204 APPENDIX. 



the sea- work afterwards executed in the ship. I need hardly allude 

 to the facilities, afforded by heights, for making eye sketches of the 

 coast line, and other details, such as the ranges of lulls, forms of 

 banks, &c. Ascending heights near the sea is advantageous in ano- 

 ther point of view ; for not a rock or a shallow escapes notice, if the 

 day is tolerably clear. While in harbour, every place in the vicinity 

 which could be examined in boats, or overland excursions, was ex- 

 plored, as far as our means and time would allow. 



Before I speak of the sea-work, it may be useful to say a word 

 about ' bases/ of four kinds, arranged according to their relative 

 value. 



The first are those derived from good astronomical or chronome- 

 trical observations, made at two stations several mUes apart. 



The second are deduced from angular measurements of small 

 spaces exactly known. 



The third are obtamed by actual measurement with a chain, with 

 rods, or with a line : — 



And the fourth are the rather uncertain bases obtained by sound. 



This statement of the relative value of bases, is only meant to refer 

 to their employment in sea-surveying. I need hardly remind the 

 reader of these notes, that the third description of bases, however 

 exact nominally, requires a host of minute precautions, in addition 

 to what I never found between Valdivia and Cape Horn, namely, a 

 nearly level and accessible space, of considerable length, on which to 

 measure. 



To attain the utmost precision is a laudable endeavour, no doubt, 

 when carrying on extensive trigonometrical operations on land ; but 

 it should be borne in mind, that every hour employed in what is 

 commonly called ' hair-sphtting' — in minute details that do not affect 

 the chart or plan which is the result of a sea-survey, is not only an 

 hour lost, but an hour taken away from useful employment. 



The second kind of bases are so quickly and easily measured, either 

 with a sextant or micrometer, across any kind of land or water, and 

 have been so repeatedly proved in every part of the Beagle's surveys, 

 that I consider them unobjectionable, when used for such limited 

 •operations as making plans of harbours, or fixing the positions of 

 objects only a few miles distant. By multiplying bases, which with 

 such easy methods is soon effected ; and by a frequent use of the 

 sextant, artificial horizon, and chronometer, material errors may be 



