Taming extreme morphological variability through coupling of molecular phylogeny and quantitative phenotype analysis as a new avenue for taxonomy.

Identification of animals is often hindered by decoupling of phenotypic and molecular evolutionary rates. The Acanthocyclops vernalis (Fischer, 1853) complex is arguably the most problematic group of cyclopoids and possibly of all copepods, with diversity estimates based on morphology ranging from 2 to 34 taxa. We reconstructed their phylogeny based on one nuclear and three mitochondrial markers, revealing only four species in the Holarctic and always the following sister-species pairs: vernalis-europensis sp. nov. and robustus-americanus. Landmarks for quantitative shape analyses were collected from 147 specimens on five structures commonly used to delineate cyclopoids. Procrustes ANOVA showed small directional asymmetry in all datasets, but large sexual dimorphism in shape and size. Allometry was also highly significant. Principal component analyses of size-corrected data almost completely separated species in morphospace based on the last exopodal and endopodal segments of the fourth leg. These two structures showed the highest amount of covariation, while modularity could not be proven and a phylogenetic signal was only observed in one structure. Spinules and sensilla have a limited use in delineating species here. Calculating mean shapes and the extent of inter and intraspecific phenotypic variability opens new horizons for modern taxonomy.