CHEMICAL APPARATUS AND PRODUCTS. 209 



of air with hydrogen, ignition being produced by the electric spark; 

 this method was improved by Ure, and is now found to be the 

 most accurate one ; and Dobereiner caused the union of the 

 oxygen and hydrogen by means of spongy platinum. These pro- 

 cesses refer entirely to the estimation of oxygen ; the analyses of 

 other gases were made by Hope, Henry, and Pepys, and the 

 whole subject has been brought to great perfection by the 

 researches of Bunsen. 



Bunsen employs two principal pieces of apparatus for the 

 analysis of gases the absorption tube, about 250 mm. long and 

 20 in diameter, and the eudiometer, from 500 to 800 mm. in 

 length and of about the same width. The latter is provided with 

 platinum wires sealed into the top of the tube to permit of the 

 passage of an electric spark to inflame the explosive mixture. 

 These tubes are graduated in millimetres from the closed end 

 downwards. 



The short tube is employed for the absorption of gases by solid 

 re-agents, which are introduced into the tube in the form of bullets 

 on the ends of platinum wires, the gas being carefully measured 

 before the introduction of the bullet and again after its removal. 

 Great care is necessary in the measurement, to ensure perfect 

 uniformity of temperature of the gas, and to determine the pressure 

 under which it is measured. In order to prevent alteration of 

 temperature by the approach of the body, a cathetometer, or a 

 telescope sliding on a vertical rod, is used for reading the height 

 of the mercury in the tube. The explosions are made in the 

 eudiometer by passing a spark from a small Leyden jar or an 

 induction coil between the platinum wires. 



A considerable time must elapse between any manipulation of 

 the tubes and the reading of the volumes of the gases; the 

 absorption by means of solid re-agents is also very slow. To 

 hasten the process Regnault and Reiset devised an apparatus 

 which has since been modified by Frankland and Ward: in 

 this the measuring tube is surrounded by a large volume of 



