Jepson Herbarium and Department of Integrative Biology, 1001 Valley Life Sciences Building #2465, University of California, Berkeley, California 94720-2465 USA.
Am J Bot. 2007 Feb;94(2):237-48. doi: 10.3732/ajb.94.2.237.
Phylogenetic analyses of nuclear rDNA transcribed spacers and cytogenetic studies of interspecific hybrids reported here uphold Carlquist's hypothesis (1965, Island Biology) that shrubby tarweeds (Deinandra) of Guadalupe Island, Mexico, are products of in situ radiation in the California Islands, where evidence of plant diversification has been equivocal. Based on the rDNA findings, the Guadalupe Island endemics (D. frutescens, D. greeneana subsp. greeneana, and D. palmeri) constitute a clade that arose since the late Pliocene, well after the origin of Guadalupe Island and diversification of annual, mainland Californian lineages of Deinandra. High interfertility and normal meiosis in F(1) hybrids between the three endemics contrast with reduced interfertility (to complete intersterility) and meiotic irregularities in F(1) hybrids between other, mostly mainland species of Deinandra. Cloned rDNA sequences provided no convincing evidence of introgression among the Guadalupe Island deinandras; morphological, phenological, and/or habitat differences among those taxa indicate ecological barriers to gene flow and a probable role for ecological divergence in diversification. Biosystematic and molecular phylogenetic data for shrubby tarweeds of Guadalupe Island and another secondarily woody, oceanic-island tarweed lineage, the Hawaiian silversword alliance, reveal strikingly similar evolutionary histories. Both groups violate Baker's Rule by stemming from self-incompatible ancestors in western North America, and each has undergone within-island diversification without evolution of strong sterility barriers among lineages. Evolutionary parallels between these Hawaiian and California Island lineages of Madiinae, first suggested by Carlquist, may reflect characteristics of tarweeds that facilitate insular colonization and adaptive radiation.
这里报道的核 rDNA 转录间隔区的系统发育分析和种间杂种的细胞遗传学研究支持卡尔奎斯特的假说(1965 年,岛屿生物学),即墨西哥瓜达卢佩岛的灌木状 Tarweeds(Deinandra)是加利福尼亚群岛原地辐射的产物,在那里植物多样化的证据一直存在争议。基于 rDNA 的发现,瓜达卢佩岛特有种(D. frutescens、D. greeneana subsp. greeneana 和 D. palmeri)构成了一个自上新世晚期以来出现的分支,远在瓜达卢佩岛的起源和 Deinandra 的大陆加利福尼亚谱系的多样化之后。三个特有种之间的 F1 杂种高度可育和正常减数分裂与其他大多数大陆加利福尼亚种的 Deinandra 之间的 F1 杂种的可育性降低(完全不育)和减数分裂不规则形成鲜明对比。克隆的 rDNA 序列没有提供有说服力的证据表明在瓜达卢佩岛的 Deinandras 之间发生了基因渗入;这些类群之间的形态、物候和/或生境差异表明存在基因流的生态障碍,并且可能在多样化中发挥了生态分歧的作用。瓜达卢佩岛和另一个次生木本、海洋岛屿 Tarweed 谱系夏威夷银剑联盟的灌木状 Tarweeds 的生物系统学和分子系统发育数据揭示了惊人相似的进化历史。这两个群体都违反了贝克规则,因为它们源自北美西部的自交不亲和祖先,并且每个群体都在岛内发生了多样化,而没有在谱系之间进化出强大的不育障碍。卡尔奎斯特首次提出的夏威夷和加利福尼亚岛 Madiinae 谱系之间的进化平行可能反映了 Tarweeds 促进岛屿殖民和适应性辐射的特征。