The present paper describes a probable budget for acetonitrile and discusses the release of this gas through biomass burning and human activity. The different loss processes in the middle atmosphere are mainly due to the reaction with hydroxyl radicals and atomic oxygen. It is shown that the destruction of CH3CN by scavenging due to precipitation is probably not more efficient than the direct gas phase reactions. Losses due to ion chemistry are very difficult to estimate at present but are probably of secondary importance, except locally, where formation of multi-ion complexes is significant. A one-dimensional calculation shows that the vertical profiles of CH3CN, deduced from ion mass spectrometry data, can be reproduced satisfactorily if an annual global emission ranging from 1.5¿1010 g to 5¿1011 g is adopted, depending on the values of the reaction rate constants and eddy diffusion coefficient. The global atmospheric lifetime of CH3CN is estimated to be about 0.5 to1.4 year. Finally the calculated acetonitrile profiles are introduced in an ion model to calculate the abundances of the major positive stratospheric ions. The results are consistent with present observations.