CHAP. XIV.] CHRONOMETRIC MERIDIAN DISTANCES 341 



In our opinion the variation of rate arising from these causes 

 is, Avith the generahty of Admiralty watches, the larger pro- 

 portion of the total change. 



The only notice that can be taken of variation of rate due to 

 this, is to consider it as detracting from the general value of the 

 meridian distance ; and the nature of the passage, whether 

 rough or smooth, should therefore be noted in the returns. 



Magnetism is another disturbing cause to which irregularities 

 of chronometers have been referred. As no trustworthy con- 

 clusion as to this has been arrived at, we do no more than 

 mention it. 



It follows as a matter of course, from the preceding observa- 

 tions, that not only will the rate of the clu^onometer as ascer- 

 tained before leaving a port be different to that found on 

 arrival at another port, but that the sea rate for the interval 

 will probably be different from either of them. 



We have now to consider the means at our disposal for 

 approximating to the true rate under different circumstances. 



The most satisfactory circumstance under which we can interpo- 

 determine meridian distance (after those already described) between 

 is when, having left a port A, called at B (the position we Places 

 Avant), where we have only stayed long enough to get Error, Longitude 

 and eventually arrived at K without fiu-ther stoppage, the i? estab- 

 longitudes of A and K are sufficiently well known to take them 

 as secondary tneridians. 



In this case, by applying the known difference of longitude 

 between A and K to the observations at A, we fmd the Error 

 on mean time at K at the epoch of starting from A. The 

 difference between this and the Error ascertained on arrival at 

 K, divided by the duration of the voyage, will give us a fair 

 ^ea rate, which we shall assume to be uniform and invariable 

 during the voyage. A simple application, then, of accumu- 

 lated rate up to the time of observations at B, will give us the 



certainly demands more space, both for the swinging of the box and to 

 allow of free access for handling the watches. 



It was very convenient for comparing, as one watch could be held to the 

 ear while the eye took the time by the standard. 



It seems probable^ also, that the temperature would be more constant, from 

 the fact of the watch being embedded in thick soft material. The lid of 

 each box was also stuffed softly, and, when in place, pressed on the glass 

 of each watch, excluding aU air. 



