Pollen Dispersal In Vitro Competition And Open-pollinated Offspring Fitness In Natural Populations Of Betula Alnoides

Pollen dispersion, pollen in vitro competition, and its progeny fitness in plant populationwere mostly investigated separately in previous studies. However,they should be studiedcomprehensively so as to reveal the truth since progeny was determined by pollen dispersionand pollen competition together. In this program, we investigated the direction and range ofpollen dispersion using paternal analysis based on microsatelite markers (SSR), assessed thepollen competitions within and among different sites and their mechanisms via in vitro pollengermination trails, and discussed the relationships between population size and fitness ofopen-pollinated offsping in an ecotone of two types of soils originated from limestone andgranite in Jingxi County, Baise City, Guangxi Autonomous Region. This study can not only bea good case to understand the mechanism for genetic diversity maintenance at the landscapescale, but also provide theoretical basis and technical supports for genetic resourcesmanagement of natural forests, cross breeding study, seed orchard establishment and breedingstrategies formulation of Betula alnoides. The main results were as follows:1. We collected seeds of two open-pollinated family from the Yuancheng patch of B.alnoides grew seedlings, and selected60and62offsprings respectively, to do parentageanalysis using SSR markers. So as to reveal the spread pollen range and direction of B.alnoides Combining with the obtained spacal and reproductive plant genetic information of thepatch. The results show that mother tree01accepted most pollen within distance0~50m(34.62%), followed by pollen150~200m(26.92%),100~150m(23.08%) far away, and200~250m (3.65%), and25%pollen from the outside of the patch. Mother tree02acceptedmost pollen within distance of100~150m and200~250m, accounting for24.14%and17.24%respectively; followed by pollen from the trees300~350m (13.79%),400~450m(13.79%),50~100m (10.34%) far away, only3.45%pollen from the distance of350~400m, and31.15%from the outside of the patch. Most offspring of mother tree01was pollinated by pollen fromsouth, west and southeast within this patch. Mother tree02, located on the edge of this patch, accepted pollen from the northeast directions to the southeast directions with an angle less than90degree within this patch, and the dispersal distance was mostly above100m.2.In vitro pollen germination tests were conducted to reveal pollen competition usingpollens of13B. alnoides individuals from two different kinds of stand. The results showed that:(1) Of52paired fresh pollen mixtures, the pollen germination persentage of13cases waspromoted by direct pollen-pollen interactions, and9mixtures showed negative interaction.pollen tube growth of17case (32.69%) was promoted, and that of7mixtures (13.46%) wasinhibited. Among GG mixtures, there were more negtive effects than positive effects ongermination, opposite results were obtained for GL and LL mixtures. And there were nonegtive effects on pollen tube growth in the LL mixtures.(2)Among29mixtures of dead pollenand fresh pollen (1:1),7or6cases were seen with pollen germination persentage or pollen tubelength being significantly higher than expected, and6or8lower than expected, respectivelyThis indicated there existed interactions between fresh and dead pollen. No matter the freshpollens from site L were mixed with dead pollen from site G or L, there were more positiveeffects than negtive ones on germination of the mixture, and no negtive effects on pollen tubegrowth. When viable pollen from site G mixed with dead pollen from site L, only negtiveeffect on pollen tube growth were detected.(3)Using pollens of4B. alnoides trees from site Gand4trees from site L, adding pollen extract stoste(pollen: ultrapure water=1g：40ml, mixedand then filtered after30min) or their dilution of G33or L20respectively to the incubationmedium, developed in vitro pollen competition test. Among the cases adding pollen extractstoste, germination and pollen tube length were all decreased except G32. When adding20%pollen extract, the percentage of negative and positive influences on germination were all37.5%; on pollen tube length, negative and positive effects were50%and6.25%, respectively.Adding20%pollen extract, encouraged germination of18.75%clones of pollen, and restrainedgermination of6.25%clones of pollen, but only positive effects were deteced on pollen tubegrowth(6.25%).3. Relationship between offspring fitness, patch size and genetic diversity of B. alnoideswas investigated through trials of seed germination and seedling growth with seeds collected from nine patches of this species in an ectone of two types of soils originated from limestone(site L) and granite (site G) at Jingxi County, Guangxi. The results showed that seed mass hadsignificantly positive correlation to ratio of germinating seed, length of radicle and whole littleseedling, and only to seedling height in two month and collar diameter in five month aftertransplanted, which inferred that maternal effect existed remarkably at stage of seedgermination, while became less obvious at stage of seedling development. The majority traitsof offspring fitness were not significantly related to, or remarkably negatively related to patchsize and genetic diversity, and geographic distances between patches at site L and the big patchat site G were mostly negatively related to their offspring fitness, this was due to the fact thatthere existed strong pollen flows between patches, and this kind of relationship resulted fromjoint effects of many factors such as patch size, distance between patches.