34 S. L. Penjield — 2£ethodsfor Determination of Water. 



The greatest deviation from the theory in this series is in No. 

 8 which shows a loss of 0*10 per cent. 



In the experiments with gypsum mixed with calcite it can- 

 not readily be told how much C0 2 is expelled, since it is not 

 known to what extent the calcite is decomposed. The only 

 correction that has been made therefore is for the time that 

 the tubes were left open. 



Gypsum 

 taken. 



•2668 



•3503 



1-0000 



Calcite 

 taken. 



•1000 



•2000 



1-0000 



H 2 



calculated. 



•0558 

 •0733 

 •2093 



H,0 



weighed. 



•0549 

 •0712 

 •2084 



Weight 

 corrected. 



•0558 

 •0721 

 •2093 



Error. 

 •0000 

 •0012- 

 •0000 



The results thus far given are very satisfactory, and they 

 represent all the determinations made after the method was 

 perfected. In the case of minerals, however, which give off 

 water only on intense ignition, it was found that the tubes 

 could not be heated sufficiently over the blast lamp. A long 

 series of experiments was made with talc, chondrodite and 

 staurolite, using lead oxide, bismuth oxide and a mixture of 

 lead oxide and sodium carbonate in the tubes as fluxes, but 

 the results were not satisfactory. It was evident that a more 

 intense heat must be obtained, and accordingly the following 



method of heating was de- 

 6. vised. A sort of oven 



was constructed of fire 

 brick lined with pieces of 

 charcoal,* fig. 6. The best 

 tube to use with this furn- 

 ace is one like fig. 3, made 

 from combustion tubing. 

 To protect the glass the 

 lower end is surrounded 

 by a cylinder of platinum 

 foil. This is applied by 

 first bending the metal about a tube considerably smaller than 

 the one to be used, and then springing it over the end of the 

 ignition tube. If properly adjusted the spring of the plati- 

 num will hold it securely in place. The tube is held in the 

 furnace as shown in the figure. An additional piece of char- 

 coal is placed on top of the tube and the blast lamp is turned 

 to a horizontal position so that the flame plays upon the side 

 of the apparatus. The temperature that can be obtained by 

 this method of heating is very intense, bringing the apparatus 



* Where fire brick are not at hand ordinary brick will answer, but they soon 

 crack with the intense heat. The charcoal that is most convenient to use is that 

 which comes prepared for blowpipe purposes. 



