APPENDIX. 



319 



One-day box chronometer, No. 3290 designated E i 



One-day 3291 fI in 



One-day 3292 G ( one box. 



One-day 3293 Hj 



Z had been going at the Observatory for many months, and had 

 preserved a very regular rate, but all the others were quite new, and 

 had scarcely settled to a steady rate when I received them. 



In addition to the above I was furnished with a pocket chrono- 

 meter, No. 553, by Mr. Murray. This watch had been at the Ob- 

 servatory for several months, and had performed remarkably well : and 

 before I sailed Messrs. Parkinson andFrodsham intrusted to my care 

 for trial a pocket chronometer. No. 1048, that was only completed in 

 time to be sent to me two days before the Expedition sailed from Ply- 

 mouth. Mr. French also lent me a pocket-watch to use for observ- 

 ing with, in order that the rest might not be unnecessarily moved. 



In the Beagle were three excellent box chronometers. Two by 

 Messrs. Parkinson and Frodsliam, Nos. 254 and 228, which had 

 been used in the Polar Voyages ; and the third. No. 134, made by Mr. 

 M'Cabe. 



The means, therefore, that were placed at my command to effect 

 this most interesting object were tolerably ample : and the result 

 will prove how admirably these machines are adapted to measure 

 such differences when a great number are employed ; because the 

 irregularities and errors of individual watches are compensated for 

 by employing the mean of the whole. 



In the observations for the determination of time, a: sextant by 

 Troughton, No. 1140, and an artificial horizon, were the instruments 

 used: and the mode, whenever it could be adopted, was that of cor- 

 responding altitudes. Occasionally, however, absolute altitudes were 

 used, but only in those places where the latitude was correctly 

 ascertained ; — and in some instances the chronometers were rated 

 by a transit instrument. 



The chronometers were always compared with the 'journeyman' 

 watch before and after the observations, and when corresponding al- 

 titudes were observed all the watches were compared at noon. Their 

 rates Vv^ere carefully observed before sailing from one port, as well as 

 after the arrival at another ; and in calculating the acceleration or re- 

 tardation of their rate of going, the correction was obtained by inter- 

 polation, upon the supposition of their having changed gradually. 

 Whenever it appeared, by comparing tiie watches with each other) 



