VISCOSITY OF WATER BY THE EFFLUX METHOD. 143 
capillary. It is almost unnecessary to say that the water used in 
the flow should be most carefully freed from organic particles. 
The pressure and temperature, which should be sensibly constant, 
ought to be read at intervals during the efflux, and also the height 
of the surface of the water in the heating vessel above J and J at 
its commencement and termination, the latter in order to afford 
material for correcting the pressures registered in the manometer. 
The terminals A B of the capillary must be set exactly horizontal, 
say by means of a cathetometer, and the parts K to A and Bto # 
firmly connected’ with the heating vessel, in such a way however 
as to permit of their adjustment—not shewn in Fig. 4. 
hd 
The tube BZ serves to ensure constancy in the terminal con- 
dition of the efflux. It should be mentioned that the method of 
allowing the effluent to drop from the tube terminal is unsatis- 
« 
The corrections to the immediate data of an experiment are :— 
(#) Corrections of the thermometers. 
(6) Reduction of the pressure registered in the manometer on 
account of the pressure conditions in the apparatus. 
In regard to (a) if we propose a precision of a higher order than 
lina 1000, we must read to -001 degree C., and therefore must 
- follow the methods of rigorous thermometry. With respect to 
{b):—if the bulb V be symmetrical in form, then since the general 
height of the water in the bath is not sensibly altered by the 
addition thereto by the efflux, we have—putting / for the height 
_ of the water surface above J at the beginning of the experiments 
_ 4nd Xi’ for the similar height above J at the end—for the mean 
__ Pressure 
Px fon te 1). ae ee sh sr 
P=g (nH-p@*") (45) 
H being the mean value of the sustained column in the manometer, 
ee 
* See Mém. des Say. é : tt 
2 . . trang. t. 9, § 10, p. 440. Also Hagen, Abhandl. 
Akad. Wiss, Berlin, 18, 
P. 398. 54, p. 17 ete.: and Meyer, Wied. Annal., Bd. 2, 
