DESCRIPTION OF A NEW HYGROMETER. 493 



ters, and there is reason to think that they will not 

 be subject to any alteration in their scales. 



Charcoal, from its known durability, likewise pre- 

 sented itself as a proper substance for the above 

 purpose ; and it was found, upon trial, to be sensi- 

 bly hygrometric, although its range was very limit- 

 ed. 



Most of the above substances were formed into 

 hollow cylinders, and cemented to the end of ther- 

 mometer tubes ; and their expansibility was tried 

 by filling the cylinder and tube with mercury, in 

 the usual manner. 



But the substance which was found to possess by 

 far the most delicate sensibility, and extensive range, 

 was the internal membrane of the Arundo Phrag- 

 mites. A small bag, made of this membrane, is at- 

 tached to the lower end of a thermometer tube, so 

 as to form, as it were, its bulb. It is then nearly 

 filled with quicksilver, which rises and falls, in con- 

 sequence of the contraction and dilatation of the 

 membrane, by any change of moisture ; and these 

 changes are indicated upon a scale attached to the 

 tube, the zero of this scale marking absolute humi- 

 dity, and the other extremity of the scale absolute 

 dryness. The lower end of the glass tube, instead 

 of being merely inserted into the top of the bag, 

 may pass through it, the quicksilver in the bag 

 communicating with that in the tube by one or 

 more openings made through the sides of the tube. 

 By this means, the bag is supported by the glass. 



