Nfzu Eleclnc Experiments. 197 



the tubes, that the fluid might afcend to the upper part of 

 them. The exhaufted fpace left by the defcent of the mer- 

 cury was fome inches in length, which appeared to him to 

 be the moil advantageous. He then held the tubes in a ver- 

 tical pofUion in a veflel tilled with mercury, and, placing a 

 ball three inches in diameter on the wires at the top of the 

 tubes, caufed the fparks from the conductor to fall upon it. 

 The quiekfilver in the tube was at the fame time connected 

 with an infulated ball by means of a wire. 



The firft experiments were made with water, carefullv pu- 

 rified by boiling and by the air-pump. When the fparks 

 (truck the water through the vacuum, a considerable quantity 

 of air feerhed immediately to be produced, fo that in three 

 minutes the mercury fell j\ inch. During the next live 

 minutes the mercury fell only £ of an inch, and the produc- 

 tion of air then totally ceafed. After three days, the air 

 which had been produced did not appear to be in the leaft 

 lefi'ened. By another experiment of the like kind fueh a 

 quantity of air was produced that in four minutes the mer- 

 cury fell 3 inches 4 lines ; but next day the air was leflened 

 1 inch 8 lines : the remainder retained its elasticity. 



In an experiment with alcohol, fuch an abundant quantity 

 of air was produced that at firft the mercury fell ; inch at 

 each fpark. The production of air, however, decreafed in 

 proportion to the falling of the mercury. Two experiments 

 of the fame kind gave more linking refalts, which in other 

 refpccls were fimilr.r to the former. 



In an experiment with cauftic ammonia, a column of air 

 of 21 inches was produced in live minutes. Carbonat of 

 ammonia gave a column of 18 inches, and camphor one of 

 6! inches in the fame time. The air produceu by alcohol, 

 when tried by a teft, was found to be pure inflammable air: 

 that from camphor was found to be nearly as he: from mix- 

 ture : that, however, from both kinds of ammonia crnfilied 

 <jf inflammable air mixed with azot. From this it appears 

 that the electricity had fcpa rated from each other the two 

 component parts of the ammonia, hvdroge:i and azot. Dr. 

 Van Martini imagined that the air produced from the water 

 would cqnfift alio of oxygen and hydrogen, and on that ac- 

 count 



