Decomposition of Water by the Galvanic Battery. 75 



through water contained -in an iron vessel, to the top of which 

 was screwed the jet belonging to our large gas microscope and 

 polariscope. The first iron vessel w^iich I used was about o|^ 

 inches high and 2 inches in diameter; its sides were an inch thick. 

 This vessel was found to be too small. On two occasions, all, or 

 nearly all the water Mas thrown out of the vessel through the 

 jet ; the flame went back and exploded the gases in the bags. 

 On these two occasions the pressure was very irregular. The 

 quantity of the gases contained in the bags was too small to be 

 forced out by weights on the pressure-boards. They were driven 

 through the jet by a person who pressed frequently and violently 

 on the bags with his arm. This violent and irregular pressure 

 forced the water through the jet. Had the pressure been uni- 

 form, it is probable the explosions would not have occurred. 

 Dangerous explosions can happen only because the water may 

 be gradually ejected from the iron vessel through the jet or into 

 the gas bag ; or because, when the gases are made to pass rapidly 

 through water in large quantities, there is a continuous series of 

 large bubbles rushing through it ; and should the uppermost 

 bubble be ignited, the flame might possibly descend through the 

 series of bubbles without being extinguished by the surrounding 

 water. In the apparatus which I have made, a dangerous ex- 

 plosion from any of these causes is impossible. The apparatus 

 consists of two wrought-iron vessels of unequal size. The smaller 

 is the one already described. The -large one is about 7^ inches 

 high and 4 inches in diameter ; its sides are about |^ths of an 

 inch thick. On the top of the vessel is laid a collar of thick vul- 

 canized India-rubber. An iron plate about f ths of an inch thick is 

 then screwed down to it by five iron bolts. The vessel is thus 

 made air-tight. The top of this vessel is connected by an India- 

 rubber tube with the bottom of the small one ; the bottom of it is 

 connected by a similar tube with the gas bag, gasometer, or volta- 

 meter. The two vessels are nearlyfiUed with water. The gas is sent 

 into the bottom of the large one, ascends through the water, passes 

 through the tube to the bottom of the small one, then through 

 the water, and issues from the jet screwed to the top of the 

 small vessel. Since the two vessels are of very unequal size, it 

 is impossible that all the water should be carried out of both at 

 the same time by the stream of the gases ; and should an ex- 

 plo.sion occur after the small vessel became empty, the flame 

 would be stopped by the water in the large vessel. In each 

 vessel the gases are made to pass through wire-gauze or perforated 

 zinc, or through small pieces of porous earthenware, in order to 

 break the bubbles, and thus prevent the gases from ascending 

 in a continued scries of large bubbles. To pre\ent the water 

 from being driven into the gas bag or voltameter which may be 

 G3 



