METAL LEVELS IN RODENTS FROM SKOMER ISLAND J. A. CAMERON INTRODUCTION When this work was being carried out, investigations into the effect upon small rodents of vegetation growing on derelict smelter spoil heaps were being conducted in the Lower Swansea Valley. Information about the levels or distribution of metals in the bodies of native small rodents was not available. Allcroft & Blaxter (1950), Egan & O'Cuill (1970) and others had reported the levels of lead in various domestic animals and had suggested the upper levels of lead which these animals could tolerate. It was decided to obtain specimens for comparison from an area which was remote from industrial activity, past or present. Skomer Island was selected as it lies to the west of any industrial area and the prevailing wind is from the south west. The island does not therefore receive regular industrial airborne contamination although, when the wind comes from the south east, some may occur from the Milford Haven oil refineries. Another advantage of Skomer Island is that there are no motor cars on it, the only mechanical transport being a tractor. METHODS The areas in which it was decided to trap were based on the maps and results of Fullagar et al. (1963). Samples of the foliage of the main plants growing in those areas in which animal specimens were trapped were also collected. A line of 12 Longworth traps was laid on the area of Bluebells (Endymion non-scriptus) starting 100m. north east of the Farm House. A further line of 12 Longworth traps was laid in an east/west direction on the north side of the end of the Lantern. Both lines of traps were prebaited for a period of 18 hours. Fourteen specimens of Woodmouse (Apodemus sylvaticus) were trapped on the Lantern and 4 males were collected. The traps on the main part of the island caught only 3 specimens, all were male Skomer Voles (Clethrionomys glareolus skomerensis), during 24 hours. The animal specimens were killed and various tissues were removed and wet ashed in aristar grade nitric acid. The samples were then analysed by atomic absorption spectro-photometry for zinc, copper, cadmium and lead. The plant material was oven dried to constant weight and then wet ashed in a mixture of aristar grade perchloric and nitric acids. These samples were analysed by the same method for zinc, copper, cadmium, lead, magnesium and calcium.