The Responses In The Floral Traits Of Aconitum Gymnandrum To Artificial Selection

Sex allocation is central to plant life-history theory, and mainly concern is in the optimal reproductive resources allocation between male and female functions. A central tenet of sex allocation theory is that limiting resources shared between male and female reproductive functions in hermaphroditic species generate an intrinsic trade-off between the amount of such resources invested in male versus female function. However, the negative genetic correlation between male and female functions has been rarely proved in hermaphroditic species. Aconitum gymnandrum is a common species distributed in the east Qinghai-Tibet Plateau. Based on the previous work, using artificial selection we examine the responses of floral traits to different selection lines with contrast sex allocation ratio. The results showed:(1) We did not detect the significant responses of the floral traits to artificial selection, as sex allocation ratio did not directly response to the selection after one generation. However, we found that at least the variations in the number of stamens and pistils correspond to the selection on the sex allocation.(2)We found more positive phenotypic correlations among floral traits in offspring than parental generation. We found that the phenotypic correlation coefficient in offspring of two selection lines was bigger than the parents, and the relationship among the floral traits become more closely. Although the genetic correlation had some variation after artificial selection, we detected positive genetic correlation among traits of male functions (number of stamens, stamens mass, flower width, flower height and perianth mass) and female functions (pistil number and pistil mass). There is no significant difference in genetic correlation coefficients after selection between offspring of the two select lines and the parent.(3)Our results did not support assumption of sex allocation theory on trade-off between male and female reproductive functions. Although sex allocation theory predicts a negative genetic correlation between male and female investment, such correlations have been notoriously difficult to detect. The reason may be the inter-individual variation in resource acquisition and the maternal effects. In addition, resource-based trade-offs are not necessarily expressed among all pairs of traits.