ON PURE AIR 193 



the vessel with a new mixture. With these experiments I 

 continued daily the whole of January, filling the glass often, 

 even in the night-time, with new air, but constantly found 

 the same proportion of pure air contained in our atmosphere. 

 Sometimes the water rose a little higher, sometimes it was 

 lower ; but this seemed to depend on the rise or fall of the 

 barometer and thermometer. I was led to think it sufficient 

 to make my experiments for the following months only four 

 times a week. In the month of February the air appeared 

 to be the same as in January, but on the 23rd of March the 

 water rose only to No. 8, which was the more remarkable, 

 as the cold had increased, and the barometer had risen. The 

 19th of April the water rose to No. 10, though neither the 

 thermometer nor barometer suffered any change during that 

 time. Thus the air continued till the 21st, after which the 

 water remained every day at No. 9. In May and June, 

 between Nos. 8 and 9. The 30th of June it rose to No. 

 10. The whole month of August it was between Nos. 8 

 and 9. But from the 3rd to the 15th of September at 

 No. 9. The 6th of October it again rose to No. 10. It 

 afterwards kept, during a continuation of very tempestuous 

 weather, between Nos. 8 and 9 till the 4th of November, 

 when it rose no higher than to No. 8. The same was the case 

 on the 5th of November, though the barometer rose very 

 high. After this it kept between Nos. 8 and 9. On the 

 10th it rose to No. 10, when the barometer was as high as 

 the 4th and 5th. The 21st it rose only to No. 8. After- 

 wards it kept between Nos. 8 and 9 till the 8th of December, 

 when it rose to No. 9, the barometer being low ; but after- 

 wards, till the 31st, it rose no higher than between Nos. 

 8 and 9. 



Our atmosphere therefore contains always, though with 

 some little difference, nearly the same quantity of pure or 

 13 



