MORPHOLOGICAL VARIATION AMONG THE MEXICAN LAMPREYS (PETROMYZONTIDAE: LAMPETRA: SUBGENUS TETRAPLEURODON)

We used univariate and multivariate statistical analyses of 16 meristic and morphomet- ric variables to compare the morphology of three populations of Mexican lampreys, one from Lago de Chapala for the endangered parasitic species Lampetra spadicea, and one each from the Rio Duero and the Rio Grande de Morelia drainages for the non-parasitic Lampetra geminis, a species of special concern. Previous studies had suggested that the two non-parasitic populations might represent sepa- rate taxa. For 11 morphometric variables, differences between L. spadicea and the two populations of L. geminis were evident. Lampetra spadicea was much larger overall, had proportionally shorter first dorsal fin height, body depth, and branchial, caudal, and post-orbital lengths, and had a proportion- ally longer eye length. The relationship between total length and pre-branchial, pre-orbital, and disk lengths also differed between L. spadicea and L. geminis. In principal components analysis (PCA), L. spadicea was clearly separate from L. geminis along the first principal component (related to size dif- ferences), but overlapped on the next two (related to shape). Lampetra spadicea also had much lower variation in scores on the second and third principal components than did L. geminis. There were no significant differences between the Rio Duero and Rio Grande de Morelia populations ofL. geminis in morphometric variables, and the two populations overlapped completely along the three principal com- ponents. For each of five meristic variables, L. spadicea had greater values than the Rio Duero popu- lation of L. geminis. In PCA, these two populations showed no overlap on the first principal compo- nent, but substantial overlap on the next three components. The Rio Grande de Morelia L. geminis population had intermediate values for all five variables and for scores on the first principal compo- nent, but had greater values than the Rio Duero L. geminis population only for numbers of lateral circumoral teeth and cusps. We conclude that L. spadicea is morphologically distinct from L. geminis in several respects, but that the differences we have found between the two populations of L. geminis are too small to warrant separate taxonomic status based on morphology alone.