A set of results about the positive ion composition of the stratosphere, consisting of some recently obtained as well as some previously reported data, is presented and analyzed. It is compared with model calculations described in a companion paper. A review of the abundances of proton hydrates and nonproton hydrates shows that the latter most probably result from the presence of acetonitrile (CH3CN) in the atmosphere. The vertical mixing ratio profile of this gas deduced from positive ion composition measurements is in good agreement with a model based on the assumption that acetonitrile is released at the earth's surface and subsequently destroyed in the atmosphere by chemical reactions (mainly with hydroxyl radicals). The results of an analysis of the possible isotropic mass peaks of mass 96 (H+(CH3)(H2O)3), is not in contradiction with this conclusion.