84 THE APPLICATIONS OF PHYSICAL FORCES. [BOOK i. 



turns out the debris into the boat outside. While this goes on the 

 stonework slowly built on the upper part of the caisson presses it 

 down by its weight and forces it to descend until it arrives at the 

 required depth. Then the workmen leave the caisson, the three 

 chimneys are filled with cement, and the foundation is finished. 



The bridge of Kehl is formed of two abutments and four piers : 

 the two extreme piers each rest on four caissons ; the two others, 

 on three caissons only. 



It must be added that work in chambers where the air is at such 

 great pressure is not without danger to the health of the workmen. 



V. MEASURING HEIGHTS BY THE BAROMETER. 



The experiments made by Pascal in 1648 at the foot and at the top 

 of the Puy de Dome, and those also made by himself at the top and 

 at the base of the tower of St. Jacques la Boucherie at Paris, were 

 intended to determine whether the pressure of the atmospheric 

 column of air was really the true cause of the rising of the mercury 

 in Torricelli's tube. As the new theory came out victorious from 

 the experiment, an important application of- the barometer was 

 realized. 



It is evident, indeed, that a barometer may be used to measure 

 heights, and that by noting the two different points to which the 

 column of mercury rises at two stations of unequal altitudes, from 

 the difference of the two levels, the difference of the two altitudes 

 ought to be determined. This, of course, supposes that the relative 

 densities of air and mercury are known, and that this density does 

 either not vary at all, or varies in a determined ratio, in the thick- 

 ness of the stratum which separates the two stations. 



Let us imagine ourselves in a place where the temperature of the 

 air is 0, the barometric pressure 760 millimetres, which is near the 

 mean pressure at the level of the sea in the south of England. 

 In these conditions the mercury, with equal volumes, weighs 10,500 

 times more than the air. A barometric height of 1 millimetre of 

 mercury is equivalent then to a column of air of 10,500 millimetres, 

 or 10 m ' 5, on the hypothesis that the successive strata of air do not 

 vary in density or temperature. This, however, is not the case. The 



