Geometric morphometric analysis of geographic variation in the Río Negro tuco-tuco, Ctenomys rionegrensis (Rodentia: Ctenomyidae)Geometrisch morphometrische Analyse geografischer Variationen des Río Negro Tuco-Tuco Ctenomys rionegrensis (Rodentia: Ctenomyidae)

A fundamental issue in the study of intraspecific variation is the analysis of how it is allocated within and between local populations. The genus Ctenomys represents an excellent model for studying such kinds of phenomena, and Ctenomys ríonegrensis is particularly interesting for the study of the processes driving geographic differentiation. This species occupies a relatively small area in Uruguay, where it is restricted to sand dunes. Throughout its distribution, these “tuco-tucos” show conspicuous phenotypic variation in the form of three easily distinguishable pelage colours: melanic, agouti and dark-backed. These observations contrast with the usual correlation between pelage and substrate colour found among small mammals, including other tuco-tucos and subterranean rodents on other continents. Different hypotheses have been advanced to explain the population structure pattern found in this tuco-tuco, including the random fixation of alternative pelage colours by drift and the possible role of an unknown type of selection to compensate possible predation pressures against melanic individuals. We analyse the geographic variation in the cranial shape of C. rionegrensis along its distribution range, using geometric morphometric techniques. Thirteen landmarks, assumed to be homologous among all specimens analysed, were chosen in each of three views of the skull. The discriminant function analyses performed over the partial warps and uniform components scores matrix do not discriminate among different pelage types regardless of the view of the skull considered. In contrast, the discriminant function showed a better separation among local populations. No correlation was found between geographic and morphological distances, or between genetic and morphological distances, among pairs of popualtions studied. This geometric morphometric approach had proved to be sensitive enough to detect a similar pattern of population structure that the molecular markers employed before.