IsSO.] 



NATURAL SCIENCES OF PHILADELPHIA. 



32T 



carbon resting loosel}' in a perpendicular position between the 

 mica and the centre of the weights connected them so that an^' 

 expansion of the mica would lift the weights from off the ring 

 on which thej^ rested. The blowpipe flame was now directed 

 from one side upon the mica. 



G, retort stand. 

 b, ring. 

 (', support. 



d, weight. 



e, pencil of chalk. 

 /, fragment of mica. 



In the following table of experiments, the first column repre- 

 sents the weight of the fragment of mica, and the second column, 

 the iron weight which was lifted by the exfoliating mica. 



Philadelphite. Weight. 



15 grains lifted 10 lbs. avoirdupois. 



readily. 



In the last experiment the four-pound weight was lifted up and 

 thrown ofl" the ring supporting it ; the weight lifted being 56, Ot 

 times the weight of the mica. 



A remarkable motive power is here deA'eloped. That it is 

 OAving solely to the escape of the combined water is shown by the 

 tact that if the weights are so arranged that the mica can onlv 

 slightly expand, and, after heating, are removed, the mica will 

 expand no more, or very slightly more, upon further application 

 of heat, the water having been in great part expelled. If the mica 

 is confined under a weight so heavj' that it is impossible for it to 



