Ecotone effect on assemblage-level tip-based metrics
While we expected higher diet transition rates, lower stasis time, and shorter transition times in ecotone than core assemblages, we found that, relative to core species, ecotone species presented 1) fewer transitions in diet over their evolutionary history, 2) quicker transition periods, and 3) slightly longer retention of the current diet.
The existence of patches of favorable habitat can prevent evolution at ecotone zones. Patches of favorable, high-quality habitat can be ephemeral and sparsely distributed along ecotones, but can sustain large population sizes with individuals presenting little or no shifts in ecological, morphological and behavioral characters over time (Eckert et al. 2008; Sexton et al. 2009). These processes result in few transitions from ancestral diet because the retention of an optimal feeding strategy enables species persistence in ecotone zones. This strategy could be a generalist diet that generally evolves as an option to explore resources from different habitats (Price et al. 2012). Also, patches of favorable habitat along ecotones can provide the stability needed to maintain the current diet since long time ago, perhaps since late Miocene or early Pliocene when major sigmodontine tribes diversified (Leite et al. 2014; Steppan and Schenk 2017) and within-clade morphological disparity increased (Maestri et al. 2017).
Shorter stasis time at ecotone assemblages indicates that trait evolution occurred at more regular periods of time along the evolutionary history of ecotone species. Speciation along regular periods prevent the accumulation of time between transition events, and can be produced by regular cycles of environment change which are first noticed by ecotone species (de Vivo and Carmignotto 2004; Karanth et al. 2006; Donoghue and Edwards 2014). As ecotones buffer environmental changes, there may be thousands of years of lag between the beginning of environmental changes and modifications of species traits. Although the difference we found here seems to be subtle (Figs. 2-4), it represents thousands of years of lag that may have profound influence on species persistence and trait evolution.
We found a stasis time of around 2.5 ma for both core and ecotone species. It is a long time period under little to no trait evolution relative to the ~10 ma of sigmodontine presence in the Neotropics. Although we do not know the exact geological period in which diet stasis occurred, cooling periods such as the one embracing late Miocene and early Pliocene (Amidon et al. 2017) may well have facilitated diet retention over large time periods. Cooling periods through the Cenozoic are related to speciation slowdowns across major tetrapod clades, likely due to the influence of temperature on the environment’s carrying capacity (Condamine et al. 2019).