A new interpretation of the lightning discharge observations made by McEachron and others at the Empire State Building suggests that the polarity of the upwardmoving stepped leaders found by them may have been negative, not positive as has been supposed, the current-recording equipment having failed in most cases to record the initial streamer stages of the discharge.
Important investigations of the current carried by lightning discharges to the Empire State Building, accompanied by Boys camera photographs of the streamers involved, have been made over a ten-year period by the research staff of the General Electric Company (1, 3, 4). 2
About 75 per cent of these discharges were found to begin with an upward-moving stepped leader. This was often followed by a continuous flow of current (~250 amp.) carrying negative charge from cloud to ground and most frequently lasting for 0.27 sec. For half the flashes, however, the continuing discharge ceased earlier and further intermittent negative leader-return strokes of the usual kind discharged the cloud.
The steps of the first leader varied in length from 6 to 23 m., the average pause-times between steps from 19 to 30/z-sec. and the effective velocities of leader advance (pilot-streamer velocities) from 0.6 X 107 to 7.1 X 107 cm/sec. (3). These quantities are in close agreement with those observed in South Africa for downward-moving negativelycharged stepped leaders (6).
From the records of the currents in the discharge and their timevariations, taken at the same time as these photographs, McEachron (3, 4) concluded that the upward step-streamers from the Empire State Building were positively-charged. This would mean that the parameters involved in the propagation of stepped leader streamers are exactly the same whether the streamers are positive or negative, a conclusion so unusual and unexpected as to need careful scrutiny.
It would appear that all the Empire State observations in which photographic and current observations were made on the same flashes date from the years 1935, 1936 and 1937 (3). The current wave-form was recorded by means of a crater-lamp oscillograph which required