Hydroxyl plays a central role in the chemistry of the troposphere. Since the importance of OH was first recognized in the early 1970s, many experiments have attempted to measure OH, but very few have been successful, and much controversy has surrounded the measurements made <Crosley, this issue>. Thus, intercomparison of OH measurements is a requirement for community acceptance of derived OH concentrations, and accurate measurements of OH are essential to provide robust modeling of tropospheric chemistry since models typically allow OH concentrations to be determined within the model for lack of measurements. The Tropospheric OH Photochemistry Experiment held in the Colorado Rocky Mountains in fall 1993 brought together four instruments that to measure OH. In order to make a proper comparison, a number of other trace species and relevant j values were measured simultaneously with hydroxyl to provide a photochemical understanding of the measurements and an understanding of the airflow in the mountainous region in which this work was conducted. This region provided opportunities to measure both clean and polluted air over a wide dynamic range of species that affect the OH concentration. The measurements extended over a long enough period of time that a large quantity of data were collected under a variety of atmospheric conditions, thus allowing for a meaningful comparison of techniques. The long-path spectroscopic laser system and the in situ selected ion chemical ionization mass spectrometry system determined OH concentrations for 4 weeks simultaneously. Generally, the derived concentrations from the two instruments agreed within the 2&sgr; error bars over half the observation time. One quarter of the time, OH differences could be explained by different air masses between the long path and in situ site which contained different concentrations of trace species that affect OH abundance. Approximately one quarter of the data disagreed outside the observational errors with no obvious explanation. The long-path experiment typically gave concentrations about 20% higher than the ion-assisted experiment. This could be caused by inaccuracies in the absolute calibrations of the two instruments, or by trace species variations along the long path that differed from those at the in situ site. Generally, agreement between ion-assisted and long-path OH measurements was good, showing that measurements of hydroxyl in the troposphere can be performed with good accuracy on a routine basis.¿ 1997 American Geophysical Union