Viscosity of a Solution of Saponine. 413 



and for the surface of water, 



MX 753 x -0266 Q Q07 

 _= M 3-927. 



At *1 millim. below the surface these numbers change to 131 



and 4 respectively. At the surface, therefore, the ratio of the 



resistances is 1261 ; and at '1 millim. below it is 33; while in 



8*42 

 the interior it is, as has been shown, ^«, or 1*2. Although, 



therefore, these numbers can only be taken as approximations 

 to the truth, we think that they enable us to make an esti- 

 mate of the magnitude of the resistance offered to a body 

 oscillating in the surface of saponine solution, for which no 

 previous experiments afforded the required data. 



They show that whereas the resistance offered to an oscilla- 

 ting disk, 2 millim. thick, in the surface of water is only 

 about half what it is in the interior, at the surface of a 2-per- 

 cent, saponine solution it is at least 600 times greater than in 

 the interior, but that this ratio is reduced to 16 by immersing 

 the upper surface of the disk to a depth of 0*1 millim. 



Special experiments proved that the logarithmic decrement 

 in air was so small that the resistance of the air might safely 

 be neglected when the comparisons of the various resistances 

 were made as above described. 



Explanation of the Curves. 



Fig. III. is the curve given by the logarithmic decrements 

 obtained from the experiments on water. The abscissae are 

 expressed in terms of hundredths of a millimetre ; they repre- 

 sent the distance of the upper edge of the plate from the sur- 

 face, and are negative when it is above it. The ordinates 

 represent the logarithmic decrement in terms of '0001, the 

 lowest horizontal line corresponding to the value of *0250. 



The gradual increase in the value of the logarithmic decre- 

 ment as the plate is more deeply immersed is clearly shown. 



Fig. IV. refers to the observations made on the saponine 

 solution. In this case the values of the ordinates are to be 

 taken from the small figures. The positions corresponding 

 to the numbers obtained when the disk was in the surface 

 cannot be shown on the scale of the diagram. Fig. V. there- 

 fore has been drawn on one tenth of the scale of fig. IY. 

 To avoid confusion it has been displaced to a convenient dis- 

 tance along the line of abscissae. The enormous increase of 

 resistance as soon as the disk touches the surface is very stri- 

 kingly shown; and it must be remembered that the increase 

 for a very small oscillation would be very much greater. 



