NEW COMPENSATION PENDULUM. 



215 



the subject, and to endeavour to construct a pendulum, cheap, and 

 which should unite simplicity and cheapness, with the capa- ®^^'*y>^t*J"»*- 

 bility of being easily and accurately adjusted. desirable. 



Wood has been long known as a substance that expands Wood advan- 

 less than any other with heat; and from this property many tageous, but no 

 pendulum rods of time-pieces have been made of wood, and compensate Us 

 found to answer remarkably well ; but it is surprising no ad- expansion. 

 vantage has hitherto been taken of this knowledge, to apply 

 a compensation, which might counteract the small expansion 

 to which a pendulum of wood is liable. 



Wood therefore, if it can be rendered perfectly imper- Wood best for 



v-ious to moisture, appears to be by far the best material, '!^® ^°^' *^"^ 



•^^ *' _ zinc as a corn- 



that can be used for the rod of the pendulum ; and as zinc pensation. 



is a metal which suffers the greatest expansion from an in- 

 crease of temperature, I consider it preferable to every other, 

 that could be employed as a compensation. 



The first step was to ascertain accurately the quantity of Expansibility 

 the expansion of wood, as I could find no experiments on ^i^^d ^^*' 

 the subject at all satisfactory. For this particular purpose 

 a pyrometer was used, which it would be unnecessary here to 

 describe, calculated to receive a rod of wood four feet in 

 length, one end of which was made to act against the shorter 

 arm of a lever, causing the longer to describe an arc, the 

 divisions of which might easily be I'ead off to the thousandth 

 part of an inch. 



A rod of very dry and well seasoned white deal was pro- Rod of whit« 

 cured, free from knots, four feet in length, three quarters of ^^ ^ ^^^ ^ 

 an inch in breadth, and a quarter of an inch thick. Each thick * 

 end was exactly squared, and covered with a thin flat plate 

 of brass. This rod was exposed in an oven to the tempera- 

 ture of 235°, and on measuring it in the pyrometer, it was 

 found to have contracted ; it was therefore replaced in the 

 oven, and sufi^ered to remain a long time till it appeared a 

 little discoloured, in order to dissipate all moisture. The 

 temperature of the oven was then examined, and found to 

 be still 235°. The deal rod was now quickly removed, and 

 placed in the pyrometer, where it remained a sufficient time 

 to acquire the temperature of the room, which was 49*',when 

 the space described in the interval by the long arm of the 

 lever, was registered ; and in this manner by two experi- 

 ment a. 



