During the attempt to transfer to tobacco a method commonly used to determine metobromuron in foodstuffs, it became apparent that only very small amounts of metobromuron and less than 0.1 ppm 4-bromoanilin were found. Nor did the modification of this method produce better results, because metobromuron, apparently linked to tobacco components, cannot be extracted with aqueous acetonitrile. The tobacco was then distilled under hydrolysing conditions and the 4-bromoanilin released was bound as hydrochloride. After gas-chromatographic separation 4-bromoanilin is determined specifically with the microcoulometer (bromine) and with the Coulson detector (nitrogen). The amount of metobromuron is calculated from the 4-bromoanilin. With this method no difference can be made between intact metobromuron and 4-bromoanilin already metabolised from metobromuron. Recovery is 90 % for 4-bromoanilin and 92 % for metobromuron, the coefficient of variation is 3.0 %. In cigarettes whose tobacco had grown in soil which had been treated with 4 kg of Patoran per hectare before transplanting, a mean of 5.8 ppm metobromuron was found by this procedure.