Abstract
Theoretical indications of the magnitude of horizontal spread resulting from the interaction between vertical diffusion and the wind profile are briefly reviewed. Assuming a fairly realistic form for the spectrum of turbulence the effect of this interaction is evaluated from a treatment by F. B. Smith, and this is combined with the theoretical growth arising directly from the horizontal component of turbulence to give a total crosswind growth curve. Two series of field experiments on dispersion in stable conditions are examined, one in Sweden using an elevated source of smoke puffs, the other in the U.S.A. using a continuous ground release of a fluorescent tracer. Both indicate that distortion of the plume by the turning of the wind became significant beyond 2-3 km downwind, but that the immediate transference of the effect to enhance the spread at a given level was not important within about 5 km for the elevated source or about 12 km for the ground source.