Paper No. 25
Presentation Time: 9:00 AM-6:00 PM

MORPHOLOGICAL VARIABILITY OF ASTARTE BOREALIS (MOLLUSCA, BIVALVIA) IN THE CAMDEN BAY, NORTHERN ALASKA


CHRPA, Michelle E., Department of Geosciences, Florida Atlantic University, 777 Glades Rd, Science and Engineering Building, Boca Raton, FL 33431 and OLEINIK, Anton, Geosciences, Florida Atlantic University, 777 Glades Road, Boca Raton, FL 33431-0991, mchrpa@fau.edu

The genus Astarte is well known for variable shell morphology and polymorphism within living and fossil species. Astarte borealis, the most common living species, has been divided into many subspecies and varieties merely based on overall shell shape. The A. borealis is easily recognizable and common among mid to high latitude North Pacific, Arctic Ocean and North Atlantic waters making the determination of species varieties a complex issue. Ascertaining the morphological variability based on shell shape within an A. borealis population may reduce the use of subjective interpolation of potential subspecies or varieties within the species. A collection of A. borealis specimens from the Camden Bay, Alaska yielding 635 specimens with outline intact were used in this study to find shell shape variability. Morphometric analysis of the shell outline seeks to determine the variants within a population of A. borealis. The computer program package SHAPE uses elliptic Fourier descriptor coefficients of the A. borealis shell outline to evaluate and visualize the shape variation among the sample specimens. The principal component analysis of the coefficients showed that the majority of the variation was summarized by 5 components; the asymmetrical and symmetrical variations had cumulative contributions of 83.1% and 90.0% respectively. The overall coefficient analysis showed the 1st principal component accounted for 38.3% of the variance related to the aspect ratio of the shell, ranging from 1.07:1 to 1.23:1. The 2nd principal component relating to the overall shell roundness from trigonal to subrounded explains 15.9% variance and the 3rd principal component relating to the convexness of the ventral margin explains 9.0% variance.