An attempt has been made to explain the departure of the observed hourly variation of phase height of E region reflections at 2.2 MHz from that expected of a simple undisturbed Chapman layer as the effect of vertical electromagnetic drift. Calculations show that a peak drift velocity ranging between the extreme limits 50.0 and 2.5 m/s located about 2 km below the electron density peak may account for the observed departure. Considering some related observational data it is suggested that the vertical drift velocity would be around 10--20 m/s, which is consistent with a value of &agr; in the vicinity of 2--8¿10-8 cm3/s.