The relationship between faecal examination for egg output and worm burden of Opisthorchis viverrini in man of 181 autopsy cases from Northeast Thailand is described. Diagnosis of the parasite infection by stool examination for the presence of eggs was less sensitive than the worm recovery technique. Using Stoll's dilution and formalin-ether technique, no eggs were detected in the faeces of 20 cases harbouring low worm burdens (less than 20 worms). The quantitative faecal egg count by Stoll's dilution technique showed a strikingly close positive correlation with the number of worms recovered (r = 0.96, P less than 0.001) indicating a strong linear association between eggs per gram of faeces (epg) and worm burden. The number of epg per worm was inversely correlated to the worm burden (P less than 0.001), suggesting that density-dependent constraints on fecundity could operate to restrict egg output in heavy infections. The accuracy of egg counts for estimating worm burden and its relevance to parasite epidemiological research are discussed.