Delineation in fuzzy boundaries: Canonical ordination analysis discriminates between cryptic species by linking taxonomy to genetics, morphology, climate, and space

Abstract

In species complexes, delineation of cryptic species remains a major challenge due to their morphological similarities despite significant genetic divergence. The related problems of defining and recognizing hidden diversity impact a range of sciences, posing theoretical and practical problems. Our article introduces a novel application of canonical ordination analysis as a powerful tool for examining and testing the interplay between genetic and morphological variability in cryptic species while accounting for their relation to climate and spatial descriptors, aiming to discriminate them by integrating multiple sources of predictors. We used data on 60 Fruticicola sp. populations across its range, belonging to three cryptic species—Fruticicola fruticum, F. similis, and F. gemina. We used five variable categories: taxonomy (response variables), genetics and morphology (response variables and predictors), climate and space (predictors). We applied distance-based redundancy analysis and canonical correspondence analysis to examine and test relationships among these variable categories, variation partitioning procedure to disentangle the effects of the considered predictors, and linear discriminant analysis to test their discriminatory power. Morphology was best explained by climate (mainly humidity), whereas genetic distances showed patterns shared between climate and space. Because the three pseudocryptic species were defined on molecular basis, taxonomy was almost completely explained by genetic distances. Although, when considered alone, morphology, climate, and space did not perform well in discriminating the species, when included together in the models, they were able to correctly classify the samples. We demonstrate how this multivariate canonical approach enhances species delimitation, offering a clearer understanding of cryptic diversity and its ecological implications. By linking these facets, we provide a comprehensive framework that connects taxonomic classifications with ecological and evolutionary processes. Hence, our results bring more insight into the processes linked to hidden diversity while providing new tools for its assessment, broadening the framework for applied research.