GRAVITY MEASUREMENTS. 
59 
same room, but sufficiently distant to not affect each 
other. The temperature and pressure coefficients were de¬ 
termined at Washington by swinging under different con¬ 
ditions in the manner before described, and the flexure co¬ 
efficient was computed from the theoretical formula. The 
flexure was always measured, using the same appliances as 
with the A apparatus and a weight of 1.5 kilograms. The 
corrections to the observed periods were applied in the same 
manner as with the half second pendulums, substituting the 
following coefficients: Temperature, 0.00000235; pressure, 
0.000000066; flexure, 0.00000037. The periods were re¬ 
duced to the standard conditions—arc infinitely small; 
temperature, 15° C.; pressure, 40 millimeters at 0° C., sidereal 
time, and inflexible support. The mean period of the 
three pendulums at Washington in June, 1894, was s.2502400 
and in November s.2502392. This change may be due in 
part to the fact that the knife-edge was nearly new when 
used in June. The following are the results for gravity as 
given by each pendulum, with the mean for the three and 
the differences from the mean of each : 
Station. 
g (in dynes). 
Differences from mean. 
Pendulum 
C 1. 
Pendulum 
C 2. 
Pendulum 
C 3. 
Mean of 
three pend¬ 
ulums. 
C 1. 
C 2. 
C 3. 
Washington... 
980.098 
Chicago. 
980.246 
980.254 
980.247 
980.249 
—.003 
+.005 
—.002 
Pikes Peak. 
978.940 
978.941 
978.955 
978.945 
— 5 
— 4 
+ io 
Denver. 
979.595 
979.591 
979.592 
979.593 
+ 2 
— 2 
— 1 
From the agreement of the pendulums and the accord of 
the results with those obtained with the more elaborate ob¬ 
servations with the half second apparatus, it is indicated 
that quarter second pendulums may give results of nearly 
the same order of precision as larger instruments. Their 
smallness, however, renders their use inconvenient in some 
respects, and the increased labor of observing resulting from 
the necessarily shorter swings tends to offset the gain in 
