Zinc and sulphate removal from synthetic wastewater was investigated by using four laboratory parallel upflow-mode reactors (referred as R1 to R4; RI contained carriers to retain biomass, whereas R2-R4 were operated as suspended reactors). All reactors were inoculated with anacrobically digested cow manure. R1 and R2 were first fed with glucose- and sulphate-containing feed for 48 days after which all four reactors were fed with wastewater containing 50 mgl(-1) of zinc in R1-R3 and 200 mgl(-1) in R4 and operated for 96 days. In all reactors, hydraulic retention time, organic loading rate, and sulphate load were 5-6 d, 0.2-0.4 kg COD m(-3) d(- 1) and 3.3-3.8 gSO(4) l(-1) d(-1), respectively, whereas the zinc load in R1-R 3 was 0. 074-0.077 and in R4 0.282 g Znl(-1) d(-1). During the runs, 30-40% of sulphate and over 98% of zinc was removed, and up to 150-200 mg H2S was produced in all reactors. Effluent pH dropped in all reactors (feed pH 6.5) to 3-5 by the end of the experiment. No significant effects on zinc removal were observed, despite differences in operating conditions and feed. It was only in the latter part of the runs (i.e. between experiment days 120-142) that zinc removal began to fluctuate, showing a negligible decrease in R3 and R4, whereas in R1 and R2 zinc was removed below the limit of detection (<0.01 mg Zn l(-1)). Qualitative X-ray diffraction analysis of the reactor sludge at the end of the runs indicated that the compounds precipitated were most probably ZnS (Code 05-0566 Sphalerite), suggesting metal removal through sulphide precipitation; this was supported by the fact that sulphate was reduced and zinc removed simultaneously. (C) 2002 Elsevier Science Ltd. All rights reserved.