The dynamo theory of the daily variation of the earth's magnetic field implies that all large‐scale atmospheric motions in the electrically conducting region should in some measure affect the field and appear in the magnetic records. If this view is correct, many of the conspicuous and even violent changes from day to day in the magnetic records of a single observatory may be a direct reflection of correspondingly marked changes in the atmospheric circulation, as well as the conductivity, in the upper atmosphere over the locality of the observatory. Thus a set of magnetic records might give, in large measure, a direct indication of upper‐atmosphere wind‐patterns. Comparison is here made of certain prominent features of geomagnetic data with the character of the large‐scale atmospheric circulation as shown by observations in the lower atmosphere. There is considerable evidence suggesting that these are interrelated and that a detailed study of this relationship should lead, on the one hand, to an explanation of certain conspicuous features of the geomagnetic variations in terms of the atmospheric circulation, and, on the other hand, to a deduction of certain large‐scale features of the atmospheric circulation from current magnetic obs