Seventy-nine patients with colorectal adenomata were randomised to receive calcium carbonate (3,000 mg) or placebo in a double-blind randomised trial to assess the short- and long-term effects on rectal mucosal proliferation measured by the in vitro metaphase arrest technique crypt cell production rate (CCPR). There was no significant difference in mean CCPR between the groups before treatment or after 3 or 12 months. In those patients randomised to calcium, CCPR fell at both 3 months [9.0 (2.8) cc c-1 h-1, t = 3.15, d.f = 76, P = 0.002] and 12 months [9.2 (3.3) cc c-1 h-1 t = 2.7, d.f. = 74, P = 0.009] compared with pretreatment CCPR [12.2 (5.5) cc c-1 h-1]. We have demonstrated that calcium had no effect on mucosal proliferation compared with placebo. The results on adenoma formation are awaited.