d.

Mr. Crosley had been frequently unwell during the passage from Madeira;
and after trying the effect of a few days on shore, he decided to remain
at the Cape of Good Hope, and relinquish the expedition. The instruments
supplied by the Board of Longitude he agreed to leave in my care; after
having consulted with the commander in chief upon the subject, and
received his approbation. The loss of the astronomer was severely felt by
me, both from being deprived in the surveys of his more accurate
observations, and from being called upon to supply his place so far as
was in my power. The duties of commander joined to the occupation of
surveyor, left little time for other employment; but through an increase
of effort, and with the assistance of my officers, I hoped to carry on
the surveys and fulfil the most essential parts of the instructions from
the Board of Longitude, at the same time. Of these instructions, Mr.
Crosley permitted me to take a copy.

SUNDAY 1 NOVEMBER 1801

The rates of going with mean solar time of the four time keepers
committed to my charge, were deduced by Mr. Crosley from three days
observation of equal altitudes, with a sextant and quick-silver horizon,
between the 21st and 27th of October. These rates, which he left with me,
I extended to November 1, by equal altitudes taken on that day; and their
respective errors were deduced by allowing 1h 13' 40.47" to be the
longitude in time of Simon's Bay. *

[* In 1763, Mr. Mason determined the longitude of his observatory in Cape
Town, from the transit of Venus, to be 18 deg. 23' 7" east; and the
difference of longitude from thence to Simon's Bay, by the Dutch survey,
is 2' 00" east.]

Earnshaw's No. 543,
slower than mean Greenwich time at noon h ' " "
there Nov. 1, 0 14 35.33 and losing 5.33
No. 520, 34 16.62 15.84
Arnold's No. 176, 50 59.29 8.96
No. 82 -------- -----
No. 1736, watch, faster 21 20.03 17.27

The watch was intended to be taken up rivers, and to such places as the
ship did not go; and in order to gain some knowledge of its probable
performance, I wore it five days in the pocket. Its rate of losing during
that time, was from 11.59" to 8.79" per day; so that upon the average, it
lost 7" less in the pocket than when in a fixed situation; for