Narrative of the surveying voyages of his majesty's ships Adventure and Beagle (vol.2- Appendix): between the years 1826 and 1836 : describing their examination of the southern shores of South America, and the Beagles's circumnavigation of the globe

APPENDIX. 329 ments of meridian distances, time was invariably obtained by series of equal, or corresponding altitudes of the sun ; observed by one and the same person with the same sextant, and the same artificial horizon, placed in the same manner, both before and after noon. A very good pocket chronometer, carried by hand, in a box, was always used for taking time. In every instance, it was compared with the standard chronometers (the two supposed to be the best) immediately before the morning observation, and again immediately aftei-wards. It was also compared at noon, and before, as well as after the afternoon, observations. This watch* was so well con- structed, that the intervals shown by it betv>^een morning and after- noon observations always agreed with those shown by the standards, (allowmg for their respective rates). Generally speaking, seven altitudes of one limb of the sun were taken, and then the same seven altitudes of the other hmb, for one set of sights, or observations. Three such sets were usually taken, at short intervals, and the mean result used, unless any marked dif- ference occ'-irred, in which case the result of each separate pair of equal altitudes (morning and afternoon) was computed, and the erroneous ones were rejected. Those were considered erroneous which differed much from the majority. Generally, however, there was the closest agreement between the results of single pairs of sights, as well as between those of entire sets. When clouds intervened the series was unavoidably irregular, but the pairs of equal altitudes were always numerous. In a very few in- stances the chronometers were rated by the results of absolute or independent altitudes, taken with every precaution at similar times of day with the same instruments, and by the same observer. In such cases the rates were obtained by comparing together the times obtained by morning observations, or those deduced from afternoon sights ; not by morning and afternoon, or afternoon and morning observations. But the time, considered to be correct, was invaria- bly deduced from equal altitudes, by the method of Professor Inman. At Paramatta, at the Cape of Good Hope, and under the walls of the Royal Observatory at Greenvvdch, we had opportunities of trying whether there was any difference between our time, thus obtained, and that of the respective astronomers ; and I feel gratified in being able * K. Parkinson and Frodsham. No. 1041 e e