The possibility that a drying of the tropical upper troposphere might accompany tropospheric warming, substantially reducing the climate's sensitivity to a change in radiative forcing, has been subject to extensive debate. A simple one-dimensional model of tropical convection and humidity originally proposed by Lindzen (1990a) and Sun (1990) is used to explore mechanisms for ''cumulus-induced drying,'' and a narrowband radiation model to investigate its radiative implications. Low-level moistening, which accompanies warming in all cases, significantly offsets the radiative effects of any upper-level drying, making the net effect on radiative forcing critically sensitive to undetermined parameters in the model. When the model is tuned to reproduce mean relative humidities from Pacific radiosonde data, no substantial drying is observed to accompany a 1 K tropospheric warming. However, the clear-sky moisture feedback is in all cases substantially less than that suggested by a scheme in which relative humidity is held constant, illustrating the importance of upper tropospheric relative humidity for climate sensitivity. Finally, the feasibility of placing constraints on critical parameters through satellite observations of interannual variability of the tropical clear-sky radiation field is investigated.