Recent years there has been published several studies that report temporal trends in animal body size: often decreases over time. The abundance of such literature leaves the overall impression that these changes are common across taxa, and might be due to climate change. Following Bergmann’s rule increasing temperatures should lead to decreases in body size. To study whether such common patterns take place, Meiri et al. investigated historical body size records in 22 carnivore species in 52 populations, but with no evidence for common temporal trends in body size. Few of the populations showed any significant trends and those that did either increased or decreased in body size in a way that could not be linked to species size, habitat, latitude or sampling intensity. The study may suffer from lack of power in detecting weak trends and be limited in the number of species and population. Nonetheless, it invokes the question of whether publication bias conveys false impressions about how common certain trends seen in wild actually are.
Meiri S, Guy D, Dayan T, Simberloff D (2009) Global change and carnivore body size: data are stasis. Global Ecology and Biogeography 18: 240-247.
3 Comments
Hm, sounds like it could be fun to throw a funnel plot at this: see if there’s any evidence for publication bias.
Well. There are many pretty solid studies showing that mean body size in many vertebrate populations has been declining over time. See e.g.
http://blogs.helsinki.fi/egru-blog/2008/08/18/birds-shrink-in-the-warming-climate-bergmann-rules-but-why/
I recall of seeing similar trends published for a whole bunch of Australian passerines recently. My concern would not be – at the first place – possibility of publication bias (why should one expect it?), but how to interpret the observed declines.
The point here is not to question the reported trends, but to ask how common it actually is that no clear trend was observed. The authors just make the point that such results are not often reported even though they do not seem entirely uncommon either. .