DEPARTMENT OF TERRESTRIAL MAGNETISM. 263 



analysis of the new data has confirmed the results announced by the author 

 in 1921^ regarding the predominance, over the oceans, of a 24-hour wave pro- 

 gressing approximately according to universal rather than local time. It is 

 found, however, that, at least in the Pacific, there is also a well-defined second- 

 ary wave during the months of northern summer. 



The present investigation has been extended to include also the diurnal 

 variation of the potential gradient over the several continental areas. From 

 harmonic analyses of the diurnal variation from many widely separated 

 stations, varjdng in latitude from 78° north to 77° south, it is found that, 

 despite the greater complexity of the variation over land, a large percentage 

 of the stations considered show sufficiently good agreement with the ocean 

 results to furnish strong evidence for the assumption of a world-wide effect 

 of 24-hour period progressing approximately according to universal time. The 

 average amplitude of this wave is about 20 per cent of the mean-of-day value 

 and the average time of its greatest phase apparently varies somewhat with 

 time of year, ranging from about IG*" to 19^ G. M. T. 



While there are land stations in the tropical and temperate regions for which 

 the analytical results are not in accord with the foregoing, there is considerable 

 evidence indicating that this may be due to local disturbances of large ampli- 

 tude rather than to an absence of the general phenomenon. The fact that for 

 (P2, the phase angle of the 12-hour wave at local midnight, marked departures 

 from the all-stations mean are largely confined to places where the amplitude 

 of the secondary wave is relatively small lends further support to the view 

 that the limitations of harmonic analyses in the form here used may be 

 partly responsible for the lack of more complete agreement between the 

 land and ocean results. 



Observatory equipment for recording photographically the electric conductivity of the air.'^ 

 S. J. Mauchly. 



An abstract of the matter contained in this paper is given on pp. 303-304 

 of the Director's annual report for 1922. 



The potential gradient at the Apia Observatory, Samoa, for the year ending April 30, 1923 : 

 PreUminary results. S. J. Mauchly. 



The potential-gradient registrations at the Apia Observatory, together 

 with the required control observations, were carried out during the period in 

 question by Mr. Andrew Thomson, of the Department of Terrestrial Mag- 

 netism. On account of the deterioration of the old instrument hut, a new 

 one was constructed under Mr. Thomson's direction in March 1922, due care 

 being taken to introduce no material changes in the location of the hut 

 and the exposure of the collector. The same ionium collector and Benndorf 

 electrometer are used as in recent years at Apia, although the instrument 

 was thoroughly overhauled and put in good condition before the present 

 series of observations was begun. 



The following preliminary results are based upon 112 selected days. No 

 negative hourly values are included, although a few hours containing short 

 periods of negative potential have been used in order to have a reasonable 

 number of days for the formation of diurnal-variation curves for each quarter 

 of the year. In accordance with the usual assumptions these curves have been 

 subjected to Fourier analysis; the results are summarized in table 4, where 

 the phase angles {<p) refer to local midnight and the amplitudes (c) are 

 expressed as percentage of Pm, the mean-of-day value. 



1 Bulletin of the National Research Council, No. 17, pp. 73-77, Washington (1922), and Amer- 

 ican Physical Review, n. s., vol. 18, pp. 161-162 and 477, August and December 1921. 



^ Presented before the annual meeting of the section of Terrestrial Magnetism and Electricity 

 of the American Geophysical Union, Washington, April 18, 1923. 



