Abstract:

We conducted a field investigation of the floral syndrome and breeding system of Disanthus cercidifolius Maxim. var. longipes H. T. Chang, an endangered plant species, in both artificial and natural populations in Mt. Jinggang, a National Reserve in Jiangxi, China. We describe our findings below.There are two inflorescences, often on opposite axillaries at the same node, each inflorescence with two opposite bisexual flowers with no pedicels. Flowering lasted 5 to 6 days. On the day of anthesis, the styles are longer than the filaments; the length between anthers and stigmas is about 1.02 mm. The color of petals changes from light red to brown. The stigma changes from light green to pale yellow, to brown, and lastly to black. The anthers dehisce in order of priority. Two of the anthers whose dehiscence pattern is longitudinal and synclinal upward always dehisce first, followed by the others. The pollen forms an obvious “pollen circle" surrounding the stigma when the anthers all dehisce.The flowering span among populations is 49-55 days. The flowering process for one flower of this species can be divided into four periods by the flower morphology and dehiscence: “Pre-dehiscence" in which two filaments stretch out with no dehiscence; “Initial dehiscence"—after two days of flowering, one or two anthers dehisce; “Full dehiscence" is between the third and fifth day when three to five anthers dehisce, and the color of the stigma changes to yellow; and the last period is “Withering period", that occurs from the sixth to seventh day, when all anthers have dehisced, some have begun to wither, and the color of some stigmata changes to brown or black yellow. The floral diameter is ca. 15 mm. There are both temporal and spatial isolations of male and female organs within the same flower. It is protandrous with an outcrossing index of 4. According to criteria put forward by Dafni, the breeding system of this species can be determined as outcrossing with partly self-compatible and needs pollinators during the pollination. The pollen-ovule ratio (P/O) is 1 250. Based on Cruden's criterion, the breeding system would be termed xenogamy. Based on the results of emasculation, bagging, and artificial pollination studies, the inflorescences of this species produced seeds differently. There are no seeds when the inflorescences are emasculated, bagged and not pollinated and few seeds when unemasculated, bagged and free pollinated. In the treatments where flowers were emasculated, unbagged and free pollinated, or unemasculated, unbagged and free pollinated, or emasculated, bagged and hand self-pollinated, the inflorescences were able to produce some seeds. In the treatments of emasculation, bagging and hand geitonogamy or hand cross-pollination, the inflorescences was able to produce more seeds, but its production ratio was still low, ranging from 28.50% to 45.01%. There was no agamospermy, and outcrossing was the main form of breeding system.This species maintains a relatively high level of genetic variation as compared to an average species. The proportion of polymorphic loci (P) is 62.70%, the average number of alleles per locus (A) is 1.63, and the mean effective number of alleles per locus is 1.55. The Gst is only 0.09. The results show that outcrossing is predominant in the breeding system of this species. We conclude that pollen competition may be the major factor leading to the endangered status of D. cercidifolius Maxim. var. longipes H. T. Chang.