200 K W. Morley— Moisture in Gas. 



within three or four months may perhaps be thought sufficient. 

 I therefore give some account of them, referring to the previ- 

 ous paper for further explanation of the principle of the 

 method. 



Part of the apparatus used is shown in the figure, c consists 

 of three parts fused together : a Liebig's bulb tube filled with 

 strong pure sulphuric acid, to effect the preliminary drying of 

 the gas ; a tube filled with powdered potassium hydroxide, to 

 remove vapors of sulphuric acid ; and a tube filled with phos- 

 phorus pentoxide mixed with glass wool, and having a plug of 

 cotton wool at the exit of the gas. dd is a tube used as a 

 counterpoise for the tube efgh, and having very nearly the 

 same volume and the same surface as the latter, efgh consists 

 of four parts. In f air dried in c is made to take up a small 

 quantity of moisture: it contains moistened calcium chloride. 

 In e, at the beginning of an experiment, is contained about a 

 gram of water: the size and length of the tube connecting e 

 with / are used to adjust the absorption of water by the cal- 

 cium chloride to the other circumstances of the experiment. 

 At g, the tube connecting/ and h consists in part of a very 

 fine capillary tube. The diameter of this tube is reduced by 

 fusion till the pressure of the atmosphere will force two or 

 three cubic centimeters of air through it in a minute, provided 

 that a vacuum be maintained at one end. In h is contained 

 phosphorus pentoxide mixed with glass wool, and having a 

 plug of cotton wool at the exit of the gas. i also contains 

 phosphorus pentoxide: it is fused to a tube leading to a large 

 reservoir connected with an air pump. 



During the passage of a current of gas through the apparatus, 

 dd and efgh have to be connected to each other and to the dry- 

 ing tubes c and i. Caoutchouc connectors would be a fruitful 

 source of error in an experiment of such delicacy as the pres- 

 ent. Whether the fact be that it is impossible to make caout- 

 chouc tubes absolutely dry, that moist air diffuses through 

 them, even against a greater internal pressure, or that gradual 

 oxidation produces water as well as carbon dioxide, I have 

 found it impossible to use such connectors when it was neces- 

 sary to avoid introducing water into the apparatus at a rate 

 of even a decimilligrarn in a day. All connections required 



