This report summarizes the analysis of prevalent winds in 16 pairs of neighboring PRRS virus-infected and PRRS virus-negative pig sites. Eight pairs were selected because airborne transmission was suspected (cases) and 8 pairs were selected because airborne transmission apparently did not occur (controls). To determine the relationship between prevalent wind direction and the potential spread of PRRS virus from infected herds to known-negative herds, information about virus similarity, meteorological data and geographic location of herds was analyzed. In all 8 cases, there was at least one wind event that was of sufficient speed and duration and in the direction of interest during the 2 week period when infection was thought to occur. The distance between sites varied from 1.10 to 12.02 km. However, there was also at least one wind event of sufficient speed and duration and in the direction of interest for all control pairs of herds. The distance varied from 1.48 to 7.01 km. The distances between sites or the mean number of wind events did not differ significantly between case and control pairs of herds. The results of this study indicate that PRRS virus spread between herds will not necessarily occur just because wind is blowing from an infected herd towards a negative site. More information is necessary to understand why some herds apparently become infected by airborne transmission while others remain un-infected under similar conditions of season and distance between sites.