278 
MR. J. B. H ANN AY ON THE MICRORHEOMETER, 
In his earlier experiments upon gases, Graham gave to the passage of gases through 
capillary tubes the name “ transpiration,” which expresses admirably what is meant; but 
when he came to work with liquids he still used the term “ transpiration,” although the 
laws relating to liquids were quite different from those relating to gases; and as the 
term “ transpiration ” is not translatable into other languages, being already hi use (as hi 
French, for “perspiration”), I have decided to abandon the title, and substitute for it 
“microrlieosis” (/ri/rpos and peco), and to call the instruments above described “microrheo¬ 
meters.” The microrlieosis of a liquid, then, is the time taken for the passage of a 
certain volume of liquid through a tube of such dimensions that the rate of flow varies 
as the pressure, the standard of comparison being in the meantime water. 
It appeared probable at the outset that the friction in the microrheometer might be 
due, in a great measure, to the cohesion of the liquid ; and this led me to seek for a 
method of determining the cohesion of all liquids (which lias been published else¬ 
where),* but I soon found that the two phenomena were not comparable—in fact, the 
cohesion had almost no effect in retarding the flow. I need not quote the work here, 
as there will be plenty of proof further on to show the real cause of the retardation. 
In this paper I shall confine myself to the work I have done on saline solutions, and on 
the substance (water) in which the salts are dissolved. The tube and measuring bulb 
with which t-liese experiments were done had the following dimensions. The capillary 
tube had an average diameter of 0'0938 millim., and deviated only ’002 from a circle. 
Its length was 21 millims., while the capacity of the glass bulb was 4’0530 cub. centims. 
Both bulb and tube were made of soda glass, whose expansion had been determined by 
weighing with mercury at different temperatures, with the following results :— 
Coefficient from 0° to 50° . . . '0000252 per degree. 
„ „ 50° to 100° . . . -0000259 
In these experiments the temperature was measured by a thermometer, graduated 
to tenths of a degree, which had been compared with a standard. The pressure 
of the air in the chamber was at first regulated by an air-pump, but I found that I 
could regulate it much more accurately by blowing’ with the mouth; and this method 
was generally used when the pressure was not over 1 metre. In the apparatus shown 
in Plate 35, fig. 2, it will be seen that during one-half of the duration of the flow the 
liquid in the bulb from which it is flowing will be at a higher level than that in the 
other bulb, thus adding to the pressure ; but as during the latter half of the flow it is 
below the level, the one effect neutralizes the other. The pressure was measured by a 
water-gauge, and the temperature of the water in the column noted, and when it 
differed from 15° a correction was made either way. In using fig. 3, Plate 35, the 
pressure was altered to take into account the difference in the level of the two bulbs. 
At a temperature of 20°, and under a pressure of 1 metre of water, it was found 
that it took 131 'S'' to empty the bulb as an average of the following 10 expe- 
* Trans. Roy. Soc. Edin., vol. xxviii., p. 697. 
