Metal (Al, As, Ca, Cd, Cu, Fe, K, Mg, Mn, Na, Pb, Zn) levels in the feeding organ or radula of the common limpet Patella vulgata L. were surveyed in 10 populations over a approximately 150-km stretch of coastline in north-east England. The most northern population was at Beadnell in Northumberland and the most southern was at Port Mulgrave in North Yorkshire; sites included unspoilt bays and areas heavily affected by industrial contamination such as the River Tees estuary. We hypothesized that the radula might be used as an indicator of environmental contamination. There were significant differences between the sites in the ratio of radula length to shell length. Limpets from Whitburn had the smallest radula fraction (mean = 1.665), while those from Port Mulgrave the largest (mean = 1.998). Such variation is common in the literature and we detected no correlate and propose no cause. Iron was clearly the dominant metal in the radulae, with an overall of mean of 1.46% of radular weight, though this is rather low in comparison to values in the literature. Iron is naturally secreted into the developing radula as a putative hardening agent. The next most abundant metals, in descending order, were Na (at approximately 2000-8000 microg g(-1)), K, Mg, Ca (approximately 1000-1500 microg g(-1)), Zn, Cu, Al, Pb (approximately 7-75 microg g(-1)), Mn, As, Cd (approximately 0-1 microg g(-1)). All but Al and Cd showed significant differences between the sites, but not in any consistent or convincing geographic manner. Nevertheless, the variations in metal levels between sites (e.g. Fe > 72%, Cu and Zn > 10-fold) suggest an environmental cause, but we are unable to offer any responsible factor, for example, there appeared little effect of the River Tees estuary. Cadmium is at a relatively low level in the radula in comparison to published data on pedal mucus and the flesh, but Pb is relatively high in pedal mucus and the radula and this might suggest that the radula is a detoxification route for Pb. Although the relationship between radula metal content and environmental metal content is unknown, the radula is constantly replaced and so may yet have the potential to be of use as a bioindicator, integrating metal exposure over much shorter periods than whole body burdens.