Barometer. 
lv 
doubtful — the needle often has a tendency to read any thing, and is very irregular 
in its motions when lowered by the Ys. (Observer W.) 
October 8. 1843. Needle No. 2. The readings are much better than with needle 
No. 1, but some of them are doubtful ; when lowered by the Ys the needle some- 
times leaps 2° or 3°. (Observer W.) 
October 9. 1843. Needle No. 1. This observation was considered fair till the 
last reading, which might have been taken at any thing from 73° to 75°. 
Same date. Needle No. 2. A fair observation ; some of the readings, however, 
were not certain to about 5'. 
The observations were, in general, so unsatisfactory, that they were discon- 
tinued till Nov. 20 ; afterwards observations were made, which were, in general, 
more satisfactory. 
81. Observations were made on April 18 and May 2, 1843, in different azi- 
muths, in order to determine the correction due to the irregularity of the needle's 
axle, or perhaps to the presence of iron in the vertical circle ; these observations 
have been already given (Table 21. and Table 22, Introduction, 1841-2.) The 
correction deduced was about — 11' for needle No. 1. No correction has been 
applied to the results in this volume. 
82. In 1846, the vertical circle was removed from the instrument and placed 
horizontally, the dip needle was suspended by a silk fibre within the circle, the 
needle and circle being in the same plane, the needle was then vibrated horizontally, 
and the zero of the graduations was placed in different azimuths ; the time of vibra- 
tion was found very little affected by the varying positions of the circle ; it seems 
probable, therefore, that the correction above is due solely to the imperfections of 
the axle. 
§ 9. Barometer. 
83. The barometer is by Newman. The tube is 0*552 inch in diameter ; the 
scale is attached to a brass rod, terminating in an ivory point, which, at each obser- 
vation, is made to meet its image in the mercury of the cistern ; the cistern is about 
3 inches in diameter ; the vernier professes to read to 0*002 inch, and that 0*001 
may be estimated, but the graduation is so inexact as to give changes in error from 
0*002 to 0*003 inch. The barometer was compared indirectly with the standard of 
the Royal Society, by means of one made by Newman for the Duke of Argyle. 
The comparisons of the Duke of Argyle's barometer with the flint and crown glass 
standards of the Royal Society are given Table 23, Introduction, 1841-2 ; they are 
not consistent ; the mean gives 
Correction of the Duke of Argyle's standard barometer to those of the Royal Society 
= - 0 009 inch. 
