ECOLOGICAL AND EVOLUTIONARY DRIVERS OF TEMPORAL VARIATIONS IN BODY SIZE

Body size, being well accessible in fossils, has been a research focus for generations of paleontologists. Most studies focus on long-term evolutionary trends within or across lineages or on short-term variations in the context of evolutionary crises. Here we address, based on a compilation of more than 6000 interval-to-interval changes of body size in marine taxa, how the oft-cited Lilliput effect compares across spatio-temporal scales. To assess body-size changes, we used the log-ratio of the geometric means of body-size estimates from one interval to the next. This measure of effect size is theoretically independent of absolute body size and symmetrical for losses and gains.

There is a tendency of effect size and variance to increase with temporal and taxonomic scale suggesting that changes in body size are largely manifested in macroevolutionary rather than phenotypic processes. Some higher taxa such as crinoids and orthoconic nautilids appear to be more prone to body size changes than others, but sampling issues prevent a conclusive statement. Effect sizes are weakly but significantly correlated with body size suggesting that larger organisms are more prone to body size change than smaller organisms.

To test if the Lilliput effect is significantly linked to environmental perturbations or rather deceived by the focus of pertinent studies on mass extinctions, we compared the large dataset of pure background changes with those of inferred background to crises changes. Negative changes of body size are significantly more pronounced during changes from background to crises intervals than in any other combination of the background-crisis-survival-recovery intervals. This maintains that the Lilliput effect in the broadest sense is a common phenomenon during evolutionary crises. The effect seems to be stronger during hypothermals (End-Permian, Pliensbachian-Toarcian crisis) than during hypothermals (end-Ordovician, end-Devonian, end-Cretaceous). However, dwarfing cannot necessarily be taken as evidence of warming-induced crises, since the underlying mechanisms for these differences still needs to be further studied.