K W. Morley — Volumetric Composition of Water. 279 



cistern containing the jars of gas is capable of the motion 

 required to bring the open end of this recurved tube inside of 

 the jar ab and to its top. When the cistern is in its highest 

 position, the recurved tube is wholly contained within the 

 well seen under the cistern, and the cistern can be moved 

 vertically by a distance equal to the height of the tallest jars 

 used. It is fitted with sliding ways, counter-poised, and moved 

 by a screw ; by means of a multiplying gear, its motion can be 

 made as rapid as is convenient ; by means of the adjustable 

 counterpoise, the cistern with its thirty-five kilograms of mer- 

 cury can be moved up and down and placed accurately at the 

 required level with ease and safety. In the center of the 

 cistern is a well for tilling the jars used : the rest of the 

 bottom is inlaid with a smooth surface of slate. Care was 

 taken to have no crevices in which air could be entangled, 

 from which perhaps it should rise into a jar of gas. The 

 mercury was always kept as much as three centimeters deep, 

 to lessen the probability that gas in the jars should be contami 

 nated by diffusion between the walls of the jar and the 

 mercury in contact with them. But the danger of this con- 

 tamination is very slight. I left two jars of hydrogen stand- 

 ing in the cistern for eleven w T eeks, after which time I was 

 unable to detect any contamination. The reason of this, so 

 different from the result obtained by Faraday,* is probably 

 the great care taken in filling the jars with mercury. The jar 

 was always put entirely under the surface of the mercury 

 while it was closed with a glass plate, so that no dust from the 

 free surface of the mercury could get to the inside of the jar. 

 The vacuum in the pressure tube was obtained by exhaust- 

 ing the tube from above while the bottom was closed, then 

 admitting mercury till it rose above the glass stopper at the 

 top of the tube ; during the exhaustion, the stopper was loosely 

 in place, and was in the vacuum. When the stopper was 

 covered by mercury, it was forced into place ; a drop of water 

 had previously been put in the upper part of the pressure 

 tube. In the pressure tube were two Jolly points ; of which 

 sometimes one was used, sometimes the other, an interpolation 

 formula having been computed for each of the two systems of 

 pressures measured by the mercury when brought to one of 

 the points. The vacuum in the pressure tube was often 

 measured or rather, the real zero of pressures was determined 

 by producing a good vacuum in the eudiometer, bringing the 

 mercury in the pressure to one of the Jolly points, and observ- 

 ing the level of the mercury in the eudiometer ; this level was 

 the zero from which pressures were counted, and could be 



* Annals of Philosophy. [II], vol. xii, p. 389, 1826; Poggendorff's Annalen der 

 Physik, vol. viii, p. 124, 1826. 



