2009 Portland GSA Annual Meeting (18-21 October 2009)

Paper No. 9
Presentation Time: 3:30 PM

MORPHOLOGICAL VARIABILITY WITHIN AND BETWEEN TWO RECENT RHYNCHONELLIDE (BRACHIOPODA) SPECIES AS MODELED BY ELLIPTICAL FOURIER ANALYSIS


SCHREIBER, Holly A.1, CARLSON, Sandra J.1, FITZGERALD, Paul C.2 and ROOPNARINE, Peter D.3, (1)Department of Geology, University of California, Davis, One Shields Ave, Davis, CA 95616, (2)Department of Biology, Northern Virginia Community College, 8333 Little River Turnpike, Annandale, VA 22003, (3)Invertebrate Zoology & Geology, California Academy of Sciences, 55 Concourse Dr, Golden Gate Park, San Francisco, CA 94118, haschultz@ucdavis.edu

The range of morphological variability within and between species of fossil and extant rhynchonellide brachiopods is poorly known. To evaluate the boundaries of morphological species in the fossil record, it is necessary to quantify variability within and among Recent species. Rhynchonellides, an evolutionarily important but neglected group, have not been analyzed using modern morphometric techniques. We used Elliptical Fourier Analysis (EFA) to examine interspecific and intraspecific variation among dorsal valve commissural and valve convexity outlines from two common Recent rhynchonellide species. Our objective is to characterize the variability present in extant rhynchonellides to establish guidelines for use in rhynchonellide classification that then can be applied to the fossil record. This study includes a total of 142 Hemithiris psittacea specimens from three samples collected from Alaska, Labrador, and Greenland and 146 Notosaria nigricans specimens collected from three localities around New Zealand. EFA simplifies the shell outline into fewer shape variables by using digitized outline data points to calculate the harmonic coefficients that are needed to complete a series of sine and cosine curves to reproduce the original outlines. The harmonic coefficients are then used as input variables in multivariate statistical analyses, using a covariance matrix. Prior to the EFA, all shell outlines were translated to be invariant to starting position, location, rotation, and size to capture true morphological variation, not variation due to errors in outline alignment. EFA was also completed without size standardization to test for size-related shape differences between groups. We predicted that populations within each species would differ significantly in their shape due to their broad geographic distribution. However, preliminary results of the Principal Component Analyses and Canonical Variate Analyses reveal considerable overlap among populations of each species, while there is clear separation between species. Initial analyses also indicate that H. psittacea and N. nigricans each occupy a similar amount of morphospace, exhibiting comparable morphological variability. Results of this study will be used to evaluate fossil rhynchonellide species boundaries in future analyses.