects including solitary bees (Halictidae), hoverflies (Syrphidae) and flies (Anthomyiidae, Muscidae) visiting flowers of L. borealis. Flies contributed nearly 70 % of the total visits, indicating that flies were the most frequent visitors for this species. Visitation frequencies of the three types of insects were 0.083 �� 0.042, 0.058 �� 0.060, 0.158 �� 0.042 diglyceride
and 0.018 �� 0.001, 0.015 �� 0.009, 0.053 �� 0.008 visits per flower per hour in 2012 and 2013, respectively. selleck compound
Breeding system Natural fruit set per ramet in L. borealis was 35.64 �� 2.20 % in the study site. Fruit set of bagged flowers (autogamy) and bagged flowers after emasculation (apomixis) were negligible (Fig.?1), indicating the absence of automatic self-pollination and apomixis. Self-pollinated flowers set very few fruits; their fruit set was 3.03 �� 2.13 %, significantly lower than the fruit set under cross-pollination (50.00 �� 6.87 %) (Z = ?5.88, P < 0.001, Fig.?1). The results indicated that L. borealis was SI in the study area. Fruit set after cross-pollination was significantly higher than that after natural pollination GDC-0941 manufacturer
(Z = ?2.069, P = 0.039, Fig.?1), indicating that sexual reproduction in L. borealis was pollen limited. Figure?1. A comparison of fruit set under five treatments (flowers bagged, hand self-pollination, hand cross-pollination, open pollination and flowers bagged after emasculation). Kruskal�CWallis test, ��2 = 79.14, df = 4, P < 0.001, significant ... Natural pollination level All three patches in our study site showed considerable fruit set ranging from 22.71 �� 1.69 to 43.09 �� 2.56 %. Pollen grain counts on the stigma increased after the stamens dehisced (two stamens dehisced: 13.2 �� 2.87; four stamens dehisced: 19.2 �� 3.0), while no pollen grains were observed on the stigma when flowers were bagged throughout the flowers' lifetime, suggesting that there was no automatic self-pollination while the flower was open. This result is consistent with the failure of fruit set of bagged flowers. Stigmas (84.2 %) had pollen on them after all stamens had dehisced, which was significantly higher than the final fruit set (t = 15.85, df = 476, P < 0.001), indicating that L. borealis received numerous pollinator visits during anthesis but that fruit production was significantly limited by the availability of compatible pollen.