TROWBRIDGE. — ELECTRICAL DISCHARGES, 439 



connector running along b and b' are brought in contact with the coatings 

 ^ and E' in such a manner as to connect them in series. The conduc- 

 tors connected with the lever arm L extend beyond the connecting arm 

 D, in order to make connection with the terminals of the storage battery 

 B. This extensi9n is necessary in order to prevent any possibility of a 

 return discliarge through the battery. T represents one of the terminals 

 of the machine, — which is a special arrangement of the apparatus con- 

 sisting of spheres one foot in diameter, and thus give a scale of estimation 

 of the size of the apparatus. There are one hundred and fifty plate 

 cmidensers forty by fifty centimeters. 



Lord Kelvin, in a paper dated April 12, 18G0,* entitled "Measure- 

 ment of Electromotive Force required to produce a Spark," states that 

 '• there is a much less rapid variation with distance of the electrostatic 

 force preceding a spark at the greater than at the smaller distance. It 

 seems most probable that at still greater distances the electrostatic force 

 will be found to be sensibly constant, as it was certainly expected to be 

 at all distances."' 



Professor Elihu Thomson, by means of transformers, has obtained 

 sparks of fifty to sixty inches in length, and has estimated the necessary 

 voltage to produce a spark of 80 cm. at 500,000. 



This estimate A. Heydweiler f thinks is very much too great, and he 

 believes that 100,000 would be nearer the truth. My investigations 

 show conclusively that the estimate of Professor Thomson is far nearer 

 the truth than that of Heydweiler, and instead of being lessened it should 

 be increased. 



In a late paper on the tension at the poles of induction apparatus, 

 A. Oberbeckt states that a potential difference of 60,000 volts under given 

 conditions can produce a spark of more than 10 cm. in length. It is 

 difficult to obtain consistent results by the use of induction coils and 

 transformers. 



My results show that Lord Kelvin's conjecture, that the electrostatic 

 force necessary to produce a spark in air remains sensibly constant for 

 all distances beyond the limit he describes, is correct up to the voltage 

 of one million ; for when the length of spark is plotted as abscissas, 

 and the corresponding electro-motive force as ordinates, a straight line 

 is obtained. Plante calls attention to the fact that the loss of energy 

 resulting from the transformation of dynamic into static electricity is in 



* Papers on Electrostatics and Magnetism. Macmillan, London, 1872, p. 258. 

 t Wied. Ann.. No. 48, p. 231, 1893. \ Wied. Ann., No. 9, p. 109, 1897. 



