208 Mr. Prideaux's Experimental Contributions 



II. Is Thermo-electricity produced at the expense of Caloric? 



3. Into the smaller neck of a tubulated receiver was fixed 

 an open glass tube, reaching obliquely to the bottom, and 

 bent, siphonwise, on the outside, so as to descend a little lower 

 than the inner end : it was then again bent upward, nearly at 

 a right angle, and to this rising end a scale was attached. 



Into the receiver was poured water tinged with cochineal, 

 and covered with a stratum of oil, to prevent evaporation by 

 the heat. Suction being now applied to the end, the liquid 

 remained, on the principle of the siphon, a few degrees up 

 on the scale. A cork thoroughly soaked in grease being fitted 

 to the open neck, the whole constituted a very sensitive air 

 thermometer. 



The cork was now perforated by two wires ; one of copper, 

 the other of tinned iron ; the latter bent off at each end, so that 

 it could be brought into contact with, or separated from the 

 copper by a slight turn with the fingers. Both ends of each 

 wire were doubled, flattened and polished to ensure good 

 contact ; they projected about three inches into the receiver, 

 and nine inches outside. 



Being now set in contact, the cork was drawn out, and the 

 ends of the wires which entered the bulb were placed between 

 a fold of sheet copper, and heated by holding them for a given 

 time (25 s ) over an Argand gas light, being turned over at 15 s , 

 to equalize the heat in both wires. The sheet copper being 

 now drawn off, the hot ends were thrust into the bulb, and 

 the cork pressed in air-tight. The fluid rose quickly to about 

 90° (more or less, depending on slight differences in the heat, 

 and in the quickness and force with which the cork was pressed 

 in). After a few seconds, the liquid began again to descend, by 

 the cooling of the wires. They were then separated, when the 

 descent of the fluid was suddenly checked : on renewing the 

 contact, it fell with renewed impetus ; and so stopped and re- 

 sumed its fall at every separation and renewal of the contact. 



Other wires of the same and different metals were tried, and 

 with many different corks; but the effect was always con- 



