CH. XV. THE THERMOMETER, 119 



place, and observed that the mercury rose above the 30 

 inches whenever the air was heavy, and sunk below when- 

 ever the air was light. When once this was discovered it 

 was easy to mark off inches and parts of inches on the side 

 of the tube so as to reckon how much the mercury rose and 

 fell each day. 



This was the beginning of the barometer^ by which we 

 measure the weight of the atmosphere. It was a long time 

 before people would believe that anything so invisible as air 

 could affect the mercury, but this was at last clearly proved 

 by a man named M. Perrier, who carried a barometer 

 to the top of a mountain called the Puy de Dome, in 

 Auvergne. As the summit of a mountain reaches to a great 

 height in the atmosphere, it has, of course, less air resting 

 upon it than the valley below has, and so the mercury when 

 carried to this height not being pressed so much up the tube, 

 fell nearly 3 inches, and then rose again gradually as M. 

 Perrier came down into the valley below where there was a 

 greater weight of air. This experiment, which was suggested 

 by the famous French writer Pascal, proved beyond doubt that 

 it was the weight of the air which caused the mercury to rise. 



If now, after reading this account, you go and look at an 

 ordinary upright barometer, you will perhaps be puzzled by 

 finding it all enclosed in wood, and you will ask how tlie 

 air can get to the mercury to press it down ; but if you look 

 carefully at the wooden box at the bottom, you will find a 

 small hole in the wood, often having a small plug of paper 

 in it to keep out the dust, and through this hole, even stuffed 

 up as it is, the pressure of the air can act. The space be- 

 tween the top of the column of mercury (b, Fig. 15) and the 

 end of the tube is a vacuum, or a space without any air in 

 it, and is still called a Torricellian vacuum. 



