“…However, amplitudes of temperature oscillation at these gradients apparently are no greater than a few hundreths of a degree C, and in reports in which consideration is given to the convective transfer of heat (e.g., Diment and Robertson, 1963;and Diment et al, 1965), it is generally concluded that, in view of the larger uncertainties that enter into such estimates, convective instability has a negligible effect on the determination of the earth' s thermal gradient or rate of heat flow. On the other hand, my investigations have confirmed the findings of others that under some conditions, temperature gradients become very large, and temperature oscillations of several tenths of a degree may occur in some wells (Diment et al, 1965;Diment, 1967;and Gretner, 1967). Thermal disturbances of this magnitude would seriously impair the accuracy of temperature measurements required to estimate groundwater velocities, for example, in a method suggested by Stallman (1963).…”