HOW ACCURATELY CAN RETENTION OF BENTHIC MACROFAUNA BY A PARTICULAR MESH SIZE BE PREDICTED FROM BODY-SIZE OF ORGANISMS

Size of organisms is frequently the prime criterion in selecting a mes h size to sample benthic assemblages. This study quantified the accura cy in estimating the sampling efficiency of screens from body size of macrozoobenthos in the upper, sandy, reaches of a small, shallow estua ry, where the community consisted primarily of peracarid crustaceans a nd polychaetes. Body size of organisms retained by a 0.25, 0.5 and 1.0 mm screen was used to predict the retention efficiency of each gear b y multiple discriminant analysis (MDA), or by simply assuming that an animal of known size will be retained by a screen of the same aperture size ('body aperture match'). MDA-models classified about 70% of all cases correctly, whereas matching of body- to screen-size gave at best spurious, and often seriously wrong estimates of retention efficiency for all tested mesh sizes. Regardless of the method employed, body si ze was found to be an extremely poor predictor of mesh retention. Cons equently, sampling performance of each mesh size in a particular habit at x community combination should be determined experimentally and not from body size measurements obtained during pilot studies.