Preingestive selection of different microalgal mixtures in Crassostrea gigas and Mytilus edulis, analysed by flow cytometry.
The potential impact of selective grazing by filter-feeding bivalves on the relative composition of planktonic and benthic algae commonly suspended in coastal areas, was examined. Six experiments were conducted between September 1993 and February 1995 using C. gigas collected from the Bay of Marennes-Oléron (France), and in November 1994 using M. edulis collected from the Exe Estuary (England). Oysters were studied within static 20-litre tanks filled with 15 litres of filtered sea water enriched with an algal mixture of Pavlova lutheri, Skeletonema costatum and Tetraselmis suecica (diet D1). Mussels were placed in smaller 2-litre tanks containing 1 litre of filtered sea water enriched with the same algal mixture (diet D2). A flow-through system was used to study feeding behaviour using 3 different algal mixtures. Diet D3 consisted of P. lutheri, Chaetoceros calcitrans, Isochrysis galbana and T. suecica while diet D4 was similar to D3 but without T. suecica. A system of recirculating sea water was used to study feeding behaviour using 2 other algal mixtures. D5 contained P. lutheri, Nitzschia closterium and T. suecica, while P. lutheri was replaced by Navicula filata in D6. The specific composition of the each algal mixture was maintained within all systems by injecting specific volumes of the original algal mixture at regular time intervals. Feeding behaviour of mussels and oysters was studied in static systems at similar seston concentrations of 9.1 and 8.6 mg/litre, respectively. Mussels preferentially filtered (P<0.001) T. suecica (ratio of observed to theoretical filtration 1.22) compared with S. costatum (0.94) and P. lutheri (0.90). Oysters preferentially filtered (P<0.001) S. costatum (1.36) and T. suecica (1.22) compared with P. lutheri (0.79). While mussels rejected more (P<0.001) S. costatum (ratio of rejection to filtration 0.30) and P. lutheri (0.26) than T. suecica (0.14), oysters rejected more S. costatum (0.25) than both T. suecica (0.10) and P. lutheri (0.08). These differences appear to depend upon differences in algal shape and flexibility. Ratios of rejection to filtration for flagellate species were influenced by the planktonic or benthic origin of the other available algal species. It seems likely that preingestive selection of algae in filter-feeding bivalves may vary according to the relative composition and relative balance of different species within the available seston.