The feedlng response of the blue mussel Mytllus trossulus exposed to environmentally relevant seston matnces were studied in the laboratory under flow-through conditions Twelve seston matnces varying In quality and quantity were prepared by mixlng 3 microalga Thalassloslra pseudonand (5, 20 and 150 X 106 cells 1 ') and 4 silt concentrations (0 5 20 and 50 m g 1 ') to represent an increasing seston organic quallty of 10 to 71 and dn increaslilg seston quantity of 1 4 to 56 6 mg I ' Clearance rates (CRs) pseudofaeces (PF) production, sorting efficiency (SEF) ingestion rates (IRs) of particulate organic matter (POhl) and particulate inorganic matter (PIM) and apparent and true carbon ass~milation efflciencies (C-AE0 ,) were determined for the varlous exposure regimes Under conditions of 0 dnd 5 mg 1 ' silt CRs decreased by 3-fold 114 4 to 4 9 1 h ' g ' dry wt (gdw ')I and 6-fold (18 0 to 3 3 1 h ' cldw ') respectively with Increasing seston quality Under condit~ons of hlgh silt loads (20 and 50 m g I-') CRs were independent of increasing seston quallty with maximum CRs (21 8 & 2 2 1 h ' gdw l ) observed at the 20 x lob cells 1 ' and 20 mg l I silt exposures PF production was dependent on seston quantity (r2 -0 63 p < 0 05) with mussels preferentially rejecting the inorganic versus the organic component of the seston This SEF was optlrnized at a seston quality of 40% organic matter As a consequence of this selective feedlng strategy, determined POM IRs under high a l g a e / h~g h s~l t (high quantity/quality) exposure reglmes were comparable to those of the high quality (lust algae) exposures (POM IRs of 48 0 as compaied to 38 1 and 91 3 mg h ' gdbv ' for high quality and high quantity/qual~ty respectively) In contrast mussels exposed to low quahty/quantity seston Ingested both seston components (SEF < 2 4 % ) I e the mussel was non-select~ve possibly ingesting both seston components to meet nutrient requirements Apparent C AE",, expressed as C ,,,,,,, -C , ,,,,., /C ,,,,,,, correlated with seston quality ( '0 POM) (r2 = 0 64, p < 0 05) However, carbon assin~ilation expressed as a true C-AE"o, C,,,,, -C,,,, ,,, iC,,,,,, \vas Independent of seston quality as was carbon assimilation rate (the product of the amount of Ingested organic matter and carbon assim~lation efficiency) Hence through a dynamic interplay between CRs and cai bon a s s~m~l a t i o n effic~ency the blue mussel was able to malntain a constant late of carbon assim~lation regardless of the quality and quantity of sebton to which it was exposed This compensatory feeding strategy displayed by the blue mussel has Impoi tant implications for predictive models w h~c h relate contaminant uptake to a diet/energy pathway Contaminant uptake may be either unde~estimated or overestimated if the abil~ty of the blue mussel to optlmize nutrlent gain under d vanety of seston compositions has not been taken into account KEY LVORDS Myhlus trossulus Feeding behaviour Sorting process Pseudofaeces Ingestion rate Ass~milatlon of carbon