Experiments and Discoveries . 75 
first, as found in the usual way without the plate, and then with 
the plate, the ship's head being on opposite sides of the meri- 
dian ; when although some difference (while the ship is changing 
her place) must be expected in the resulting variations, yet that 
change day by day will be but small, and we shall not fail to 
consider those results to be the most accurate, which agree near- 
est with each other.*” Several such examples are then selected 
out of the preceding table, as follows. 
Experiments. 
Difference in 
the variation 
without Plate. 
Difference in 
the variation 
with Plate. 
Experiments. 
Difference in 
the variation 
without Plate. 
Difference in 
the variation 
with Plate. 
3 & 4 
1° 5 ' 
0° 23 ' 
4 0 & 4 1 
1°53 / 
0°40 / 
7 8 
1 56 
0 4 
65 66 
1 25 
0 17 
15 16 
1 0 
0 1 
70 71 
1 35 
0 27 
25 26 
2 41 
1 23 
71 72 
1 28 
0 2 
26 21 
2 54, 
1 30 
75 76 
2 13 
0 14 
29 30 
2 15 
0 53 
80 81 
3 10 
1 24 
34 35 
3 40 
0 31 
These deductions are obviously perfectly satisfactory, as far 
as they go, but they are not a fair specimen of the value of the 
plate ; for although, in the situation selected for the azimuth- 
compass, the local attraction did not exceed 3° or 4°, yet the 
compasses in the binnacle were at the same time at variance with 
each other to the amount of 5°, 6°, or 7°; and as it is by these 
that the vessel is steered, the actual correction in the course was 
to the same amount, as appears from the following example, fur- 
nished by Lieutenant Mudge. 
44 On the 2 2d of May, at noon, we were in latitude 41° 46' 
N., and long, by chronometer 9° 53' W. Taking this as our 
departure, we sailed by the starboard compass S. 46° W. 183 
miles ; this placed the ship on the 23d (allowing the variation 
21° W.) in lat. 38° 58' N., and long. 11° 26' W. Whereas £he 
observation at noon for latitude gave 38° 39' N. and long. 
10° 58' W. So great a difference in 24 hours was attributed to 
a current, till I compared the steering or starboard compass 
with the one with your plate, when I found no less than 7° er- 
ror to be subtracted from the course steered, making the true 
course S. 17° W., instead of S. 24° W., which had been taken 
