2.2. Seston matrices
Based on our previous findings ŽArifin and Bendell-Young, 1997., four treatments
representative of non-selective and selective feeding strategies, and one control experiment
representing a Astandard basal metabolic rateB were used ŽTable 1.. Of the
five-seston matrices, two had been previously shown to evoked a non-selective feeding
strategy ŽArifin and Bendell-Young 1997.—Ž1. no algae and 50 mg ly1 silt and Ž2.
150=106 cells ly1 of algae and no silt, and two evoked a selective feeding strategy—Ž3.
150=106 cells ly1 of algae and 50 mg ly1 silt and Ž4. 20=106 cells ly1 of algae and
20 mg silt ly1. Seston quality of the first two matrices Žnon-selective. is equal to diet quality, 2% and 60% organic content, respectively, where diet quality is that portion of
seston that is actually ingested by the mussel. However, as a consequence of selective
feeding, for the two matrices, which evoke a selective feeding behaviour, seston quality
does not equal diet quality. In this case, mussels are capable of converting a seston
organic matter content of ca. 20% to 51% and 33% for the 150r50 and 20r20 algae to
silt ratios, respectively ŽArifin and Bendell-Young, 1997, Table 1.. The fifth, a control
treatment in which mussels were exposed to only 0.45-mm filtered seawater, i.e., no
algae and no silt, was included to represent metabolic maintenance