1S80.J NATURAL SCIENCES OF PHILADELPHIA. 321 



(3) The finely powdered mica was divided into two portions, 

 one of which Avas spread out on an open w^atch-glass, the other 

 placed in a crucible. Both were weighed, put in a desiccator over 

 sulphuric acid, and let stand unopened for two months. That in 

 the crucible lost 2.76 per cent, of water. That on the watch-glass 

 had lost 3. 87 per cent. On standing 3 or 4 minutes upon the 

 scale-pan it gained .53 per cent, of water from the air. Upon ex- 

 ]iosure over sulphuric acid in the desiccator 24 hours longer and 

 then being quickly weighed, it was found to have lost 3.99 per 

 cent. It Avas now placed in an air-bath and kept at a temperature 

 of 100° C. for 4 hours. After cooling 15 minutes in the desiccator, 

 it was found to have gained in weiglit about ^ per cent., indicating 

 that the desiccation over sulphuric acid was more complete than 

 that in the air-bath at 100° C. That in the crucible lost on igni- 

 tion 5.97 per cent, of its weight. 



(4). The powdered mica w.as placed in a watch-glass in a desic- 

 cator over sulpliuric acid. 



Aftev27 days it had lost 2.28 per cent. 

 " 40 " " " 2.36 " 



During weighing; it was enclosed in clasped w^atch-glasses. It was 

 now^ put in a crucible and ignited. 



The dried mineral lost on 1st ignition, 5.18 per cent. 

 u u a 2d " 5.36 " 



" " " 3d " 5.47 " 



(5). The following direct determinations of water of crystal- 

 lization w^ere made from the mica, dried in a glass tube, corked 

 while weighing, and then ignited in a crucible. 



Tliis determination is thought to be too high, including some 

 iiygroscopic water, since the mica in a tube cannot be perfectly 

 desiccated. 



A mean of the three determinations of hygroscopic water ab- 

 sorbed over sulphuric acid gives 3.24 per cent., which deducted 



