Haemolymph samples and haemocytes collected via the adductor muscles of bivalve molluscs are extensively used in ecotoxicological studies. Withdrawal of haemolymph from mussels, Mytilus edulis, via the posterior adductor muscle, may lead to contamination with the intracellular contents of adductor myocytes. Lysopine dehydrogenase (LyDH) activity, an adductor myocyte marker, was used to investigate the impact of this potential contamination on levels of total glutathione, glutathione peroxidase (GPx) and acetylcholinesterase (AChE) measured in cell-free haemolymph. The mean glutathione content of cell-free haemolymph from 28 mussels was 3.2 +/- 1.8 microM (mean +/- SD). There was a linear relationship (slope = 0.28 +/- 0.03 min; mean +/- SE; P < 0.0001, n = 28) with haemolymph LyDH levels suggesting that at least some of the glutathione measured in cell-free haemolymph had arisen from contamination. Haemolymph LyDH activity was significantly higher in samples extracted using larger diameter needles, and also in samples where there had been some difficulty in the extraction. Exposure of mussels to oxidative stress using 40 microg l(-1) Cu for 5 days resulted in a 1.7 fold increase in glutathione (P = 0.033), but no increase (P = 0.810) in LyDH activity in adductor muscle. This was reflected in a similar increase in the slope of a plot of cell-free haemolymph glutathione versus LyDH activity (P = 0.011), consistent with both of these having originated from the adductor muscle. Cell-free haemolymph GPx and AChE activities also correlated with LyDH activity (Spearman rank correlation coefficients of 0.531 (P = 0.0068) and 0.537 (P = 0.0062), respectively, n = 27) suggesting that these also arise from contamination of the haemolymph. For GPx there was a significant linear relationship (P = 0.025) with haemolymph LyDH levels consistent with both enzymes originating from the myocytes. However, there was hyperbolic relationship (P = 0.0004) between haemolymph AChE and LyDH activities. It appears that this is because the AChE originates from a different compartment to the LyDH, i.e. cholinergic neuromuscular junctions in the adductor muscle. We conclude that it would be prudent, when considering the possibility of using a biomarker in cell-free haemolymph from bivalve molluscs, to check whether contamination could be an issue.