On the Viscosity of a Solution of Saponine. 407 



The brass rod by which the plate was suspended could be 

 unscrewed and replaced by others of different sizes. 



The plate was suspended in the centre of a circular glass 

 dish 15 '5 centim. in diameter and 8 centim. deep. This was 

 fitted with a wooden cover, through which a thermometer was 

 introduced. The whole was surrounded by a case having glass 

 sides, which served to ward off air-currents. The time of the 

 oscillations was measured by a stop-watch indicating quarter- 

 seconds. 



Three rods were used in turn to suspend the plate. They 

 were of the same length, but of different radii. 



The radius of (1) was *1337 centim. 

 (2) „ -2360 „ 



(3) „ -3362 „ 



The scale on which the beam of light was reflected was 

 about 2*5 metres from the mirror ; the maximum half-amplitude 

 was about 200 millim. ; so that the half-oscillation was 

 about 2°'3. 



The moment of inertia of the apparatus was determined by 

 means of a brass ring of the same external diameter as the 

 plate, and was found to be 753*09 (C.Gr.S.). 



The plate was first carefully levelled, and distilled water 

 poured into the vessel to within a few millimetres of the bot- 

 tom of the plate. Water was then added with a pipette until 

 the bottom of the plate and the surface of the water touched. 



By a simple calculation the difference in height caused by 

 the addition of a measured quantity of water was determined. 

 An addition of 2 cubic centim. of water was found to cause a 

 rise of "1 millim. in the level. 



The elevation was of course increased when the disk was 

 in the surface; but the change is allowed for in the calcu- 

 lations. 



The temperatures of the air and of the water were carefully 

 noted at the commencement of each observation, and were 

 found to remain nearly constant. To attain this end the 

 experiments were conducted in a cellar, and special pre- 

 cautions were taken to prevent any marked variations in 

 temperature. The variations therefore did not exceed 1° C; 

 and the small corrections thus rendered necessary were made 

 by means of the table of the values of the coefficient of 

 viscosity of water at different temperatures given by Grotrian 

 (Pogg. Ann. Bd. 157, S. 242). All the observations were 

 thus reduced to 16° as a standard temperature. The time of 

 oscillation was ascertained by observing the time of 10 swings. 



2G2 



