164 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 11, NO. 7 



Owing to considerable cloudiness during the first hour of the eclipse, the 

 conditions at Sobral were not so favorable for the detection of an eclipse- 

 effect as those which favored the observations made for the Department of 

 Terrestrial Magnetism by the author and others, in connection with the eclipse 

 of June 8, 1918. However, in spite of the fact that clouds still remained, 

 they did not obscure the Sun after a time corresponding to about 15 minutes 

 before totality, and it was possible to note, in the results for the remainder 

 of the eclipse, variations of the several elements similar to those observed at 

 Lakin in June, 1918. 



Briefly enumerated, the Sobral results indicate: 



(i) A general potential-gradient minimum beginning just before totality 

 and continuing until about 20 minutes after totality. The values during this 

 period were abnormally uniform and about 20 per cent lowei than those which 

 obtained during the equal periods just preceding and following the one under 

 consideration. (2) For both the positive and negative conductivity as de- 

 termined by the ordinary Gerdien apparatus, a maximum of the order of 

 20 per cent, which set in just after totahty and continued throughout most 

 of the period of potential-gradient minimum. (3) That the air-to-earth 

 current-density, as computed from potential-gradient and total conductivity, 

 remained near the normal station value throughout the period in which 

 obscuration exceeded 50 per cent, and was more constant than during arny 

 equal period of the forenoon. 



In view of the general agreement between the results obtained at Lakin, 

 in 1918, and at Sobral, in 1919, a detailed study was made of all available 

 data concerning atmospheric-electric observations made during solar ecHpses. 

 It was found that the evidence as a whole is of a very conflicting nature un- 

 less special care is taken to note the apparatus and methods employed by the 

 various observers, and the meteorological (especially cloud) conditions which 

 prevailed during the observations. 



Following these lines it was found that almost without exception the con- 

 ditions attending previous observations were less favorable for the detection 

 of an eclipse-effect than at Sobral and Lakin, and in nearly all cases where 

 conditions were reasonably favorable the results for potential-gradient and 

 conductivity were in substantial agreement with the foregoing. 



The paper was illustrated by lantern slides. 



L. A. Bauer: Results of magnetic and meteorological observations during 

 solar eclipse of May 2Q, igig (illustrated). 



With the aid of magnetic data from nine stations within the region of visi- 

 bility of the solar eclipse of May 29, 1919, five of these stations being those 

 at which observations were made by the expeditions of the Department of 

 Terrestrial Magnetism of the Carnegie Institution of Washington, and from 

 about eighteen cooperative stations distributed outside the region, the fol- 

 lowing main conclusions weie drawn: 



(j) Magnetic effects of appreciable and determinable magnitude were ob- 

 served during the solar eclipse of May 29, 1919, at stations inside the region 

 of totality as well as at certain stations in the sunlit region, the magnitude 

 and character of the effects being similar to those observed during previous 

 solar eclipses and showing a distinct connection with the eclipse circumstances. 

 The magnetic data for stations in the night region of the globe did not exhibit 

 similar effects. 



