CHAP. XIV.] CHRONOMETRIC MERIDIAN DISTANCES 337 



both journeys, and the intervals are not long, we shall probably 

 get a very good result. 



Travelling rates obtained thus should always, as already 

 remarked, be used when the scale of the chart depends on 

 the observations. 



The method is very simple, and, used for this purpose, none 

 of the considerations of temperature, etc., hereafter mentioned 

 need be thought of, as the time is short. 



We now come to the consideration of the rates to be used other 

 on other occasions, especially when voyages are long, and 

 circumstances change much during them. 



This is a very wide subject, and besides the fact that it 

 has already been fully discussed by Captain Shadwell, in his 

 masterly treatise before referred to, neither space nor the 

 intention of these pages permits our going very far into it, and 

 we shall content ourselves with giving general descriptions 

 of cases, together with formula) for them, with just sufficient 

 reasons to allow of their being understood. 



The whole question rests on : What makes chronometers 

 vary ? 



The labours of many observers show us that the answer is : Why 



1. Imperfection in the workmanship of the watch. change"^ 



2. Changes of temperature. their 



3. The quality of the oil in the pivots, and its age {i.e., the 

 time elapsed since the watch was last cleaned). 



4. Accidental shocks or vibrations imparted to the watch. 



A supplementary question may be asked : Which of these 

 is the most important ? To which the general answer is that, 

 according to circumstances, any one may be. 



1. Imperfection of Worhnanshvp. — For tliis manifestly there imperfec- 

 is nothing to be done. A badly made chronometer will go so 

 erratically that we shall soon lose confidence in it, and reject it 

 from all results, returning it as soon as we can. There are, 

 however, but few chronometers that pass through the hands of 

 the Royal Observatory which will come under this head, and 

 doubtless many a chronometer has been classed in this category 

 from ignorance of the circumstances of its compensation, and its 

 resulting variation under change of temperature. If on a 

 voyage during -which tetnperahire is uniform a chronometer 

 placed with others, under the same conditions of protection 



22 



