Central axis depth dose and isodose curves around a 90Sr applicator used for neurosurgery were measured in a polystyrene phantom, using LiF teflon dosimeters having the shape of discs, 1 mm in diameter and 0.1 mm thick. The minimum distance between the applicator and the dosimeters 0.1 mm. The results were checked at a number of points using a simplified method of calculation. Agreement within experimental error was obtained between theoretical and experimental results at source-detector distances of 2 to 5 mm. At smaller distances, the approximations involved resulted in too high theoretical values. At larger distances the consistency was again poorer. The measured values of absorbed dose have experimental errors of a few per cent.