Polymorphism in Corbicula fluminea (Bivalvia: Corbiculacea) from North America

Britton, Joseph; Morton, Brian

Populations of Corbicula, an Asiatic clam, were evaluated from populations in Texas, Montezuma Well, Arizona, and Hong Kong. Possibly, more than one species of this clam appears in North America, one of them, the type found in Montezuma Well, has a slightly narrower body and darkly pigmented shell exterior. Evidence has recently appeared (Fontanier, 1982; Hillis and Patton, 1982; McLeod, 1984) suggesting that more than one species of Corbicula occurs in North America. In addition to the typical C. fluminea, characterized herein (Form A), there is a second Corbicula morphotype (Form B) characterized by dark periostracum and a darkly pigmented shell interior, more closely spaced concentric sulcations in young specimens, and a slightly narrower body. We found Form B in North America usually restricted to permanently flowing, oligotrophic, calcium and carbonate-rich, often spring-fed and travertine depositing streams in arid, warm temperate environments. A similar morphology is also known to occur in C. fluminea from flooded furrows of Hong Kong vegetable gardens. Population and reproductive studies of Forms A and B were initiated at two Texas localities, the Brazos River near Granbury, Texas, consisting exclusively of Form A, and Pinto Creek, Kinney Co., Texas, consisting of both morphotypes. Additional populations from Phoenix and Montezuma Well National Monument, Arizona and Hong Kong provided additional data. Sulcation spacing on the shell exterior serves to differentiate Form A and B morphotypes less than 15 mm in length. The distance between sulcations on Form B shells gradually increases with shell size, and becomes unreliable for morphotype discrimination on specimens larger than 15 mm in length. Arcsin transformed shell dimension ratios (calculated from length, height, and width measurements) demonstrate low within population variability, but moderate to considerable difference between populations separated by space or time. Differences are most apparent between Form A and B morphotypes, but are also commonly significant (p < .05) among Form A or B comparisons. Large samples of sympatrically distributed Form A and B morphotypes displaying significantly different arcsin transformed mean ratios nevertheless also display considerable overlap with respect to these ratios. These data support the concept of self-fertilization as an important reproductive mechanism (producing low variability within populations). Because of at least some outbreeding, however, differences between populations with low internal variability are more apparent than would be if each population displayed somewhat greater variability. Under these conditions, shell dimension comparisons are unreliable for taxonomic discrimination in Corbicula. The relatively consistent habitat relationship with Form B Corbicula, the presence of morphologies intermediate between Forms A and B when some but not all of the conditions normally associated with Form B are present, and the virtual absence of Form B in localities where most of the conditions normally associated with this morphotype are all strong indications that the Form B morphology is a habitat-specific response derived from the predominant (Form A) North American phenotype. In addition to differences in shell morphology, Form A and B also display differences in sexual expression which may be an effect of habitat. The tendency for C. fluminea to develop different reproductive strategies in different environments may help reinforce phenotypic differences between morphotypes. We conclude that there is still insufficient evidence to justify a two species model for North American Corbicula. The preponderance of evidence supports the view that a single highly variable species, C. fluminea, is the only introduced corbiculid in North America.