The performance of the DNPH-coated C-18 cartridge with respect to sampling of ppb-ppm levels of carbonyls has been tested in dry air (RH = 3-7%) and in humid air (RH = 55 +/- 10%). In humid air, experiments with cartridge (upstream)-impinger (downstream) sampling trains indicated no cartridge breakthrough, and good agreement was found between measured and nominal concentrations. In dry air, there was good agreement between measured and nominal concentrations for ketones but not for aldehydes. For carbonyls produced in-situ as reaction products in ozone-alkene experiments, formation yields in dry air and humid air were the same for ketones but were lower in dry air than in humid air for aldehydes, especially for formaldehyde. For the ca. 20 carbonyls studied, the results indicate good performance of the DNPH-coated C-18 cartridge when sampling carbonyls in humid air and poor performance when sampling aldehydes, especially formaldehyde, in very dry air. Implications of the results are briefly discussed.