VOL. LXIX.] PHILOSOPHICAL TRANSACTIONS. 507 



water not only possesses the property of diminishing the noxious part of tainted 

 air, but has also the power, and that in a very high degree, of dephlogisticating 

 common air; which must certainly be one of the methods by which nature 

 keeps the atmosphere in a state constantly fit to support animal life; it being 

 certain, that the water in various circumstances, must lose either a part or the 

 whole of that air which it hath absorbed from the atmosphere. 



It might be suspected, that in these experiments of extracting the air from 

 water by the action of fire, the air might be considerably altered by the vapour of 

 the water itself. As this difficulty was of some force, he endeavoured to remove 

 it in the following manner. He introduced into a tube, through water, a quan- 

 tity of common air of known goodness, and he caused the steam of water boiling 

 in a matrass, from which the air had been previously extracted, to pass through 

 it. The heat of the steam sometimes made the water occupy above 5 times the 

 space it did when cold; yet the air so treated was not at all altered by it, as ap- 

 peared by the test of nitrous air. The event of the experiment, repeated at va- 

 rious times, was constantly the same. 



He observes, lastly, that having once caused the air of boiling water to pass 

 into receivers filled with, and standing in, quicksilver, he found that the air was 

 better than usual. He had observed the same thing when he caused the air to 

 go through distilled water into receivers filled with it: which observation, if the 

 event of the experiment should be constantly the same, would induce him to 

 believe, that the air loses some of its good properties by going through water not 

 very pure; or, which seems to be rather more probable, that a quantity of air 

 less good is, by the action of the vapours and the heat, extricated from that im- 

 pure water, and is mixed with the air that comes out of the matrass, whence this 

 air is debased. 



Mr. F. mentions a new character of equal importance with that which distin- 

 guishes dephlogisticated from common air. This new character has been equally 

 unknown, and deserves the attention of philosophers, because at the same time 

 it discovers a new property of the atmospherical air, which he should never have 

 suspected if experience had not offered it to him. He had found, that common 

 air shaken in water, instead of being diminished, is sensibly increased in its bulk. 

 The increased space is in proportion to the time the air is shaken in water, and 

 it begins to be sensible even from the beginning, that is, after a few seconds. 

 This augmentation he had sometimes brought to be -^ of the bulk of the air, 

 and even more: it must however be confessed, that he met with great variety in 

 the experiments of this kind made at different times. After the bulk of the air 

 shaken in water is increased to a certain degree, it then begins to decrease conti- 

 nually ; and, in proportion to this decrease, the air becomes gradually less good. 

 When the experiment is tried in close vessels, the diminution cannot be observed. 



