A combination of stabilizing selection and random walk are associated with phylogenetic signal in hard pines.

Abstract

Background and aims: Although phylogenetic niche conservatism is widely accepted, in many cases the cause of this conservatism is unclear. The presence of phylogenetic signal could extend to morphological and anatomical characters, chemical soil properties and DNA content. However, as sessile organisms, especially those that migrated toward drier or tropical conditions, are subject to environmental heterogeneity, phylogenetic signal in these characters could be explained by a combination of different evolutionary forces.

Methods: We assessed the phylogenetic signal in DNA content, chemical soil properties, climate and morphoanatomical characteristics, tested different evolutionary models and performed an ancestral state reconstruction in 30 species of hard pines geographically distributed in North America, Mexico, Central America and the Caribbean. To reinforce our hypothesis of phylogenetic niche conservatism, we applied a niche similarity test among different species pairs, performed ecological niche modelling and projected these models in geographical space.

Key results: We found strong phylogenetic signals in the characters evaluated, indicating a retention of characteristics throughout the evolutionary history of these pines. The best models to explain these phylogenetic signals were Ornstein-Uhlenbeck, Brownian Motion and Early Burst, indicating the action of stabilizing selection, with an input of random walk. The detection of niche overlap supported our hypothesis of phylogenetic niche conservatism; however, we found more similarity than expected in more phylogenetically distant species.

Conclusions: Overall, we detected strong phylogenetic signal, and our results supported the hypothesis of phylogenetic niche conservatism, but there was more similarity in some species that have evolved under similar selective pressures independent of phylogenetic relationships. No single evolutionary model fully explains trait divergence; depending on the specific trait, divergence could be explained primarily by stabilizing selection or random walk in these hard pines.