﻿254 Method of Investigating Transpiration of Gases. 



From these equations the viscosity can be calculated directly 



or by comparison. 



It will be seen that only a small quantity of gas is required 

 for use with this apparatus, as the bulb C avoids the necessity 

 for preliminary experiments : gases soluble in water can be 

 used, and the volume of gas transpired can be measured very 

 accurately. The vessel B should be bedded firmly in its 

 socket, otherwise its sinking, as it tills with mercury, will 

 cause an increase in the pressure-head. 



This apparatus was first designed by Prof. TTilberforce and 

 constructed here, being originally used by Mr. R. H. Albery 

 for a series of observations on the effusion of gases through 

 holes in thin plates ; the writer has performed a large 

 number of experiments with it, and found it very convenient 

 in practice. In reproducing the apparatus care must be 

 taken that the bulb CJ is not made too small, otherwise the 

 steady state will not be attained before the mercury reaches 

 the lower mark ; for the adjustment of T 3 a preliminary 

 experiment will then be necessary in every case, and one of 

 the chief advantages of the apparatus will be lost. 



An idea of the behaviour of the apparatus may be obtained, 

 in conclusion, from the following figures. The capillary 

 tubes used varied in length from 2S1 mm. to 10 mm., and in 

 radius from 0*2504 mm. to 00964 mm., the radii being 

 measured in the microscope. 



In several experiments at pressures of 37*8 cm. of oil 

 (density 0'873) the pressure varied by 1 mm. • in others at 

 the same pressure the variation was imperceptible, and in 

 some at a pressure of 32*3 cm. it varied bv less than 1 mm. 

 To show the reliability of the regulator T 3 a time of 

 56 seconds was taken six times with variation not greater 

 than 0*3 second, and a time of 75' 5 seconds was obtained six 

 times within the same variations. 



The following illustrates a good set of readings for a direct 

 determination of the viscosity of air : — 



Mean pressure of barometer 741-0 mm. 



Mean pressure of manometer in mm. of 1 ->_ , ~ 



oil (density 0-873) } -' 4 ' 2m 



Volume of air transpired 347-0 c.c. 



Time taken 2420 seconds. 



Effective length of capillary 21 20 cm. 



Mean radius 0*280 mm. 



Mean temperature 1P-75 C. 



From these data the direct determination of the viscosity 

 of air gives 



M = -000180. 



