44 INTRODUCTION. 



berger's Introduction to the Geographical Deter- 

 mination of Places." The determination of the 

 longitude is made partly by chronometers, and 

 partly by lunar distances ; with respect to the 

 first, it is well known that the best instruments of 

 this kind, when they are exposed to a considerable 

 and continued change of temperature, gradually 

 alter their rate of going, and it seems too as if the 

 effects of the cold or heat upon the watches do 

 not become observable, till after some days have 

 elapsed, which is probably to be ascribed to a gra- 

 dual coagulation, or an incre..sed fluidity of the 

 very small portion of oil which these instruments 

 still require, even when the friction is diminished 

 by means of diamonds. When there are two or 

 more good time-keepers, a daily comparison of 

 them is very advisable, partly to determine the 

 longitude by each of them, and particularly be- 

 cause, from the change of their daily difference, 

 the time may be discovered at which they began to 

 alter their rate of going. By degrees one is able 

 to fix in each of them the direction, and in some 

 measure the amount, of the change of their daily 

 going, for a certain increase or decrease of the 

 temperature of the air ; and, hence, to deduce the 

 most probable correction of a longitude found. 

 Lunar distances, however, are still the best 

 means of determining the longitude ; only they 

 must be observed in considerable numbers, and, 

 if possible, with instruments that magnify power- 



