Limits...
North Andean origin and diversification of the largest ithomiine butterfly genus

View Article: PubMed Central - PubMed

ABSTRACT

The Neotropics harbour the most diverse flora and fauna on Earth. The Andes are a major centre of diversification and source of diversity for adjacent areas in plants and vertebrates, but studies on insects remain scarce, even though they constitute the largest fraction of terrestrial biodiversity. Here, we combine molecular and morphological characters to generate a dated phylogeny of the butterfly genus Pteronymia (Nymphalidae: Danainae), which we use to infer spatial, elevational and temporal diversification patterns. We first propose six taxonomic changes that raise the generic species total to 53, making Pteronymia the most diverse genus of the tribe Ithomiini. Our biogeographic reconstruction shows that Pteronymia originated in the Northern Andes, where it diversified extensively. Some lineages colonized lowlands and adjacent montane areas, but diversification in those areas remained scarce. The recent colonization of lowland areas was reflected by an increase in the rate of evolution of species’ elevational ranges towards present. By contrast, speciation rate decelerated with time, with no extinction. The geological history of the Andes and adjacent regions have likely contributed to Pteronymia diversification by providing compartmentalized habitats and an array of biotic and abiotic conditions, and by limiting dispersal between some areas while promoting interchange across others.

No MeSH data available.


Ancestral reconstruction of the mean and boundaries of the 95% elevational range (left: lower boundary, middle: upper boundary, right: mean).For the lower and upper boundaries of the elevational range, trees were rescaled according to the δ value inferred (Table 2). The figure was generated with R (https://cran.r-project.org/).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5384087&req=5

f4: Ancestral reconstruction of the mean and boundaries of the 95% elevational range (left: lower boundary, middle: upper boundary, right: mean).For the lower and upper boundaries of the elevational range, trees were rescaled according to the δ value inferred (Table 2). The figure was generated with R (https://cran.r-project.org/).

Mentions: We investigated the evolution of the elevational range and mean elevation of Pteronymia species on the MCC tree and on 100 trees from the posterior distribution. The phylogenetic signal and the tempo of evolution of the elevational range and mean elevation were assessed by estimating simultaneously the maximum likelihood values of the λ and δ branch scaling parameters47, respectively. A λ value of one indicates that the phylogeny correctly represents the trait covariance among species (Brownian motion model of evolution), while λ < 1 indicates that the phylogeny overestimates the trait covariance among species. A δ value of one means that the trait evolves at a constant pace along branches of the tree; δ < 1 indicates early changes in the character values followed by a slowing down of the evolution rate; while δ > 1 indicates accelerated evolution rate and species-specific adaptation. For the mean elevation and the elevational range, estimates of λ across all trees from the posterior distribution and for the MCC tree were not significantly different from one, meaning that trait evolution does not differ from a Brownian motion model. Estimates of δ for the mean elevation were also not significantly different from one (Table 3). For the lower and upper boundaries of the elevational range, the estimates of δ were significantly higher than one in >50% of the trees of the posterior distribution (Table 3).The MCC tree showed significantly higher estimates of δ for the lower boundary of the elevational range, but only marginally significant for the upper boundary (Table 3). Values of δ higher than one indicate an acceleration of the rate of evolution of elevation range and species elevational specialisation. Results differ between the mean elevation and elevational range perhaps because the mean is less variable than the range (e.g., species with different elevational ranges may have similar elevation mean). Reconstructions of ancestral mean elevation and range boundaries accounting for the inferred δ values are depicted on Fig. 4.


North Andean origin and diversification of the largest ithomiine butterfly genus
Ancestral reconstruction of the mean and boundaries of the 95% elevational range (left: lower boundary, middle: upper boundary, right: mean).For the lower and upper boundaries of the elevational range, trees were rescaled according to the δ value inferred (Table 2). The figure was generated with R (https://cran.r-project.org/).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5384087&req=5

f4: Ancestral reconstruction of the mean and boundaries of the 95% elevational range (left: lower boundary, middle: upper boundary, right: mean).For the lower and upper boundaries of the elevational range, trees were rescaled according to the δ value inferred (Table 2). The figure was generated with R (https://cran.r-project.org/).
Mentions: We investigated the evolution of the elevational range and mean elevation of Pteronymia species on the MCC tree and on 100 trees from the posterior distribution. The phylogenetic signal and the tempo of evolution of the elevational range and mean elevation were assessed by estimating simultaneously the maximum likelihood values of the λ and δ branch scaling parameters47, respectively. A λ value of one indicates that the phylogeny correctly represents the trait covariance among species (Brownian motion model of evolution), while λ < 1 indicates that the phylogeny overestimates the trait covariance among species. A δ value of one means that the trait evolves at a constant pace along branches of the tree; δ < 1 indicates early changes in the character values followed by a slowing down of the evolution rate; while δ > 1 indicates accelerated evolution rate and species-specific adaptation. For the mean elevation and the elevational range, estimates of λ across all trees from the posterior distribution and for the MCC tree were not significantly different from one, meaning that trait evolution does not differ from a Brownian motion model. Estimates of δ for the mean elevation were also not significantly different from one (Table 3). For the lower and upper boundaries of the elevational range, the estimates of δ were significantly higher than one in >50% of the trees of the posterior distribution (Table 3).The MCC tree showed significantly higher estimates of δ for the lower boundary of the elevational range, but only marginally significant for the upper boundary (Table 3). Values of δ higher than one indicate an acceleration of the rate of evolution of elevation range and species elevational specialisation. Results differ between the mean elevation and elevational range perhaps because the mean is less variable than the range (e.g., species with different elevational ranges may have similar elevation mean). Reconstructions of ancestral mean elevation and range boundaries accounting for the inferred δ values are depicted on Fig. 4.

View Article: PubMed Central - PubMed

ABSTRACT

The Neotropics harbour the most diverse flora and fauna on Earth. The Andes are a major centre of diversification and source of diversity for adjacent areas in plants and vertebrates, but studies on insects remain scarce, even though they constitute the largest fraction of terrestrial biodiversity. Here, we combine molecular and morphological characters to generate a dated phylogeny of the butterfly genus Pteronymia (Nymphalidae: Danainae), which we use to infer spatial, elevational and temporal diversification patterns. We first propose six taxonomic changes that raise the generic species total to 53, making Pteronymia the most diverse genus of the tribe Ithomiini. Our biogeographic reconstruction shows that Pteronymia originated in the Northern Andes, where it diversified extensively. Some lineages colonized lowlands and adjacent montane areas, but diversification in those areas remained scarce. The recent colonization of lowland areas was reflected by an increase in the rate of evolution of species&rsquo; elevational ranges towards present. By contrast, speciation rate decelerated with time, with no extinction. The geological history of the Andes and adjacent regions have likely contributed to Pteronymia diversification by providing compartmentalized habitats and an array of biotic and abiotic conditions, and by limiting dispersal between some areas while promoting interchange across others.

No MeSH data available.