The Nonmethane Hydrocarbon Intercomparison Experiment (NOMHICE) was designed to evaluate current analytical methods used to determine mixing ratios of atmospheric nonmethane hydrocarbons (NMHCs). A series of planned experiments, or tasks, were implemented to test the analytical methods in a graduated fashion. Tasks 1 and 2 involved relatively simple standard gas mixtures prepared by the National Institute for Standards and Technology (NIST). Results are presented here for task 3 in which a complex mixture containing 60 commonly observed NMHCs at concentrations of 1-30 parts per billion by volume (ppbv) in nitrogen diluent gas was prepared and distributed for analysis to 29 participating laboratories throughout the world. Reference mixing ratios were determined jointly by scientists from the National Center for Atmospheric Research (NCAR) and from the U. S. Environmental Protection Agency (EPA). Participants were asked to identify and quantify the hydrocarbons present in the mixture and submit their results to NCAR-NOMHICE scientists. The results were encouraging overall. Some laboratories performed extremely well during this exercise whereas other laboratories experienced problems in either identification or quantification or both. It is evident from the comparison of the NCAR-NOMHICE results with both the EPA analysis and the top 11 analyses in the study that very good agreement is achievable between laboratories for mixtures in this concentration range. Some of the largest analytical discrepancies were from laboratories that used in-house standards for their calibration and/or syringe sample injection techniques. A major conclusion from this study is that the use of high-quality gas phase standards, introduced into the measurement instrument in a similar manner to air samples, is an important prerequisite for an accurate analysis. ¿ 1999 American Geophysical Union