“The building in which Dr Priestley made his chemical and philosophical experiments was detached from his house to avoid the danger of fire. It consisted of several apartments on a ground floor. Upon entering it we were struck with a simple and ingenious apparatus for making experiments on inflammable gas extracted from iron and water reduced to vapour. The tube, which was thick and long, was made of red copper and cast in one piece to avoid joinings. The part exposed to the fire was thicker than the rest. Into this tube he introduced cuttings or filings of iron, and instead of dropping in the water he preferred making it enter in vapour. The furnace destined for this operation was supplied with coke made of coal, which is the best of all combustibles for the intensity and equality of its heat. By these means he obtained a considerable quantity of inflammable gas of great lightness and without any smell. He observed to me, that by increasing the apparatus and using iron or copper tubes of a large calibre, aerostatic balloons might be filled with far less trouble and expense than by vitriolic acid. Dr Priestley allowed me to take a drawing of this new apparatus for the purpose of communicating it to the French chemists who are engaged in the same pursuit.... “Dr Priestley did not regard the experiments made relative to the decomposition of water as satisfactory. He could not admit the fact to be demonstrated so long as the gas was only obtained through the medium of iron, a metal which is itself susceptible of inflammability; but he waited with impatience for the result of the experiments of the French chemists, particularly those of Lavoisier, who had invented and caused to be constructed an extensive apparatus for the same object. “‘The decomposition of water,’ said this indefatigable philosopher, addressing himself to me, ‘is of so much importance in Natural Philosophy, and would occupy so distinguished a place among the phenomena of the universe, that far from admitting the fact upon slight evidence, and as it were from enthusiasm, it were rather to be wished that all objections that may be made, and which will still long continue to be made against this theory were completely refuted; in the conflict of opinions, truth may at last be obtained. But I have still so many doubts upon this subject, and I have so many experiments to make, both _pro_ and _con_, that I can as yet regard the greatest as only started.’