2 SO Messrs. J. Trowbridge, T. C. and I. C. Howe on tJie 



and one yielding sparks 180 to 200 centimetres in length. 

 These transformers are connected to a battery of ten thousand 

 cells ; and can afford voltages from twenty thousand to three 

 million. Various methods of measurement were tried. On 

 account of the great tension all galvanometric and electro- 

 metric means had to be abandoned; and what is known as the 

 electric thermometer was finally adopted. This consists, as is 

 well known, of a hermetically closed glass tube provided with 

 a manometer-gauge. The electrical discharge passes through 

 the tube by means of a fine wire, or by a spark-gap. The 

 term electrical thermometer is a misleading one : for, as we 

 shall show, the quick movement of the manometer-gauge is 

 not due to heat. 



At first we passed the discharge through very fine wire. 

 The gauge, which consisted of a bent glass tube containing an 

 index of bichromate of potash, rose suddenly at each discharge 

 and returned approximately to the zero-point. When, how- 

 ever, the wire had been heated by repeated discharges, the 



Fiar. 1. 



index showed a slow rise in temperature. Even when the 

 terminals in the glass tube were connected by a fine wire a 

 very strong electrostatic field was created in the tube. This 

 is shown in an interesting manner by fig. 1. This photograph 



