Limits of 
accuracy. 
548 ASTRONOMICAL OBSERVATIONS. 
Amplitude was S.E.bE.^E. Required tlie True Amplitude, and the 
error of the compass. 
Month. Day. h, m. secs. 
Date at place, A.T., January 5th. 16 44 27 
Correct for long, in time 36° 34' (tab. 17, Raper) + 2 25 36 
Gr.D., 5th ... 19 10 2 
Sun’s declin. (p. i., N.A.) . 22 39 49S, and decreasing. 
Hourly difference. — 5 16 
Corrected declination. 22 34 33S. 
O / // - 
Declination. 22 34 33 S. Sin. 9*584210 (Tab. 68, Raper.) 
Latitude. 37 59 0 S. Sec. 10*103369 
True amplitude, E. ... 29 8 30 9*687579 (Tab. 68, Raper.) 
Compass bearing, E.... 28 7 30 
Error of compass . 1 1 0 E. 
As this subject of Variation is somewhat confusing, it will be found 
convenient to adopt the following plan :— 
Draw a cross, marking the four cardinal points with their initials; 
on this, project the true and magnetic amplitudes ; then, turning the 
point where they meet in the centre towards you, if the true amplitude 
is seen to the right of the magnetic amplitude the variation is E., but if 
the true amplitude is to the left the variation is W., as in figures:— 
E. variation. W. variation. 
In the example above the true amplitude is to the right of the 
magnetic amplitude, and the compass variation is E. 
The foregoing examples are those of observations which have been 
taken in an artificial horizon ; in such cases, as has been shown, it is 
always necessary—1st, to apply the index error of the instrument 
to the observed angle, and then to divide the angle by 2. 
In conclusion, any one who has had some previous experience in the 
use of the sextant and artificial horizon, ought, by means of such 
observations, to fix his position within one mile of Latitude, and two 
miles of Longitude, always supposing that he is acquainted with the 
index error of his sextant, and the error and rate of his watch. 
