An array of five surface moorings was set in the subtropical convergence zone southwest of Bermuda with spacings of 16 to 53 km. Meteorological instrumentation on each of the surface buoys recorded wind velocity, barometric pressure, solar radiation, air temperature, sea temperature, and relative humidity. One objective of the deployment was to look for horizontal variability in the meteorological fields on the scale of the array. In support of that objective, both a high data return from the instruments and a quantitative evaluation of the quality of the measurements were sought. To maximize data return rates, two meteorological instruments were placed on each buoy. To determine the accuracy of the measurements, careful predeployment and post-deployment calibrations of all instruments were carried out, and, during the experiment, meteorological data were collected from ships stationed near the buoys. From the two redundant instruments it was possible to construct one complete dataset for each mooring. The results of the calibrations and intercomparisons provided estimates of the errors in the measurements. Significant horizontal variability was occasionally observed in some of the surface meteorological variables and in the wind stress and air-sea heat flux fields. More often, observed spatial gradients in the meteorological fields were not significantly larger than the experimental uncertainty in those gradients. Larger than anticipated errors were encountered in measuring wind speed and barometric pressure, and the preformance of anemometers, barometers, relative humidity sensors, and other sensors for use on buoys could be improved.