Bokma Folmer
Department of Biology, University of Oulu, PO Box 3000, FIN-90014 Oulu, Finland.
Evolution. 2002 Dec;56(12):2499-504. doi: 10.1111/j.0014-3820.2002.tb00174.x.
Of the approximately 9500 bird species, the vast majority is small-bodied. That is a general feature of evolutionary lineages, also observed for instance in mammals and plants. The avian interspecific body size distribution is right-skewed even on a logarithmic scale. That has previously been interpreted as evidence that body size evolution has been biased. However, a procedure to test for unbiased evolution from the shape of body size distributions was lacking. In the present paper unbiased body size evolution is defined precisely, and a statistical test is developed based on Monte Carlo simulation of unbiased evolution. Application of the test to birds suggests that it is highly unlikely that avian body size evolution has been unbiased as defined. Several possible explanations for this result are discussed. A plausible explanation is that the general model of unbiased evolution assumes that population size and generation time do not affect the evolutionary variability of body size; that is, that micro- and macroevolution are decoupled, which theory suggests is not likely to be the case.
在大约9500种鸟类中，绝大多数体型较小。这是进化谱系的一个普遍特征，在哺乳动物和植物等类群中也有观察到。即使在对数尺度上，鸟类种间体型分布也是右偏的。这一点以前被解释为体型进化存在偏差的证据。然而，缺乏一种从体型分布形状来检验无偏差进化的方法。在本文中，精确地定义了无偏差体型进化，并基于无偏差进化的蒙特卡洛模拟开发了一种统计检验方法。将该检验应用于鸟类表明，按照所定义的无偏差进化，鸟类体型进化几乎不可能是无偏差的。文中讨论了这一结果的几种可能解释。一个合理的解释是，无偏差进化的一般模型假定种群大小和世代时间不会影响体型的进化变异性；也就是说，微观进化和宏观进化是解耦的，而理论表明情况不太可能如此。
