Data from several operational black fly abatement programs using Bacillus thuringiensis var. israelensis (B.t.i.) in New York State's Adirondack Mountains were analyzed to determine what the concentration of formulation in the stream might have been if application rates were determined by a stream's width rather than its discharge. There was a high correlation (magnitude of R = 0.87) between discharge and width measured at 315 treatment sites. Had the applications been based upon the stream width, 96% of the actual concentrations in the streams would have been within plus or minus a factor of 5 and none exceeding a factor of 10, a range in which B.t.i. remains environmentally safe and effective.