128 ti 
the ulgé¢braic sum for the whole piston is zero, corresponding to an oscillation which travels up and down 
in the piston. If pressure has been applied to the end of the piston long enough for a wave to travel up 
and down the piston a number of times, the second movement will be very sinall compared with the first. 
The energy of movement (2) is gradually dissipated, a small fraction veing communicated at each oscillation 
to the water in contact with the exposed face of the piston, in which it produces alternate condensations 
and rarefactions, but this dissipation of energy does not involve any loss in the algebraic total momentum 
of the piston, which is entirely comprised in movement (1), and it is only this momentum that the gauge is 
required to measure. 
When the pistons of a gauge are driven in, the air in the gauge is compressed and offers a certain 
resistance to the movement. It is necessary to show that this. effect is negligible. As a typical example 
one may take the case of a GZ gauge with 1§-inch pistons; the volume of air in this gauge is 8 cubic 
inches ; if,the gauge is hung at a depth of 35 feet in the water the volume of air is reduced to 4 cubic 
inches ; when the six pistons are driven in, the volume of air is further reduced to 3 cubic inches; if this 
second compression is assumed to take piace adiabatically it absorbs 3-05 foot-pounds of energy; the 
greater part of this work, 2°45 foot-pounds, is supplied by the normal external hydrostatic pressure, and 
it is only the difference, 0-6 foot-pounds, that is taken from the kinetic energy of the pistons, amounting 
to a luss of 0-1 foot-pound by each piston ; this is a very small fraction of the energy of the piston, which 
in practice was generally from 5 to 10 foot-pounds. In most cases the error due to the cushioning effect 
of the air was less than in the above example ; it was generally smaller than | per cent. 
When a gauge of the kind described in Section 18 is put more than 50 or 60 feet deep, the water 
inside the gauge rises above the tops of the pistons. The gauges can, however, be used at greater depths 
if a few pellets of lead-sodium alloy (6 parts of lead to 1 of sodium by weight) are dropped into the 
air-chamber. This alloy reacts with water to generate hydrogen, and so keeps the air-chamber full of gas. 
The same device was used with the gauges described in Section 19, when these were at a depth of more 
than 40 feet. 
