On Some Observations of Atmosi)heTic Electricity. 89 



Art. X. — Notes on Some Observations of Atmospheric 



Electricity. 



By R. L. J. Ellery, Esq. 



[Read before the. Royal Society of Victoria, 16th November, 1876.] 



Some years ago I described to you an apparatus which I had 

 arranged for obtaining a continuous record of the electrical 

 condition of the atmosphere at the Melbourne Observatory, 

 which was a modification of the exquisite electrometers 

 devised by Sir William Thompson. This apparatus was in 

 operation for several years with most satisfactory results, 

 and a valuable series of records were obtained. It was 

 found, however, almost impossible to maintain the instru- 

 ment in perfect working condition in some states of the 

 atmosphere, through .the subtle nature of the force dealt 

 with and the difficulty of maintaining the requisite insula- 

 tion of all parts of the apparatus. In consequence of this, 

 the working of the instrument had to be frequently inter- 

 rupted for improvements in the methods of insulation and 

 of collecting the electricity from the air; and, I regret to say, 

 eventually stopped altogether until a more efficient plan 

 for insulation could be obtained. 



It is, however, with respect to the results of some obser- 

 vations with this instrument that I now wish to say a few 

 words; but I will at first briefly refer to the generally 

 accepted theory of the distribution of electricity over the 

 earth's surface. 



As a rule, the potential of the earth's surface is negative 

 relative to that of the air above it. Exceptions to this, 

 however, sometimes occur. Generally speaking, I" have found 

 in quiet and fine weather that if the air has a certain 

 electric potential, say six feet from the ground, a contour of 

 an equi-potential line traced over the gTound, buildings, 

 trees, &c., will be approximately six feet from the surface of 

 such portions of the earth's surface ; the line will, however, 

 usually approach the summit of a building, hill, or tree, to 

 something less than six feet ; and as the potentials of higher 

 strata are contoured this difference decreases, so that at a 

 few hundred feet the equi-potential Hnes will probably be 

 found to be parallel to the earth's surface. This is only the 

 case in very serene weather, for in wind, rain, fog, or dust, 

 the case is very different, and nothing more variable than 

 the electric condition of the air can well be conceived, and 



