in the length of the pendulum vibrating seconds. 385 
Unst, on the 22d July, 50 s , 2. The mean of the rates of the 
chronometer at Unst, Portsoy, and Leith, gives i*,8i for the 
mean daily rate, which being multiplied by 57, the number 
of days between the 22d of July and the 17th September, we 
have i“.43*,i7 for the loss of the chronometer during that 
period. This being added to 8*.4i',6, we obtain io m . 24 s ,77 
for the error of the chronometer on the 22d July, for the 
meridian of Leith, and subtracting 50*, 2 (the error at Unst) 
the remainder 9“. 34 s , 57 will be the longitude of Unst, east of 
Leith. Now the longitude of Leith being i2 m .46*,7 west, the 
difference 3". 12*, 13 will be the longitude of Unst, west of 
Greenwich. This agreeing so nearly with the preceding 
result, may perhaps be considered as not very far from the 
truth. 
At Portsoy on the 3d August, the chronometer was found 
to be 7“.52*,3 too fast, by altitudes of the sun, which are 
detailed in the Appendix. The loss of the chronometer from 
the 15th June to the 3d August, at the daily rate of i%8 
is i".28%2 ; which, together with i“.i5%75 (the error of the 
chronometer at Greenwich on the 15th June) being added to 
7“.52',3, we obtain io“. 36',25 for the longitude of Portsoy, 
west of Greenwich. 
In order to deduce the longitude of Portsoy from that of 
Unst, we have the chronometer fast at Unst on the 22d July 
50*, 2, and at Portsoy on the 3d August 7“*52 s ,3. The mean 
of the daily rates at Unst and Portsoy is i*,£i, and the loss 
from the 22d July to the 3d August, at this rate, is 18*, 12. 
Hence we have Portsoy west of Unst 7“. 20*, 22, and the lon- 
gitude of Unst from Greenwich being 3”. 18 s , 87, we have the 
mdcccxix. 3 E 
