ME. J. B. HANNAY ON THE MICRORHEOMETER. 
279 
riments. Tlie time was marked by a stop-watch, the bulb being filled to above the 
mark, and the watch started when the level of the fluid passed the upper mark, and 
again stopped as it passed the under mark— 
131-2 + 131 ■3 + 131-2 + 131-2 + 131-4 + 131-4 + 131-3 + 131'3 + 131-4+131-2 /, ^-10 
= 131'29". 
It will be seen that this method gives very concordant results, the deviation in these 
experiments being only xg-th of a second, and that most likely due to personal causes. 
The pressure was then halved—that is, only half a metre of water was used—when 
the following times were obtained :— 
263'0 + 262-9 + 26 2’8+ 263 - l + 263-2 + 263H + 263'1 + 262-9 + 263-0 + 263'l"-M0 
= 263-02". 
As this number is nearly double the former result, the tube may be said to have been 
capillary, according to the above definition of that term. 
The probable error of the mean of the above experiments, calculated by the formula 
•6745 a/ 
amounts to '060" in the first series, and -256" in the second. It appeared that before 
going into the examination of complex organic bodies, it would be better to attempt to 
gain some knowledge of the causes of the phenomena presented by the microrheosis 
of liquids, by thoroughly examining one liquid, to be used as a standard and as a 
menstruum, and then examining a series of comparable inorganic salts about whose 
constitution a good deal is already known, dissolved in the standard menstruum. 
Water was the liquid chosen. 
Poiseiulle had noticed that as the temperature rose the rate of passage increased, 
and at first thought that the loss of density was one of the causes of this; but on 
examining water at temperatures below 4° he found that the retardation increased 
very rapidly, whereas the density decreased. Graham also determined the rates for 
water at different temperatures, but offered no remarks on the phenomena presented. 
As neither of these investigators determined the rate above 70°, I undertook another 
examination from 0° to 100°. The temperature was regulated by a large glass water- 
bath, as shown in Plate 35, fig. 2; and at least 10 experiments were done at each 
temperature. As a pressure chamber, a large glass reservoir of about 20 litres capa¬ 
city was used, placed very close to the microrheometer, and connected therewith with 
tubes of pure caoutchouc (such as toy balloons are made of), of only -g-th inch bore. 
By this arrangement no leakage occurred, and a very slight addition to the air in the 
reservoir was required after reversing the order of the stop-cocks. The reason for 
using glass as a reservoir was because it is not nearly so sensitive to slight variations 
of temperature as copper (used by Poiseiulle), and one has always time to regulate 
the pressure, even when the temperature of the laboratory is quickly rising or falling. 
