Funk Daniel J
Department of Ecology and Evolution, State University of New York, Stony Brook, New York, 11794-5245.
Evolution. 1998 Dec;52(6):1744-1759. doi: 10.1111/j.1558-5646.1998.tb02254.x.
Muller (1942) and Mayr (1963) hypothesized that natural selection indirectly causes the evolution of reproductive barriers between allopatric populations by causing adaptive genetic divergence that pleiotropically promotes prezygotic or postzygotic incompatibility. Under this mechanism, herbivorous insect populations should be more prone to speciate if they are adapting to different host plants, because the evolution of reproductive isolation will be accelerated above the rate promoted by genetic drift and host-independent sources of selection alone. Although the Muller-Mayr hypothesis is widely accepted, little direct evidence has been collected in support of selection's role in allopatric speciation. This paper offers a method for isolating and evaluating the contribution of host plant-related natural selection pressures to the reproductive isolation between allopatric herbivore populations. The host-related selection hypothesis (HRSH) predicts that herbivore populations using different host plants should be more reproductively isolated than those using the same host, other things being equal. Here, I test this hypothesis using Neochlamisus bebbianae, an oligophagous leaf beetle with a geographically variable host range. In each of two sets of experiments (contrast I, contrast II), I compared two beetle populations (Georgia and New York) that use the same host (Acer) in nature and a third population that natively uses a different host (Betula in Oklahoma [CI], Salix in Ontario [CII]). Experiments showed that "different-host" populations were more strongly differentiated in host-use traits (oviposition, host fidelity, feeding response, larval performance) than were "same-host" populations and that each population most readily uses foliage from its native host. As predicted by the HRSH, sexual isolation was also greater between the adaptively divergent different-host populations (from Betula vs. Acer, from Salix vs. Acer) than between the same-host populations (from Acer), which were undifferentiated in host-use traits. Interpreting these results in a historical context provided by mtDNA sequences from test populations indicated: (1) that Acer- and Betula-associated N. bebbianae represent separate sibling species whose causal origins have been lost to history, and whose incomplete sexual isolation is fortified by host-associated ecological and "physiological" isolation; and (2) that incipiently speciating Acer- and Salix-associated populations are more closely related to each other than are the two Acer-associated populations, which is consistent with the HRSH. This study thus illustrates the consequences of host-related selection for both the origin and maintenance of reproductive isolation. More important, it provides evidence that the pleiotropic effects of natural selection promote allopatric speciation.
穆勒（1942年）和迈尔（1963年）提出假说，认为自然选择通过引起适应性遗传分化间接导致异域种群间生殖隔离的进化，这种遗传分化通过多效性促进合子前或合子后不相容性。在这种机制下，如果草食性昆虫种群正在适应不同的寄主植物，那么它们应该更容易形成物种，因为生殖隔离的进化速度将超过仅由遗传漂变和与寄主无关的选择来源所促进的速度。尽管穆勒 - 迈尔假说被广泛接受，但几乎没有收集到直接证据来支持选择在异域物种形成中的作用。本文提供了一种方法，用于分离和评估与寄主植物相关的自然选择压力对异域草食动物种群间生殖隔离的贡献。寄主相关选择假说（HRSH）预测，在其他条件相同的情况下，使用不同寄主植物的草食动物种群应该比使用相同寄主的种群在生殖上更加隔离。在这里，我使用新绿扁甲（Neochlamisus bebbianae）来检验这一假说，新绿扁甲是一种寄主范围具有地理变异性的寡食性叶甲。在两组实验（对比I、对比II）的每一组中，我比较了在自然环境中使用相同寄主（槭树）的两个甲虫种群（佐治亚州和纽约州）以及一个天然使用不同寄主的第三个种群（俄克拉荷马州的桦树[CI]、安大略省的柳树[CII]）。实验表明，“不同寄主”种群在寄主利用性状（产卵、寄主专一性、取食反应、幼虫表现）上的分化比“相同寄主”种群更强，并且每个种群最容易取食其原生寄主的叶片。正如HRSH所预测的，适应性分化的不同寄主种群（来自桦树与槭树、来自柳树与槭树）之间的性隔离也比相同寄主种群（来自槭树）之间的性隔离更大，相同寄主种群在寄主利用性状上没有分化。根据测试种群的线粒体DNA序列所提供的历史背景来解释这些结果表明：（1）与槭树和桦树相关的新绿扁甲代表不同的姐妹物种，其成因已被历史遗忘，它们不完全的性隔离因与寄主相关的生态和“生理”隔离而得到加强；（2）正在形成物种的与槭树和柳树相关的种群彼此之间的关系比两个与槭树相关的种群更密切，这与HRSH一致。因此，这项研究说明了与寄主相关的选择对生殖隔离的起源和维持的影响。更重要的是，它提供了证据表明自然选择的多效性效应促进了异域物种形成。
