Keeping up with the flowering times: Examining correlated evolution between flowering phenology and morphological traits in bluebells (Mertensia spp., Boraginaceae)

Background/Question/Methods  Flowering phenology is an important aspect of plant adaptation and diversification due to its interactions with resource and pollinator availabilities. Fundamental trade-offs between growth and reproduction mean that across species, flowering phenology and vegetative size at flowering should be positively correlated (i.e., larger species should flower later). Meanwhile, seasonal changes in pollinator assemblages could result in pollinator-mediated selection on flowering phenology and floral traits, resulting in correlated evolution of phenology, floral traits, and size at flowering. In our study, we asked if flowering phenology is evolutionarily correlated with size at flowering and floral traits in the perennial herbaceous genus Mertensia (Boraginaceae). We visited populations of 12 Mertensia species in the central and southern Rocky Mountains and measured several size (plant height, maximum leaf area, and number of flowers) and floral traits (anther height, stigma height, stigma-anther separation, and flower volume). We also quantified flowering phenology of each species based on the phenological stage of the Mertensia population and on the presence of other flowering species occupying the same habitat. We tested for correlated evolution among the traits using phylogenetic principal component analyses and phylogenetic generalized least-squares analyses.

Results/Conclusions  We found a significant phylogenetic correlation between flowering phenology and size traits in Mertensia, with later-flowering species being taller and producing larger floral displays. This is in accordance with previous studies showing that smaller plant species require a lower threshold of vegetative size before the onset of flowering; therefore, larger species typically flower later than small species but are able to produce more flowers. We found no signal of significant correlated evolution between flowering phenology and floral traits, suggesting that these two traits did not evolve as correlated responses to similar mechanisms, such as pollinator-mediated selection. This may be because pollinators of Mertensia are generalists that do not exert strong selection on the floral traits measured in this study, or because pollinator assemblages are too variable to impose consistently different selection on early- and late-flowering species. Furthermore, we did not detect correlated evolution between size and floral traits, suggesting that selection pressures on these two traits did not co-occur across these Mertensia species. Our results show that diversification in flowering phenology within Mertensia has been associated with changes in plant size but has not necessitated concomitant changes in floral traits.