Fig. 6. Copper accumulation by intestinal cells exposed to external [Cu] (Cu_o) ranging from 0 (no added Cu) to 800 µmol l^–1 for 15 min in normal saline (NaCl=140 mmol l^–1 at pH 7.4, filled triangles), compared with the effects of either low external chloride (Cl^–_o=6.6 mmol l^–1, Cl^–_o replaced by sodium gluconate, filled circles), 0.1 mmol l^–1 DIDS in normal saline (open squares), or with low external pH (pH 5.5, open circle). Values are means ± S.E.M. of N=5–6 separate experiments using fresh cells from different individual fish. Rapid Cu accumulation in/on cells at time zero was not deducted from the data. Different letters (a, b, c or d) indicate a statistically significant difference between adjacent treatments within Cu_o (looking up at adjacent data points between plots at each Cu_o, Kruskal–Wallis test, P<0.05). For clarity, statistical differences between treatment effects at 10 µmol l^–1 Cu_o are not shown, but are identical to those at 50 µmol l^–1 Cu_o. Cu accumulation within treatment (Cu-dose effects within each experiment) were significantly different from the control with no added Cu at Cu_o=50 µmol l^–1 or greater for the low Cl^–_o and low pH experiments; and at 10 µmol l^–1 Cu_o or greater for the DIDS experiment (labels not added for clarity, Kruskal–Wallis test, P<0.05). The effect of DIDS reached a plateau at 200 µmol l^–1 Cu_o because Cu accumulation rates between 200–800 µmol l^–1 Cu_o were not statistically different from each other within DIDS treatment (from post-hoc box whisker plots following Kruskal–Wallis test). Similarly, the low pH effect reached a plateau at 400 µmol l^–1 Cu_o because Cu accumulation rates between 400–800 µmol l^–1 Cu_o were not statistically different from each other within the low pH treatment (from post-hoc box whisker plots following Kruskal–Wallis test). The above statistics for Cu-dose effect within treatment are not shown for clarity.