VISCOSITY OF WATER BY THE EFFLUX METHOD. 121 
The volumes given by Poiseuille have therefore been multiplied 
by the factor 13-5697/13-5674 = 1-00017. 
Pressures.—Pressures under 15 centimetres of mercury or 200 
centimetres of water were measured by a water-manometer, those 
above that amount by a mercury-manometer.’ The relative 
density of mercury to water at 10° ©. was assumed by Poiseuille 
to be 13:576981.2 I have taken it as 13°5746, and since the effect 
is to make Poiseuille’s pressures under 15 centimetres too small, 
have multiplied them by the factor 13°5770/13-5746 = 1-00017. 
In the case of tube /’ the uncertainty of the times of efflux is 
relatively so large that the application of this correction would 
be futile. 
The acceleration of gravity for the place where Poiseuille made 
his experiments is given by him as 980°8 centimetres per second.® 
Taking the absolute density of mercury at 10° C. to be 13-5711, 
and of water to be 0:99975, we get for the absolute pressures FF; 
by multiplying these numbers into the value of the acceleration 
£2 138100 Ay = 98050 Ay icieicnsicas (33) 
hm being the pressure expressed in centimetres of mercury, and 
Aw in centimetres of water, both at 10° C. 
Dimensions of tubes and bulbs at different temperatures.—The 
dimensions of the tubes and bulbs used by Poiseuille were evalued 
a the temperature 10° C.4 Four of the tubes were tested by 
filling with mercury and finding the area of the section from the 
Volume.5 The results were then compared with the direct 
Measurements of the diameters of the sections, on the assumption 
that their contours were ellipses, for it is on this assumption and 
m the direct measurement alone that the dimensions of the 
tubes, may be found. In the illustrative case given by Poiseuille 
er 
1 
‘ay cit., p. 460 (note), and § 65, p. 4 
id. p. 448~ 452: table 11, p. 475; i § 16, p. 532. 
: Tid. p. 531, 
This is oily stated by Poiseuille, vide § 83, p- 497 of his memoir: 
See also § 32 457. 
ips Pp. 457. 
