An observationally constrained box model has been constructed in order to investigate the chemistry of the marine boundary layer at Mace Head, a remote location on the west coast of Ireland. The primary aim of the model is to reproduce concentrations of the hydroxyl (OH) and hydroperoxy (HO2) radicals measured by an in situ fluorescence assay by gas expansion (FAGE) instrument, and the sum of peroxy radicals &Sgr;(2>+2>) determined by a peroxy radical chemical amplification (PERCA) instrument. The model has been constructed based on observed concentrations of a suite of non-methane hydrocarbons, measured in situ by gas chromatography. The chemical mechanism for the model is a subset of a comprehensive master chemical mechanism (MCM). This paper describes in detail the construction of the model, as well as the underlying approach. Comparisons of modeled and measured concentrations of radical species, from a recent field campaign held at the Mace Head Atmospheric Observatory during July and August 1996 (EASE 96), are also presented. For the limited OH data available from this campaign, the model tends to overestimate the observations by about 40%, although this discrepancy is within the uncertainties of the model (¿31%, 2&sgr;) and the FAGE measurements (¿75% on average, 2&sgr;). For HO2 the model reproduces the concentrations well on one day but less well on another. Low HOx concentrations compared to model results have been observed previously, with greater than expected heterogeneous losses invoked to explain the differences. Comparisons between measurements of peroxy radicals made by chemical amplification and model predictions show good agreement over a wide range of conditions. ¿ 1999 American Geophysical Union