Shell-shape variation in the Chilean blue mussel Mytilus chilensis was examined in eight populations covering the totality of its distribution range, which represents over 1800 km of latitudinal gradient in the southeastern Paci¢c. The shell outline shapes were assessed using elliptic Fourier analysis followed by canonical variates analysis. Fourier coe⁄cients showed a highly signi¢cant morphological variation between the populations studied. Canonical variates analysis showed a visual association of the First canonical with convexity of ventral edge and umbo shape, while the second axis was associated with shell elongation. Previously described genetic patterns were marginally congruent with our morphometric pattern, though geographic distance had a stronger e¡ect on morphology. Shape change (elongation) was correlated with latitudinal clinal variation. The morphometrical analysis proved to be an important tool for evaluating the components of shell-shape variation in mussels and to document association patterns with geographical and ecological variables. Such patterns are useful to establish awide ¢eld of work, including sexing, stock recognition, traceability studies and productive management in general.FONDECYT-Chile through project 1050848

Shell thickness in mollusks is generally considered adaptive because of their effects on fitness. However, little is known about the genetic basis of shell thickness. This is important, because the response to selection and the subsequent adaptive microevolution of a trait, such as thickness is only possible when that trait exhibits additive genetic variation. Here, we estimated the narrow-sense heritability (h 2 : ratio between additive genetic variance and phenotypic variance) for the traits 'shell thickness' and 'shell length growth' in a 34-month-old cohort of the mussel Mytilus chilensis obtained by using a half-full sib design and grown in the field. Also, phenotypic and genetic correlations were estimated between both traits. We found that h 2 showed significant values for shell thickness (0.294 AE 0.194) and length (0.731 AE 0.379). The phenotypic correlation between both traits was positive and significant; however, the genetic correlation between these traits was not. These results suggest both traits can evolve adaptively by selection, but because these traits did not show genetic correlation, it is possible that selection pressure affecting one trait may not affect the other.

No abstract