Decomposition of Water by the Galvanic Battery. 79 



attacking the positive electrode. When the plates were connected 

 with the opposite ends of a battery, the water was rapidly de- 

 composed, and a considerable quantity of the mixed gases obtained. 

 From the intensity of the lime light produced by the gases, it 

 appeared that the potash effectually prevented the combination 

 of the oxygen with the positive electrode. The iron plates worked 

 well for about an hour ; the decomposition then began to decline 

 very rapidly, though the battery was in good order. I then 

 took up the iron plates, separated them from each other, and 

 removed the linen cover. One of the plates was coated with a 

 non-conducting black deposit. This was the first time I found 

 such a deposit on cither of the plates ; on all former occasions 

 they were separated from each other by slips of wood or gutta 

 percha. I afterwards got a pair of lead plates, which I rolled 

 up in the same way as the iron plates; they, too, ceased to act 

 after some time, because one of them became covered with a 

 black non-conducting substance. I then got four concentric 

 hollow cylinders made of sheet iron ^th of an inch thick, so that 

 the largest of them fitted in the iron vessel. The first or inner- 

 most one was connected with the third, and the second with the 

 fourth ; they were separated from each other by wedges of wood, 

 and also insulated from the bottom of the vessel. The fii'st and 

 third were connected with the top plate, and the second and 

 fourth with the side of the vessel. These acted tolerably well ; 

 but when I used a pair of lead electrodes with a battery of ten 

 6-inch plates, the lime light was considerably larger and more 

 steady than when I employed the iron cylinders. Hence I 

 resolved, if possible, to give xip the use of iron electrodes. I 

 saw at the same time that lead plates would not answer; for 

 when they arc placed near each other, a slight pressure, and 

 sometimes even their own weight, brings them into contact with 

 each other. It then occurred to me to try strong tin plates 

 immersed for a few seconds in melted lead, or in a liquid alloy 

 of lead and tin, in which alloy the quantity of tin would be small 

 compared with that of lead. Not having any new tin plates at 

 hand, I cut up some old tin vessels, and made three plates, each 

 nearly 4 inches wide and G long. I coated one of them, by 

 means of a soldering iron, with an alloy, containing about seven 

 parts by weight of lead and one of tin ; and another with an 

 alloy containing about three parts of lead and one of tin. On 

 the third, after being coated with the alloy, I sprinkled some 

 powdered sulphur, and held the plate over the fire until the sul- 

 I)hur was infiaiiied : the plate was thus covered with a black 

 coating of burnt sulphur. 1 then tried each of the three, and 

 also a plate of lead about the same size, as the negative element 

 of a nitric acid battery ; that is, I put each successively, instead 



