McAdam Andrew G, Boutin Stan
Department of Biological Sciences, University of Alberta, Edmonton, Alberta T6G 2E9, Canada.
Evolution. 2003 Jul;57(7):1689-97. doi: 10.1111/j.0014-3820.2003.tb00374.x.
Selection will result in observable changes in traits only if it acts consistently in space and time, but few estimates of selection in natural populations have been temporally replicated. Here we estimate viability selection on nestling growth rates for 13 cohorts (1989-2001) of red squirrels (Tamiasciurus hudsonicus) from a natural population located in southwestern Yukon, Canada. Directional selection on nestling growth rates varied in magnitude and direction from one cohort to the next. The magnitude of directional selection was relatively weak in most years (median beta' = 0.24), but there were episodes of very strong viability selection (beta' > 0.5) in some cohorts. We found no evidence of significant stabilizing or disruptive selection on this trait. Examination of viability selection episodes over shorter time periods suggested that the strength of selection on juveniles in this population was positively related to the time scale over which selection was measured. Viability selection from birth to emergence from the natal nest (50 days of age) and from emergence to successful recruitment (100 days of age) were positively correlated, but were both independent of selection on nestling growth rates from recruitment to potential breeding age (one year). The strength of directional selection on growth rates prior to recruitment was negatively correlated with spring temperature whereas selection from recruitment to breeding was positively correlated with the abundance of spruce cones produced in the previous fall. Episodes of strong directional selection from birth to breeding age appear to be due to potentially rare combinations of environmental conditions. As a result, predicting the occurrence of very strong episodes of selection will be extremely difficult, but predicting the microevolutionary responses to observed selection on individual cohorts remains feasible.
只有当选择在空间和时间上持续起作用时,才会导致性状发生可观察到的变化,但对自然种群中的选择进行的时间重复估计很少。在这里,我们估计了来自加拿大育空地区西南部一个自然种群的13个红松鼠(北美红松鼠)队列(1989 - 2001年)雏鸟生长率的生存选择。雏鸟生长率的定向选择在不同队列之间在大小和方向上有所不同。在大多数年份,定向选择的强度相对较弱(中位数β' = 0.24),但在一些队列中存在非常强烈的生存选择事件(β' > 0.5)。我们没有发现对该性状有显著稳定或破坏选择的证据。对较短时间段内生存选择事件的研究表明,该种群中对幼体的选择强度与测量选择的时间尺度呈正相关。从出生到离开出生地巢穴(50日龄)以及从离开巢穴到成功招募(100日龄)的生存选择是正相关的,但两者都与从招募到潜在繁殖年龄(一年)的雏鸟生长率选择无关。招募前生长率的定向选择强度与春季温度呈负相关,而从招募到繁殖的选择与前一年秋季产生的云杉球果丰度呈正相关。从出生到繁殖年龄的强烈定向选择事件似乎是由于环境条件潜在的罕见组合。因此,预测非常强烈的选择事件的发生将极其困难,但预测对各个队列中观察到的选择的微进化反应仍然是可行的。