On Sal Nitriim and Nitro-A'erial Spirit 87 



other. It must not be supposed here that the vapours 

 expired by the animal occupy the top of the glass and 

 exclude the air from that part ; for these vapours 

 soon condense and adhere to the sides of the glass ; 

 for otherwise the underlying water would be depressed 

 by them. But it would be reasonable to think that 

 the particles of air expelled from the lungs of animals 

 become lighter, because the nitro-aerial particles are in 

 part removed from them, and that they rise in conse- 

 quence to the top of the glass ; and that, being more 

 densely crowded there, they are capable of resisting 

 the pressure of the air below and of excluding it, but 

 are nevertheless unfit to sustain life ; while at the 

 same time the air at the bottom of the glass remains 

 unchanged and the animal placed there is still able to 

 breathe. It is also worthy of notice that when a small 

 animal, say a mouse, is shut up in a glass and suffers 

 from want of air, it turns its mouth hither and thither 

 in an upward direction in quest of breath ; but when 

 it perceives that it suffers more there from want of 

 breath it is wont to bring its mouth downwards, and 

 when it gets a little refreshment there it pushes its 

 mouth as far down as it can into the glass and keeps it 

 there. 



As an animal, so also a lamp expires sooner when 

 placed in the upper than in the lower part of the 

 glass, although this should perhaps be ascribed in part 

 to the smoke which occupies the top of the glass. Nay, 

 if an inverted bell-jar be suspended in the air and then 

 a lamp from which scarcely any smoke proceeds be 

 .placed in it, you will soon observe the lamp going 

 out, because the air contained in the glass is soon 

 rendered incapable of sustaining fire on account of the 

 burning of the lamp. But as it is lighter than the rest 

 of the air, the surrounding air forces it upwards and does 



