232 
MR. J. J. MANLEY ON THE APPARENT 
added to B to establish equilibrium ; the beam was then duly fatigued, and the 
differential bolometer placed within the beam case read; if the reading = 0, then, 
with the aid of a reading-telescope, a first set of observations of the oscillating pointer 
was taken, and the temperature of the air surrounding the beam noted; after this, 
the bolometer was again read ; if the difference between the first and second bolometer 
readings did not exceed 1/5,000° C., then, and then only, was the set of observations 
retained for calculating the resting-point, B.P., of the beam. 
July 18, 1910. 
Table I. 
Vessel A in left pan. 
Temperature of f before observation = 17°’40 C. 
beam case „ = 17°‘40 ,, 
Mean R.P. reduced for 18° C. = 114 "2. (1). 
Other reduced R.Ps. similarly f =115-1. (3). 
obtained during the same< 
day were h = 114-7. (5). 
Vessel A in right pan. 
Obser- 
vation.s. 
Pointer 
readings. 
Obser¬ 
vations. 
Pointer 
readings. 
R.Ps. 
Obser¬ 
vations. 
Pointer 
readings. 
Obser¬ 
vations. 
Pointer 
readings. 
R.Ps. 
No. 1 
164 
No. 2 
68 
115-5 
No. 1 
152 
No. 2 
28 
89-3 
„ 3 
162 
71 
116-0 
„ 3 
149 
,, 4 
31 
-5 
„ 5 
160 
6 
72 
115-5 
,, 5 
147 
„ 6 
33 
-3 
„ 7 
158 
„ 8 
74 
-5 
„ 7 
144 
„ 8 
37 
-5 
„ 9 
156 
„ 10 
76 
-5 
„ 9 
140 
„ 10 
39 
-0 
„ 11 
154 
12 
78 
-5 
» 11 
138 
„ 12 
42 
-8 
„ 13 
152 
„ 13 
137 
Mean = 
115-6 
Mean = 
89-4 
Temperature of f before observation = 18°-42 C. 
beam case ]_ after ,, = 18°-44 „ 
Mean R.P. reduced for 18° C. = 90-4. (2). 
Other reduced R.Ps. similarly f =90-8. (4). 
obtained during the same< 
day were 
t = 90-5. (6). 
Taking the sensibility of the balance = 45, we find from (1) and (2) the difference in the weights of 
the two vessels to be 
114-2-90-4 . 
- = 0-264 mgr. 
2 X 45 ^ 
Using convenient lifters, the positions of the vessels A and B were now interchanged 
and the beam again fatigued ; observing the precautions and same method of procedure 
as before, new data were obtained for calculating the R.P. In general, the data 
secured both before and after reversing the positions of the two vessels enabled us to 
make six independent calculations of the R.P. The average value deduced from each 
set of six R.Ps. was then, by means of tlie temperature coefficient of the balance, 
