Björklund Mats
Museum of Comparative Zoology, Harvard University, Cambridge, MA, 02138, USA.
Evolution. 1991 May;45(3):608-621. doi: 10.1111/j.1558-5646.1991.tb04332.x.
According to theory, two consequences of sexual selection are sexual dimorphism in size and secondary sexual characteristics, due to either intra- or intersexual selection. In this paper I suggest three criteria for the test of an evolutionary hypothesis involving quantitative morphological characters. First, the postulated change must be shown to have occurred in evolutionary time. Second, this change must be positively correlated with a change in the proposed selective agent. Third, given two taxa with different degrees of sexual size dimorphism and different mating system, the possible influence of drift must be rejected. If the hypothesis is not rejected by these three criteria, then we still have no proof of causality, but we can at least be more confident about its plausibility. This is applied to the particular hypothesis that sexual dimorphism in the Boat-tailed and Great-tailed grackles (Quiscalus spp; Icterinae; Aves) is caused by the highly polygynous mating system in these species. In relation to an outgroup, both species have increased disproportionately in male tarsus and tail size, creating an increased sexual dimorphism. This has cooccurred with the evolution of their particular mating system. However, the variance among species in male tarsus size can be accounted for by drift, and need not be a result of selection for increased size. In contrast, the variance among species in male tail size was much larger than expected under a null model of drift, indicating directional selection for long tails. The variance in female tail size was not larger than expected by drift, whereas the variance in female tarsus size was in fact lower than expected by drift, indicating stabilizing selection. The data are consistent with the hypothesis with regard to tail size, but not with regard to body size.
根据理论，性选择的两个后果是体型上的两性异形和第二性征，这是由种内或种间选择导致的。在本文中，我提出了三个标准来检验一个涉及定量形态特征的进化假说。首先，必须证明所假定的变化在进化时间内已经发生。其次，这种变化必须与所提出的选择因子的变化呈正相关。第三，对于两个具有不同程度两性体型差异和不同交配系统的分类群，必须排除遗传漂变的可能影响。如果该假说没有被这三个标准所否定，那么我们仍然没有因果关系的证据，但我们至少可以对其合理性更有信心。这一方法应用于一个特定的假说，即舟形长尾拟八哥和大尾拟八哥（Quiscalus spp；拟黄鹂科；雀形目）的两性异形是由这些物种高度多配的交配系统所导致的。相对于一个外类群，这两个物种的雄性跗骨和尾羽尺寸都不成比例地增加，从而导致两性异形加剧。这与它们特定交配系统的进化同时发生。然而，雄性跗骨尺寸在物种间的差异可以用遗传漂变来解释，不一定是体型增大选择的结果。相比之下，雄性尾羽尺寸在物种间的差异比在遗传漂变的零模型下预期的要大得多，表明存在对长尾的定向选择。雌性尾羽尺寸的差异不大于遗传漂变预期的差异，而雌性跗骨尺寸的差异实际上低于遗传漂变预期的差异，表明存在稳定选择。这些数据在尾羽尺寸方面与该假说一致，但在体型方面则不一致。
