S. L. Pen field — Methods for Determination of Water. 35 



to a full white heat. The flame glances off from the charcoal 

 and it is necessary to have the tube carefully screened by 

 asbestus board, in addition to having a wet cloth around the 

 bulbs. After the ignition is finished a glass tube is fused to 

 the platinum to hold it, and the end of the tube is pulled off. 

 The platinum foil can be separated from the glass by plung- 

 ing it while hot into cold water, when the glass readily cracks 

 away. For testing the efficiency of this method of heating, 

 experiments were made with talc from Fowler, !N\ Y. The 

 material was not absolutely pure and, therefore, the degree of 

 accuracy must be determined by the agreement of a series of 

 determinations. 



Talc taken. 



Wa 



.ter weighec 



[. Per cec 



1. -5000 



Heated alone 



•0232 



4-64 



2. -5000 



Mixed with 50 gr. CaO 



•0224 



4-48 



3. -5000 



" " 0-75 " PbO 



•0230 



4-60 



4. 1-0100 



" " 1-50 " " 



•0451 



4-47 



5. '9637 



" " 2'00 " " 



•0426 



4-42 



The talc contained some fluorine, since when heated alone 

 the tube was slightly etched and the water distinctly acid, the 

 percentage in JSTo. 1 is certainly, therefore, somewhat high. A 

 mean of Nos. 2, 4 and 5, 4/46 per cent is probably very nearly 

 correct. The amount of lead oxide to be taken as a flux de- 

 pends upon the fusibility of the mineral. With an infusible 

 compound like talc, an amount equal to twice the weight of 

 the mineral would be near the limit, while with a more fusible 

 one very much less should be used. All that is necessary is to 

 have the contents of the tube fuse so as to insure a complete 

 decomposition, and if the mixture is too fusible it may eat its 

 way through the glass, not only spoiling the experiment but 

 probably injuring the platinum. Lead oxide, taken in small 

 quantities, cannot always be relied upon to hold fluorine. 

 Thus in trying to determine water in topaz, 1 gram of the min- 

 eral was fused with twice its weight of the oxide but the 

 water that came off was very strongly acid. The lead oxide 

 used in these experiments was a good quality of litharge, 

 heated to near its fusing point to drive off water or any vola- 

 tile matter. The material is not hygroscopic, and can be kept 

 for a considerable time exposed to the air without suffering 

 any appreciable increase in weight. 



The intense heat which can be obtained with the charcoal 

 furnace already described makes it possible to decompose a 

 mineral completely by fusion with sodium carbonate in a com- 

 bustion tube, and to collect the water in a weighed sulphuric 

 acid or chloride of calcium absorption apparatus. For this 

 purpose a piece of combustion tubing of 15 mm internal diam- 



