280 
MR. J. B. H ANN AY ON THE MICRORHEOMETER, 
The following table represents the result of experiments upon water. Each number is 
the mean of 10 experiments, the probable error from 0° to 60° being under ' 2 ", and 
from 60° to 100° under - 4 ,/ :—- 
Temperature 
Time. 
Temperature. 
Time. 
o 
0 
U 
235-0 
O 
30 
106-0 
1 
220-8 
35 
96-8 
2 
211-7 
40 
88-7 
3 
205-0 
45 
81-8 
4 
198-5 
50 
75"5 
5 
192-5 
60 
65-0 
6 
187-1 
70 
57"5 
7 
181-0 
80 
49-8 
10 
167-2 
85 
47-1 
15 
146-0 
90 
45-5 
20 
131-3 
95 
44-3 
25 
115-5 
100 
43-8 
It becomes plainly apparent from this table, as well as from curve No. I., that the 
retardation increases very rapidly as the temperature approaches the freezing point; 
but we know that the contraction of water as the temperature falls ceases at 4°, and 
below that becomes negative. It thus appears probable that we have two separate 
forces at work which act in the same direction in retarding the rate of flow in the 
microrheometer, but which act in opposite directions in the case of the thermal expan¬ 
sion of water, the one being due to the vibrations we call heat, and the other being 
du.e to the gradual arrangement of the water particles to form crystals of ice. The 
thermal force is much greater if we have regard to the space between 0° and 100°, but 
we know that water remains liquid below 0° if undisturbed, and could we cool it till 
its density were as low as ice the crystallic force would then be enormously greater 
than the thermic. An attempt was made to trace the microrheometrical curve below 
0°, but the motion of the water through the tube always caused freezing. Below is a 
formula shorter than Poiseiulle’s, which agrees with the curve pretty well. Let 
6 = temp. C°. and T= time, then 0+6-36 = 158'93('98618) T , which gives— 
perature. 
Found. 
Calculated. 
O 
0 
235 
231 
1 
220-8 
220-6 
2 
211-7 
211-6 
4 
198-5 
196-2 
10 
167-2 
163-4 
20 
131-3 
1300 
40 
88-7 
88-4 
50 
75-5 
74-5 
60 
65-0 
62-7 
80 
49-8 
43-8 
