276 DR. E. ANGUS SMITH ON AN APPARATUS FOE, &c. 



the direct estimation of the metal when there is a com- 

 plicated mixture. 



Some years ago there was read to the Royal Society a 

 paper of mine on the mode of estimating ammonia, cyanogen, 

 urea, &c., by means of the nitrogen obtained, or decomposing 

 them with chloride of lime or soda. This apparatus gives 

 the power of completing the process, and obtaining much 

 finer results. If even very small quantities are used with 

 this apparatus a correct result may be obtained. If chloride 

 of lime is used it is well not to take too much, as the 

 oxygen given out on boiling must afterwards be absorbed. 

 Water which has been carefully boiled must be used to fill 

 the apparatus. 



Sometimes gases may be collected from liquids by the use 

 of a simpler apparatus still. The flask of water may have 

 over it a cup of oil, into the lower part of which the mouth 

 of the flask projects. Over this mouth may be placed a 

 receiving tube, filled with oil or melted paraffine. When the 

 water boils, all the gas goes directly into the tube, and is 

 immediately cut off from the water below. There are cases 

 in which this may be used with advantage, but in general I 

 prefer the apparatus above described. 



