114 



same tube and under similar conditions, tliereby preventing an\' reabsorption 

 of moisture while cooling. 



Determinations were made on from one to three grams of coal, giving results 

 of which the following are fair examples: 



The gain in weight of the calcium chloride tube corresponds with a slight 

 difference to the loss in weight of the coal, giving per cent, of moisture, which 

 agreed veiy well with separate determination of moisture in the same coals 

 by the ordinary method of heating for one hour in a toluene bath, results being 

 slightly higher, due perhaps to current of air passing over coal, removing moisture 

 more completely. 



The gain in weight of the calcium chloride tube exceeded the loss in weight 

 of the coal more than what would arise from the combustion 6f any volatile 

 hydrocarbons present, as indicated by carbon found and calculated as methane. 

 This excess of water found may indicate presence of free hydrogen, but a correct 

 explanation of its presence, as well as of the source of carbon found, can not be 

 made without an accompanying analysis of the gases given off from the coal at 

 lOO'^ C, as the small percentage of carbonic oxide, carbon dioxide and hydro- 

 carbon volatilized at 100^ varies widely with different coals. The per cent, of 

 volatile hydrocarbons given off in these experiments — .0299c to .0779c — calculat- 

 ing as methane, is probably higher than occurs in the regular determination of 

 moisture because of the air current, while on the other hand the somewhat lower 

 temperature of these experiments would tend to modify the difference. As in no 

 case, however, does the loss of volatile matter other than water nearly reach one- 

 tenth of one per cent., the error in calculating the same as moisture is of no prac- 

 tical importance. 



]S^OTES ON DiPHENVLSELENON AND SeLENTHREN. By RoBERT LyON. 



