GRAVITATION. 79 



a great many curious effects.* If a vacancy is anywhere 

 produced, the surrounding air is forced by this pressure 

 violently into it. If the air is removed from a bladder, 

 the sides are forced together, and no effort can separate 

 them so as to leave within an empty space. When the 

 piston or box of a pump rises, it brings up with it the air 

 within the pump ; the load of air upon the water around 

 it forces the liquid up into the space thus left. When 

 the air is exhausted by a suitable apparatus from a thin 

 glass vessel, its sides will often be crushed inwards by 

 the pressure of the surrounding air. 



It was well known that the atmosphere would rush 

 with violence into any vacant space long before the facts 

 were referred to the right cause, and there was a long 

 and obstinate controversy among the philosophers, wheth- 

 er the phenomena in question were really owing to the 

 weight and pressure of the air, or to What one party called 

 Nature's abhorrence of a vacuum. This controversy was 

 at last settled by experiments made upon a certain moun- 

 tain in the south of France, by which it appeared that 

 the tendency of the air to rush into the vacant space was 

 decidedly less upon the elevation, than at the ordinary 

 level of the ground. Now, as in ascending an eminence, 

 we pass above a considerable portion of the atmosphere, 

 it was natural that what remained above the summit, 

 should press less heavily than the whole. The difference 

 of the effects was therefore very easily accounted for, on 

 the supposition that they were both owing to the pressure 

 of the air ; and as it was absurd to suppose that Nature's 

 abhorrence ef a vacuum would be less upon a mountain, 

 than in a valley, the advocates of this latter theory gave 

 up the point.f 



The gravitation of the air may also be proved, as it often 

 has been, by a very simple experiment. A vessel, filled 

 as usual with air, is weighed. The air is then removed 

 by an air pump, and the vessel, now empty, is weighed 

 again. The difference, which is very sensible, shows 

 the weight of the air which had been removed. 



It may be at first imagined that there are some excep- 



* See No. II., p. 54. t See No. III., p. 41. 



