670 THE POPULAR SCIENCE MONTHLY. 



have abundant reason for great leniency in judging of the work of 

 these really marvelous machines. 



Remembering these things, we shall be prepared to accept the 

 stories about the wonderful running of some watches with more or 

 less allowance. That watches sometimes seem to go with very slight 

 variation for long periods of time is often accounted for by the fact 

 that the accelerating and the retarding effects in the carrying of them 

 have nearly counterbalanced each other. A good watch may be so 

 nicely regulated as to keep a mean time between its variations, which 

 will be very accurate. I have heard of two watches in the course of 

 my life for which their owners claimed that they varied only four 

 seconds a month, and I can not help adding that they sought to touch 

 the regulator just once more to overcome even this variation. With 

 one of these men I had a personal interview, and succeeded in draw- 

 ing out that, after all, he was not telling what his watch varied, but 

 only how it stood at the end of a month ; for aught he knew, it might 

 have ranged a minute or two either way during the time. The other 

 man I never met ; but I trust these are the only men in the world who 

 ever imagined that a pocket time-piece can be made to have a uniform 

 variation of less than a second a week. But such men serve as an ex- 

 ample of the misfortune it really is if one possesses too good a watch. 

 When a man gets to reading the time on his watch by the second- 

 hand, he is likely to feel discouraged and out of joint with the world 

 much of the time, for his watch will not bear such scrutiny. A far 

 happier man is he who can only afford a poor old " turnip " which, like 

 Sam Weller's, must be set twice a day. A man who possesses one of 

 these wonderful watches, supposed to run to the second, could hardly 

 feel worse to find himself coming down with the cholera than to find 

 that without seeming provocation his watch has gained or lost twenty- 

 five seconds. 



I add a few remarks upon the use and care of time-pieces, their 

 visits to the jeweler, etc. 



It is to be observed that there are many unscrupulous men working 

 at the bench who have very little real knowledge of their trade, and 

 who, to make a living, must make the most of what they get to do. 

 Such men uniformly declare that any watch which comes into their 

 hands to be repaired needs cleaning, which will cost you from a dollar 

 to a dollar and a half. Sometimes this is true, and sometimes it is not. 

 If you have a stem-winder in a close-fitting case, it is probably not 

 dirty, and yet, if it has run two years or so, it ought to be oiled. This 

 most jewelers will do, if you ask them to, for a small fee. If I owned 

 a fine watch, my practice would be to take it to a jeweler, and have it 

 thus oiled once a year, having it cleaned perhaps once in three. 



If a jeweler tells you that there is some very serious trouble or 

 break in your watch, which it is going to cost several dollars to get 

 repaired, ask him to take the watch *' down," as he terms it, and let 



