An assessment of the effects of clouds on photolysis rate coefficients was carried out using three different experimental methods involving two different aircraft platforms. This evaluation was based on data recorded during NASA's Pacific Exploratory Mission (PEM)-Tropics A program in August-October 1996. On the NASA DC-8, upward and downward looking J(NO2) filter radiometers and spectroradiometers were employed. For the NASA P-3B, the instrumentation consisted of Eppley radiometers. Although each aircraft typically sampled the same geographic region, coincident measurements occurred for only one brief period in the marine boundary layer near Christmas Island (2 ¿N, 157 ¿W). All three methods were compared for this flight period; however, only the J(NO2) filter radiometers and spectroradiometers could be compared for the entire campaign. For the Christmas Island sampling period, all three radiometric measurements disagreed in magnitude but exhibited trends consistent with model-calculated photolysis rate coefficients. Overall, the results showed that the J(NO2) filter radiometers and spectroradiometers exhibited a consistent disagreement of 30%, the J(NO2) filter radiometers being higher. Eppley-derived values of J(NO2) fell between those of the J(NO2) filter radiometers and spectroradiometers. An examination of the variation in J(O1D) and J(NO2) based on the output from the spectroradiometers and J(NO2) filter radiometers suggests that differences between each photolysis rate coefficient's response to cloud effects tend to be smaller than model uncertainties. Thus J(O1D) and other photolysis rate coefficients can be corrected for cloud effects based on the response of J(NO2). ¿ 1999 American Geophysical Union