324 PROCEEDINGS OF THE ACADEMY OF [1880. 



A mean of these 14 determinations gives for hj'groscopic water 

 3.17 per cent. 



Subtracting tliis fx'om the mean total water, 7.48 per cent., we 

 liaA^e for water of crystallization 4.31 per cent., an amount closely 

 agreeing with that deduced from the writer's experiments. The 

 desiccation over sulphuric acid in Exp. I is for too short a time to 

 completely extract the hj^groscopic water. 



The exact state of the water cannot yet be regarded as certainly 

 established. There is no reason why a fixed temperature of 100° 

 C. should divide the hygroscopic water from the water of crystal- 

 lization. The above experiments show that the loss of water as 

 the temperature is raised above that point is a very gradual one. 

 It is difficult to see in what manner the water driven off at 190° 

 in experiment Y, differs from that driven off at 100°. It will be 

 seen hereafter that much of the water can be driven off without 

 exfoliation. Again, there is no sufficient reason why some of the 

 water absorbed by sulphuric acid in the desiccator may not be 

 water of crystallization. It has been long known that sulphate of 

 copper either at 100° C. or in a desiccator over sulphuric acid 

 loses much of its water of crj'stallization. M. Damour has shown 

 that chabazite loses nearl}^ half of its water in a desiccator. It 

 seems probable that Philadelphite, with other vermiculites, holds 

 its water in a similar manner. From the experiments here de- 

 tailed it would seem that we maj" define water of constitution to 

 be the more closely combined, and hygroscopic water the less 

 closely combined water of cr3-stallization ; and the distinction 

 between the three states of water then becomes a theoretical rather 

 than a practical one. 



Temperature of Exfoliation. — The temperature at which exfo- 

 liation occurs is from 1,50° to 160° C. It has been found that the 

 exfoliation temperature is proportional directly to the original 

 volume of the substance, and inversely to the rapidity ^Df the ap- 

 plication of heat. The larger the piece experimented upon, the 

 higher the temperature necessary to make it exfoliate, and the 

 more rapidly the heat can be applied, the sooner will it exfoliate ; 

 as the following experiments will show. 



(a) Yery small fragments heated on a watch-glass in an air-bath 

 began to exfoliate at 150° C. 



{h) A large piece heated similarly did not exfoliate at 210° C. 



(c) A piece was immersed in melted parafflne. At 100° C. 



