Summary Sagittal otoliths from the Atlantic and Mediterranean regions of Serranus cabrilla L. were compared using shape indices (form‐factor, roundness, circularity, rectangularity, ellipticity and eccentricity). Regional differences were best described by form‐factor, circularity and eccentricity variables. The canonical discriminant functions were built with form‐factor or rectangularity indices. The results indicated slight regional variations. The G‐test was the criterion more useful to identify the otolith origin in the discriminant analysis.

Variations in the morphology of saccular otoliths (sagittae) among three sympatric species of the genus Serranus (S. atricauda, S. cabrilla and S. scriba) from the Canary Islands were investigated. Although the otolith gross morphology was similar among species, S. scriba was distinct in having a rostrum which had a slight turning at the tip and a more funnel-like ostium. The shallower water species (S. scriba) had otolith and sulcus areas which were smaller than the deeper water species (S. cabrilla and S. atricauda). The sulcus acusticus and ostium size were correlated with the habit depth of the species, with the highest values in the deepest species, S. cabrilla. The otolith outline shape indices changed with size (total length) of the species, and allowed the separation of the species by means of a discriminate function.

Biological invasions have become major players in the current biodiversity crisis, but realistic tools to predict which species will establish successful populations are still unavailable. Here we present a novel approach that requires only a morphometric characterisation of the species. Using fish invasions of the Mediterranean, we show that the abundance of non-indigenous fishes correlates with the location and relative size of occupied morphological space within the receiving pool of species. Those invaders that established abundant populations tended to be added outside or at the margins of the receiving morphospace, whereas non-indigenous species morphologically similar to resident ones failed to develop large populations or even to establish themselves, probably because the available ecological niches were already occupied. Accepting that morphology is a proxy for a species' ecological position in a community, our findings are consistent with ideas advanced since Darwin's naturalisation hypothesis and provide a new warning signal to identify invaders and to recognise vulnerable communities.