During the Joint Air-Sea Interaction (JASIN) experiment conducted in the northern Rockall Trough in the summer of 1978, oceanographic moorings with surface buoys carrying wind recorders were deployed in an array designed to investigate the variability of the near-surface wind field at scales of from 2 to 200 km. The wind records together with observations taken on board the research vessels participating in JASIN have provided ground truth measurements for the sea surface wind velocity sensors on the Seasat satellite. During most the experiment the wind field was characterized by spatial sclaes large in comparison with the separations between the buoys. On several occasions, spatial differences associated with cold fronts were identified, and it was possible to track the passage of the front through the array. However, quantitative analysis of the variability of the wind field was complicated both by a lack of data due to mechanical failures of some instruments and by significant differences in the performance of the diverse type of wind recorders. Reevaluation of the instruments used in JASIN and recent comparison of some of these instruments with more conventional sets of wind sensors confirm the possibility that there is significant error in the JASIN wind measurements made from the buoys. In particular, the vector-averaging wind recorder on W2, which was one of the few instruments to recover a full length record and which was chosen during a Seasat-JASIN workshop as the JASIN standard, had performance characteristics that were among the most difficult to explain.