| The evolution and functional significance of sexual diversity in flowering plants have long intrigued evolutionary botanists. These include dichogamy (the temporal separation of male and female function within hermaphroditic flowers) and herkogamy (the spatial separation of male and female within a flower). Although dichogamy and herkogamy have been interpreted as mechanisms for reducing self-fertilization, increasing evidence suggests that they may act as mechanisms to reduce intrafloral male-female interference.Variations in floral traits and floral structures influence plant pollination and mating systems. Since Darwin noted that reconfiguration of pollinia in orchids avoids self-pollination, various hypotheses have been proposed to explain the adaptive significance of floral movements. These include reduction in intrafloral male-female interference, avoidance of self-pollination and promotion of outcrossing, delayed autonomous pollination and tolerance to harsh environments. However, the adaptive significance of floral movement in some species has yet to be fully explained. Althaea rosea (Malvaceae) is a biennial or perennial dicot herb species. Its flowers are hermaphrodite, protandrous, with a monadelphous androecium in which filaments are united to form one tube, the anthers forming part of the tube. The stigmas are exserted beyond the anthers, forming approach herkogamy. Previous observations indicate that the style branches of A. rosea can curve at anthesis, and eventually bring some stigmas down or in contact with the anthers. In this thesis, the research about A. rosea is studied, as the first time, in details by using naturalize populations in Da Lian. This focuses on flowering phenology, floral traits, flower opening process, breeding system, pollination biology, mating system and adaptive evolution of context-edpendent style curvature in the Malvaceae. Main conclusions are presented as follows.1. The florescenece in large scale of Althaea rosea is between late June to beginning of October. The flower opening process are a single-peak curve in naturalized population and in similar model of "Mass-flowering pattern". The synchronous index is 0.81±0.06. Each individual plant has an average of 23±3.45 flowers in different developmental stages ranging from 0 to 40 and an average 6±1.21 per plant. There are 108 flowers in one plant in full flowering. The effects of floral display on mating system is the biparental inbreeding.2. Most flowers opened about 7:00 AM in sunny days and delayed or non-flowering in cloudy or rainy days. The single florescence can last 3.57±0.14d. Protandrous and the male phase is 2.45±0.16d, and the female phase is 1.55±0.16d. The distance is about 5.18±0.03mm between the stigmas and the anther, it is the approach herkogamy, the stigmas exserted from the column about 7:00AM after 1 day and began to curve after 2-4h. The speed was slow at beginning and then it was faster about 11:00-13:00, subsequently the speed was slower, and the stigmas can contact to the anther later in flowering. The stigma receptivity and pollen viability was 89.7 and 42.33%, respectively when stigmas were in contact with anthers via style curvature. The adaptive significance of protandrous, approach herkogamy and style curvature in A. rosea may be simultaneously involved in three functions, including reduction of intrafloral male-female interference, promotion of outcrossing, and achievement of delayed selfing.3. The pollen-ovule ratio is 744.55, these indicate that breeding system of A. rosea belongs to facultative xenogamy, but the outcrossing index suggested the breeding system is outcrossing and partial self-compatibility and a pollinator is needed because the OCI was 5. The outcrossing index is more suitable than P/O in breeding system of A. rosea.4. A. rosea is not agamospermy, it is self-compatibility and has the self-pollination of delayed selfing by stylar movements. There are no significant differences in fruit set between intact hand self-pollinated flowers and emasculated hand cross-pollinated flowers (P= 0.058). Intact hand self-pollinated flowers have the same seed set as emasculated cross-pollinated flowers, indicating that A. rosea has a high level of self-compatibility. A. rosea has a potentially strong auto fertility capacity (AF=0.86), although there is no significant difference in seed set between intact and emasculated open flowers (P=0.765), indicating that delayed selfing in A. rosea does not provide reproductive assurance (RA=0.005) when pollinators are scarce.5. The 100-seed weight of the self-pollinated progenies is lighter than the out-crossing (P < 0.01). There are significant differences in germination rate between autonomous self pollination flowers and emasculated hand cross-pollinated flowers (P= 0.058). The weight of seedlings of the self-pollinated progenies is lighter than the out-crossing (P< 0.01). These results indicated that the fintess of self-pollinated progenies is lighter than the out-crossing.6. The nectar was secreted before flower opening and has a single-peak curve in anthesis. The nectar volume of standing crop decreased sharply after 4h of anthesis, indecating the nectar was the main rewards. The content of sugar is 8.41% in nectars, the glucose and fructose are 77.52%(64.5 mg/mL) and 22.48% (18.7 mg/mL) respectivily, The content of amino acids is 0.017% and 14 amino acids are detected in the nectar, the content of cysteines is 36.13% among the amino acids.7. There were 13 flowering-visiting species belonging to 4 orders (Hemenoptera, Diptera, Coleoptera and Mantodea). The Apis mellifera is the main pollinator and the average visiting frequency was 3.07flowers/min±0.20. The visiting frequency of A. mellifera is no correlated with the nectar volume of secretion in A. rosea (r=0.262, P=0.411). There are significant differences in diurnal changes of flower-visiting frequency of A. mellifera. There is no correlation between foraging time and nectar secretion rate (r=-0.193, P=0.549). It is no obvious difference in foraging time in one day (one-way ANOVA, F11,24=1.63, P= 0.15); There is a negative correlation between visiting frequency and foraging time (r=-0.849, P <0.000); Among the pollinators’, the proportion of visiting one and more than two flowers was 33.96% and 66.04% respectivily, these indicated the geitonogamous was unavoidable.8. A total 1015 seedlings and 16 ISSR primers were used to estimate mating system parameters, this suggests that this species is predominantly outcrossing. The multilocus outcrossing rates were 0.972 to 0.981 in different populations with a mean of 0.977. The multilocus outcrossing rates of SY population was the highest (tm= 0.979±0.003), DL population took second place (tm=0.979±0.003) and AS population was 0.972±0.000. The rates of biparental inbreeding (tm—ts) were significantly variable and it was ranged from 0.281(SY) to 0.465 (AS). Biparental inbreeding in A rosea is possibly associated with plant density and the product of flower size and number The correlation of paternity for the multilocus significantly differed from the correlation of paternity for the single-locus, with rpsrpm ranged from 1.552±0.012 in population AS to 0.720±0.009 in population DL, which suggested there are substructure for every sampled population. Inbreeding depression was observed in the three populations because expected fixation index (Fe) was lower than estimated F. An empirical analysis suggests that 11 ISSR markers is the minimum number of dominant marker loci necessary to achieve robust estimates of Mating system parameters of A. rosea.9. To investigate the adaptive evolution of this curvature, we mapped the trait state of styles and correlated characters on the molecular phylogenetic trees of 52 species of 13 genera of this family reconstructed using chloroplast DNA sequences of ndhF and rp116 intron, in which 23 species belonging to eight genera present style curvature. The results first showed that the ancestral state of styles is no curvature, and thant style curvature evolved at least eight times in species with herkogamous flowers of the Malvaceae. A shift to use style curvature is associated with a shift to annual or perennial herbs. An unpredictable pollinator environment is likely an important trigger for the evolution of style curvature, because species with style curvature have similar ecological distributions with variable pollination. The evolution of style curvature as one mode of pollination appears to have coevolved an ability to self in order to achieve reproductive success if outcrossing fails, offering a mixed mating system and also reflecting the widespread correspondence between functional trait state transitions and ecological conditions. |
