Mean low-level temperature and wind profiles were constructed for 44 cases of free convection using the data for ONeill, Nebraska, during July and August 1956. Based upon the expression for the normalized logarithmic wind shear first suggested by Ellison and later refined by Panofsky, a theoretical formula for the Deacon profile number as a function of the Richardson number was derived, and values of the Deacon profile number were computed. One of the parameters entering into this theoretical formula is the ratio of the eddy diffusivities for heat and momentum. This parameter was, in turn, computed from Priestleys expression for the dimensionless heat flux for free-convective cases. In using observed wind data from the mean profile in order to verify the theoretical computations of beta Deacon profile number, some marked discrepancies occurred above the 100 cm level. These were due to inconsistent wind speed readings, and it was necessary to employ control data based on neutral profiles to correct the wind speeds. When this was done, the theoretical and observed Deacon profile numbers were in very good agreement.