790 
DR. T. R. ROBINSON - ON THE DETERMINATION OF 
rubbers, which becoming condensed by the pressure tend to act like a hard body. On 
laying down the curve of this coefficient, it looked so like an equilateral hyperbola 
with coordinates parallel to its asymptotes, that I tried the Armagh values of it by 
the equation — B being the load on the brake in ounces — 200 grains, the 
F 
force required to bring the rubbers into contact ; each value of — gives an equation of 
condition, combining which we get values of the constants. Substituting these in 
F 
each, we get values of — whose errors enable us to approximate still more by the 
tZF civ clu 
equation — = db?+— — Thus we obtain y=315*52 ; x=7 7'814; u=6" 662. 
^ B B + u (B + %) 2 j 
F 
These give when B=0, -=125 ‘28 ; when B=co , 77*81 ; and afford values of F differ- 
ing in most cases from the observed ones far less than the probable errors of the latter. 
I have therefore used these values in reducing the observations. The following table 
gives these : their differences from the observed ones, the number of observations, 
F 
and — for each value of B. 
-D 
Table IV. 
No. 
On Brake. 
Computed 
0 — c. 
No. of 
Friction 
Friction. 
Observations. 
B 
1 
oz. 
3 
284-75 
+ 2-3 
7 
11303 
2 
6 
574-3 
+ 13-1 
6 
105-85 
3 
9 
841-7 
-18-2 
4 
96-42 
4 
12 
1098-1 
-14-0 
7 
93-51 
5 
15 
1347-6 
+ 11-2 
5 
93 45 
6 
18 
1593-5 
-15-S 
5 
89-95 
7 
24 
2026-9 
+ 13T 
4 
86-66 
8 
30 
2556-0 
- 8-5 
5 
86-24 
9 
36 
3031-1 
+ 7-5 
5 
85-50 
3. A lateral friction is produced by the wind pressing the anemometer’s axis against 
its outer bearing. Omitting the consideration of the arms, this pressure will be the 
sum of the mean pressures on the cups during a revolution, and the same reasoning as 
in the case of I. shows that it =eV' 2 — 2/cVT+ev 2 . If, therefore, the constants of I. be 
determined by the observations, the effect of this friction will merely be to diminish 
a and /3 and to increase y ; e and k are larger than the other, and k much less than e, 
so that probably 2 kV'v — ev~ is small in comparison of the first term, and the pressure P 
is simply = eV ,s . This is not merely confirmed by these experiments, but a good 
measure of it is obtained. In these the force which turns the vertical axle is ij>D — the 
frictional resistance of the driving apparatus : this when the motion has become 
uniform, acting at the end of the horizontal arm=air’s resistance =P at the point of 
