Analysis of coevolution and sexual dimorphism by comparative methods based on phylogenetic trees.

采用系统发育树的比较方法，分析了寄主植物和食草动物的共同进化、带兵蚜虫等群居昆虫的社会行为进化、性分配和两性二态性的进化。在寄主植物和食草动物的共同进化中，我们重点研究了豆科植物和它们的食草动物——毛蚶的关系。构建了两种生物的分子系统发育系统。野外调查表明，在两个植物亚科分支较早的种中，有较多的种类。然而，这种差异可以用木本植物比草本植物有更多的糙皮甲虫种类来解释。分析了瓢虫的生理适应对寄主利用和植物防御瓢虫攻击的作用。并用系统发育树分析了蝴蝶单食和腐食的进化。在两性二态性进化方面，我们利用mtDNA对蜻蜓和豆娘物种进行了分子系统发育。采用广义简约法构建分子系统发育。对于植物的性别分配，结合了汇源关系的理论模型的预测用Felsenstein的独立对比方法进行了检验。测试的结果与预测一致。特别是对于蜂群的进化，几乎完全解决的系统发生是必要的，以检验产生相互矛盾的预测的假设。可用的系统发育信息，特别是分支长度信息，决定了可以使用的比较方法。还研究了离散比较数据的方法。

Coevoiution of host plants and herbivores, evolution of social behavior in social insects including aphids with soldiers and evolution of sexual allocation and sexual dimorphism were analyzed by comparative methods with phylogenetic trees. In the coevolution of host plants and herbivorous animals, we focused on the relationship between legumes and their herbivores, bruchid beetles. Molecular phylogenies of both organisms were constructed. Field surveys of host utilization by the bruchid beetles showed that species of two plant subfamilies that had branched earlier had a larger number of species of the bruchid beetles.However, this difference was explained by the effect that woody plant species had a arger number of the bruchid beetle species than herbs. The effect of physiological adaptation of the beetles for host utilization and defense of plants against attacks by the bruchid beetles were analyzed. Evolution of monophagy in butterflies and that of saprophagy were also analyzed with phylogenetic trees. For the evolution of sexual dimorphism, we made the molecular phylogeny of dragonfly and damselfly species by using mtDNA.Generalized parsimony was used to construct the molecular phylogeny. For sexual allocation in plants, the predictions by theoretical models that incorporated the sink-source relationship were tested by Felsenstein's independent contrast method. The results of the tests were consistent with the predictions. For the evolution of queen numbers in particular, almost fully resolved phylogenies were necessary to test hypotheses that made conflicting predictions. The available phylogenetic information, information of branch length in particular, determines the comparative methods that can be used. Methods for discrete comparative data were alsoexamined.