VI 
APPENDIX. 
The ships having arrived in Davis’ Strait, using the temporary rates 
assumed from the five weeks of previous trial, a sufficient number of 
lunar observations were obtained during the lunations of June and 
August, to entitle the mean result to considerable confidence ; being 
in all 1,209 distances, whereof 640 were with the sun west of the moon, 
and 569 with the sun east of the moon ; the difference of each chronometer 
was noted on the mean Greenwich time, thus determined, at noon of the 
22d of July, being the middle day of the observations ; and these differences 
being compared with those at midnight on the 6th of May, the gain or loss 
of each chronometer in the interval of seventy^six days and a half became 
known, and consequently their corrected daily rates on board ; the detail 
of these comparisons will be found in a memorandum annexed to the 
abstract of lunar observations in June and August, 1819. 
The daily comparison of each chronometer with all the others, had also 
served in this interval, to guide the selection of those which had gone most 
steadily and uniformly, and had best preserved a mean rate, to be used 
in the determination of longitudes ; these were Nos. 228, 254, and 259 
of Messrs. Parkinson and Frodsham, 25, 369, and 404 of Arnold ; No. 228 
was also chosen as the watch by which the time of all observations should 
be noted, its rate being small and very uniform. 
No other opportunity of lunar observation and consequent comparison 
with Greenwich time occurred to any extent, until the ships were secured 
for the winter, about the end of September, in a small harbour in Melville 
Island, which had been passed early in the month, but to which they were 
obliged to return ; it had fortunately happened, that Captain Sabine had 
landed on a low point within three miles of this harbour, when passing 
it on the 6th of September, and had ascertained its longitude by the six 
chronometers above-mentioned ; this point was sufficiently near to be 
included in a very careful survey of the harbour and adjacent coast, made 
by Captain Parry in the spring ensuing, whereby the bearing and distance 
of the point from the ship’s winter station was correctly determined, and 
its chronometrical longitude thereby brought in comparison with its true 
longitude, deduced by a mean of 6,862 lunar distances observed during 
the winter, and carried to the point of land by means of the survey. 
The detail of this comparison will also be found in the memorandum 
alluded to ; the error of the chronometrical longitude, using the corrected 
rates ascertained by means of the lunars of June and August, proved in 
