35 



without it, it is not too much to say that there would have been no 

 better time-keeping than could be obtained from a common kitchen 

 clock made more carefully than usual ; at least, until some of the 

 latter inventions of Hardy and others, which might have been 

 more generally adopted if Graham's had not existed." 



The other capital invention of Graham was the " mercurial 

 compensating pendulum," to which I have already alluded. Its 

 principle is very easily understood. Galileo's attention is said to 

 have been attracted by the swinging to and fro of some lamps 

 suspended from the roof of a cathedral, and he observed the fact 

 that the same lamp performed its oscillations in the same time 

 whatever might be the arc of vibration. Further investigation 

 revealed another law, viz. — that different lamps swing in times 

 proportioned to the square roots of the lengths of the cords by 

 which they were suspended. The subsequent study of these laws 

 gradually led to the application of the pendulum as a means of 

 regulating the rate of a clock ; and it is probable that for this 

 purpose it will never be replaced by anything more efficacious. 

 The length of a pendulum vibrating seconds in the latitude of 

 London is 39-6 inches; and if it were possible to make a pendulum 

 on which variations of temperature had no effect, it would be 

 theoretically perfect, and would be (certain other conditions being 

 complied with,) an accurate method of measuring out time. But 

 there is no substance in nature which is insensible to changes of 

 heat and cold, and therefore it was found that the rate of a clock 

 varied with changes of temperature. In hot weather the seconds 

 pendulum lengthened, and the clock lost time ; in cold weathe r 

 it shortened, and the clock gained. The problem which Graham 

 set himself to solve was this: to construct a pendulum which 

 should practically be insensible to changes of temperature ; and 

 this he effectually accomplished by a contrivance wonderfully 

 simple and beautiful. Lindley Murray tells us that "two nega- 

 tives in English destroy one another." Applying this principle to 

 Mechanics, he righdy concluded that two expansions in opposite 

 directions would likewise destroy each other, with the result that) 

 virtually there would be no expansion at all. Fortunately, he 



