264 Dr. Marshall Hall on the 



The bulb of a retort was filled with small portions of iron, 

 the interstices between the pieces of metal, together with the 

 greater part of the tube of the retort, were then occupied with 

 freshly- boiled distilled water, and a small portion of olive oil 

 was poured into the tube to exclude the atmospheric air ; the 

 retort was then left in an inverted position. It was very long 

 before any chemical change was observed. But, at length, 

 bubbles of gas were seen to ascend to the upper part of the 

 bulb of the inverted retort, and the pieces of iron became 

 tarnished. This effect was observed to take place much more 

 quickly in a similar experiment in which no oil was poured 

 into the tube of the retort, and in which the atmospheric air 

 was consequently not excluded ; the first bubbles of gas were 

 much earlier observed at the top of the bulb of the retort. 



Suspecting, in these experiments, the superadded influence 

 of carbonic acid, of which a minute quantity might be retained 

 by the boiled distilled water in the first of these experiments, 

 and of which a larger quantity might be absorbed from the 

 atmosphere in the second, T wished to adopt some unequi- 

 vocal mode of abstracting every portion of the carbonic 

 acid from the water used in them. In a third experiment^ 

 I took the retort, into the tube of which no oil had been 

 poured, and in the bulb of w'hich bubbles of gas were now 

 rapidly forming, and dropped into it a small lump of lime, 

 which I concluded would seize the whole of the carbonic acid. 

 From this moment the evolution of gas entirely ceased, 

 although it had been previously proceeding for many days 

 uninterruptedly. 



It appeared to me quite plain, from this last experiment, 

 that the appearance of bubbles of gas, observed in the two 

 preceding ones, had arisen from the decomposition of the 

 water, the oxygen being seized by the metal, and the hy- 

 drogen being evolved in the form of gas ; that this decom- 

 position had been effected by the superadded agency of 

 carbonic acid, and that it was arrested when the influence of 

 this acid was abstracted by the lime. To determine these 

 points still more distinctly, I resolved to try the comparative 

 effect of lime-water, and of water mixed with Ume- water, and 



