Ii2 On the Longiiitdinal Eapansion [Jan. 



7. — From llie rfite <.>i going- v\Iiicii this table exhibits fi'om January 

 to the end of March, it rn.iy be inferred that the clock movements and 

 working parts were well made, but in the month of April the clock ran 

 wild. On taking it down I discovered that one of the pallets, which 

 were of steel and were screwed on to the arms of the escapement, had 

 got loose ; I made both pallets firm and tight, and put up the clock 

 again ; and its rate of going for the month of Maj^ it will be seen, was 

 beautiful. On the first of June the land winds set in, and the effect was 

 soon felt by the clock, which began rapidly to change its rate of going 

 from minus 0' 5"' daily, to plus 0' 5'' ; 0' 7" ;— 3' 0", and 5' 2". — Find- 

 ing that the hot dry winds had so great an effect on the pendulum, 

 1 paid little attention to the clock afterwards, but let it go on until I 

 should see what effect the damp weather of the monsoon would pro- 

 duce. In the beginning of November a great deal of rain fell, and as 

 might be expected the rate of the clock changed again, from gaining to 

 losing, thereby clearly shewing that the wood I employed for the pen- 

 dulum, though well varnished, and I think well seasoned, vvas so much 

 aflected by the hygrometric changes of the atmosphere as to render it 

 unfit for the pendulum rod of a correct time-keeper. 



8. — There appears to be some doubt,as well as difference of opinion,, 

 as to the successful use of a wooden rod pendulum ; some thinking it 

 nearly as good as the compensating pendulum, and others putting no 

 faith in it. The experiment I have made will determine the point only 

 as far as the particular specimen of wood I employed was concerned, 

 and no further. But, as being closely connected with the subject, I 

 subjoin some experiments on the longitudinal expansion and contrac- 

 tion of wood, occasioned by changes in the atmosphere, which changes, 

 with respect to temperature, alter the rate of a clock's going, when a 

 common metallic pendulum is attached to it, and with respect to dry- 

 ness or humidity, chiefly, when the clock is supplied with a wooden 

 pendulum. The specimens were as follow;5 : viz. 



No. 1. — Dry v/ell-seasoned teak. 



2. A very hard wood, called by the Malabar workmen uH'ei^LDn'La 

 Pal Maram, or milk u'ood. 



3. Dry well-seasoned chittagong wood. 



4. Mahogany, from a clock-case : had been at Madras for half a cen- 

 tury. 



5. * American i)ine or fir had been at Madras 4 or 5 years. 



6. * The same varnished, 



* Both these were t'.vin pieews fiom the ?arae piece of fir ttiat the clock pcnduhira \vn:- 

 inade o f. 



