THE SUCTION-PUMP 



61 



You see pumps were used long before they were explained. 

 It remained for Torricelli to explain them, and also to dem- 

 onstrate the limits of atmospheric pressure. He did this 

 by means of an experiment with mercury, which, being a 

 much heavier liquid than water, does not require so long 

 a tube to measure the atmospheric pressure against it. 

 This experiment may be repeated as follows (Fig. 27): Fill 

 with mercury a strong glass tube closed 

 at one end. The tube should be about 

 thirty-two inches long. Invert the 

 tube, closing the open end with a finger 

 (A). Immerse this end in a dish of mer- 

 cury and then remove the closing finger. 

 The mercury then falls, leaving a vac- 

 uum at the top (B). But if the experi- 

 ment is well performed, the mercury 

 will fall only a short distance. It will 

 stop at about thirty inches above the 

 surface of the mercury in the dish. 

 This thirty-inch column of mercury is, 

 then, a measure (in mercury) of at- 

 mospheric pressure. Thus you see that 

 the Torricelli experiment is actually the construction of a 

 mercurial barometer. 



Pascal, a great French mathematician (1623-62), rea- 

 soned that if the mercury in the Torricelli experiment is 

 really supported by atmospheric pressure, then the column 

 should become shorter at a higher altitude. So he per- 

 formed the experiment at the top of a high tower in Paris 

 and found a decrease in the column which corresponded 

 to his computations. Soon afterward this principle was 

 confirmed by performing the experiment on a mountain. 



FIG. 27. Diagram illus- 

 trating Tonicelli's ex- 

 periment. 



