ON EUDIOMETERS. 291 



the mercury. The only danger is that the face may be a 

 little damaged by pieces of broken glass. You will notice 

 a slight film of moisture on the tube when the explosion takes 

 place, and that is all. a You see the diminution of the volume, 

 which has to be measured. Before the measurement takes 

 place you must leave this eudiometer standing for at least 

 three-quarters of an hour, so as to acquire the temperature 

 of the air. Then you come into the room, measure off the 

 temperature and calculate the result exactly as I described 

 before. One precaution which I did not take in this case 

 was to properly coat with mercury the surface of the india- 

 rubber, which is best done by putting upon it a few drops of 

 a solution of corrosive sublimate. This causes the mercury 

 to adhere to the india-rubber, and then there is no danger, 

 when contraction has taken place, of any air being sucked 

 up,. In order to save time I did not take this precaution, 

 and it has had the advantage of forcing me to describe one 

 point which is essential. When the explosion took place, 

 it sucked up a small quantity of air from the surface of 

 the india-rubber, and you saw the bubbles rising. All 

 these processes require very great precautions. One thing 

 necessary is to have the hydrogen very pure. The pro- 

 cess employed by Bunsen for the preparation of pure 

 hydrogen for the determination of oxygen is by an 

 electrical process, in which the positive pole of the battery 

 consists of some amalgam of zinc. This tube contains a 

 platinum wire connected with one pole which is bent 

 round below which will be covered with the amalgam of 

 zinc, the rest of the tube being filled with dilute sulphuric 

 acid. The negative pole, from which the hydrogen is 

 evolved, consists of a plate of platinum. The apparatus 

 is surrounded by a vessel of water which prevents a rise 

 of temperature, and the gas when evolved is allowed to 

 pass through this washing apparatus, which may be filled 

 with sulphuric acid so that the gas is obtained in a dry 

 state. The hydrogen can be prepared pure enough for 

 ordinary purposes by the action of dilute sulphuric acid 

 on zinc in a small apparatus, from which the air must, of 

 course, be carefully expelled before the gas is collected. 

 Occasionally it may happen that the gas which is being 

 analysed contains so little oxygen that its mixture with 

 hydrogen will not produce an explosive mixture, and in 



u 2 



