VOL. LXIX.] PHILOSOPHICAL TRANSACTIONS. bQg 



trous air ; yet the air of Islington is esteemed to be much better. He examined 

 the air of London, taken at different heights; for instance, in the street, at the 

 2d floor, and at the top of the adjoining houses, and found it to be of the same 

 quality. Having taken the air at the iron gallery of St. Paul's cupola, at the 

 height of 313 feet above the ground, and also the air of the stone gallery which 

 is 111 feet below the other; and having compared these '2 quantities of air with 

 that of the street adjoining, he found that there was scarcely any sensible difference 

 between them, though taken at such different heights. 



In this experiment a circumstance is to be considered, which must have contri- 

 buted to render the above-mentioned differences more sensible; this is, the agi- 

 tation of the air of the cupola, for there was felt a pretty brisk wind on it, which 

 he observed to be stronger and stronger the higher he ascended; whereas in the 

 street, and indeed in all the streets he passed through, there was no sensible wind 

 to be felt. This experiment was made at 4 in the afternoon, the weather being 

 clear. The quicksilver in the barometer at that time was 28.6 inches high, and 

 Fahrenheit's thermometer stood at 54°. After having related all these circum- 

 stances, it will be necessary to give the mean result of all the various experiments 

 made on each of those quantities of air treated after his method with the nitrous 

 air. The air of the street gave ii — 13, ii -f- 6; the air of the stone gallery, 

 which was 202 feet high, gave ii — 14, ii -|- 5; and the air of the iron gallery, 

 which was 313 feet high, gave ii — 14, ii -f 5. The results of the last 2 expe- 

 riments are exactly the same; and that of the first is hardly at all different from 

 them. From this is clearly seen how little the experiments hitherto published, 

 about the difference of common air, are to be depended on. In general, the air 

 changes from one time to another; so that the differences between them are far 

 greater than those of the airs of different countries, or different heights; for 

 instance, he found that the air of London, in the months of Sept. Oct. and 

 Nov. 1778, when treated with the nitrous air, gave u — 6, ii -f 15, which is 

 a mean result of many experiments which differed very little from each other. 

 The 26th of Nov. last, he found the air for the first time much better, for it 

 gave II— 12, II H- 12; but the 14th of Feb. the air gave 11 — 18, 11 -f- 7 

 whence it appears, that the air of this 14th of Feb. was better than it had been 

 for 6 months before. There can be no doubt of the accuracy of the experi- 

 ments, because he compared the air taken at different times with that which he 

 had first used in the month of Sept., and which he had preserved in dry glass 

 bottles accurately stopped. Now if the formulae expressed above are compared 

 together, it will be found, that the difference between the first terms is of 12 

 parts, and that between the latter of 7 ; that is, of -^ and -^ of the whole 

 quantity of air: which are much greater differences than those above-mentioned. 

 Yet he could not perceive any particular change of health, or facility of breath- 



VOL. XI v\ 4 D 



