\ 



46 BULLETIN OF THE 



Mr. S. Newcomb thought the difficulties of parallactic determi- 

 nations rendered the estimate of the elevation of auroral streamers 

 quite unreliable, and that the height had been greatly exaggerated, 

 the British observations not indicating much over 100 miles. In 

 regard to the tabulated monthly frequency, as auroras were mainly 

 seen betv^^een 8 and 10 o'clock, the fact that these hours fell in 

 twilight during the months of June and July in the northern 

 countries might of itself account for the small number of auroras 

 seen in those months. 



Mr. J. E. HiLGARD perfectly agreed with Mr. Newcomb. On 

 one or two occasions Prof. Henry, with others, had attempted to 

 obtain the parallax of notable and characteristic auroras, but while 

 a distance could, of course, be assigned from the angles given Jby 

 any two observers, so soon as a third observation M^as combined 

 there was no accord, and no parallax possible. In addition to 

 the uncertainty of the monthly numbers from the changed lengths 

 of the evenings, it appeared that no allowance had been made for 

 the average cloudiness of nights, which would also be found to 

 have an annual law, and the omission of which would introduce 

 another element ot uncertainty and source of error in comparing 

 monthly numbers. 



Mr. J. Henry said that the phenomena of the aurora were evi- 

 dently electrical, and as such thei'e were two facts to consider : 

 first, the electrical discharge; and second, the matter illuminated 

 by the discharge. He had, during a visit to Lake Superior a few 

 years ago, repeated an experiment in which a beam of the auroral 

 light, concentrated by a small concave mirror, fell on a paper on 

 which were letters written with sulphate of quinine, and which 

 became visible as in the experiments with the same invisible 

 writing when illuminated by a discharge of electricity. 



He had also made an observation on the effect produced by the 

 aurora on the needle of a galvanometer, one end of the wire of 

 which was connected with the water pipes, and the other with 

 the gas pipes of the city. In the exhibition of the aurora, on 

 one occasion the needle was deflected 90°, and was only stopped 

 by two pins placed at this degree to prevent farther motion. 



A similar effect was always observed v/heii a flash of lightning 

 took place within the visible horizon of Washington. 



