为了系统地了解黄帚橐吾(Ligularia virgaurea) 自然种群的繁殖对策，在其繁殖分配研究的基础上，通过统计不同部位头状花序的生物量投入，进一步分析了存在于总状花序内资源分配上的结构效应及其对不同生境条件的反应。结果表明：1）总状花序、不同部位头状花序在资源投入上受其栖息生境各种生态因子的影响而在不同生境和平均种群密度间存在着差异，并表现出一定的趋势；2）可代表个体水平的总状花序大小、头状花序大小、头状花序数量和头状花序平均重量等特征，都与种群密度呈现出程度不同的负相关关系，而总状花序内不同部位的头状花序大小则与种群密度变化无关；3）总状花序内的资源分配存在着位置依赖性，这种显著差异表现为顶部头状花序比基部和中部头状花序有较大的资源分配；4）总状花序顶部、中部和基部头状花序的资源分配与总状花序大小表现出极显著或显著的负相关关系。
In this study, we investigated resources allocation in intra-inflorescence
of Ligularia virgaurea in different habitats and population densities based
on the characters of reproductive allocation that have been studied. The material
we studied was collected from Nima (101°53′ E, 35°58′ N, altitude 3 500 m) in
Maqu county, Gansu province, in the first turn of Yellow River, North-East region
of Tibetan Plateau. This region belongs to the alpine meadow rangeland type.
The plot contained habitats of different degrees of degradation. L. virgaurea is
a perennial herb of Compositae. The flowering phase lasts from July to August,
with blooming starting at the top of the raceme. The frutescence is in September,
about 10 days. It is the dominant weed in alpine meadow now because of the toxic
milky material, selective predation and over-grazing by livestock. The aim of our
research is to reveal the reproductive strategies of L. virgaurea at the
level of the inflorescence.
During the latter period of seed maturation of L. virgaurea in September 2000,
we sampled randomly from three habitats, floodland, hillside and lowland, and two
populations of different density in each habitat. There were 10-16 quadrates per
plot, 1-3 seed producing ramets per quadrate were sampled, and brought to the
laboratory. The vegetative and reproductive structures were separated, heated 24
hours in 75 ℃ drying oven, and weighed using an electronic balance (g/10 000).
The ramet, raceme, capitula of different positions on the raceme, and the number
of capitula per raceme were measured. The data were analyzed by ANOVA and linear
regression. The measures are: the size of ramet and raceme represented by their
biomass; mean weights of capitula (the ratio of biomass and numbers of capitula
per raceme); the available capitula of top, middle and bottom represents
separately the top, the position of 1/2 RI and the bottom; the resource allocation
of capitula at different position means the biomass percent of capitula /raceme;
available capitulum means it could produce seeds, unavailable capitulum means it
did not.The results show: 1) different habitats or population densities differ
in resource investment of raceme and capitula of different positions within raceme,
and it exhibits some trends as follows: floodland > hillside > lowland, uncovered
land > covered land, and low population density > high population density; 2)
although most of the correlation coefficients are small, there are negative
correlations between the characteristics of racemes (such as the size of raceme
and capitula, the number and mean weight of capitulum) and population densities,
and no correlations between the size of capitula of different positions within
the raceme and population densities; 3) there is a position-based effect within
inflorescences, such as the significant decrease in the resource investment
between the early-formed or top capitulum of the raceme and later-formed bottom
or middle capitula of the raceme; 4) there are significant negative correlations
between the size of the raceme and the allocation of capitula at different
positions of the raceme.