INFLUENCE OF GRAVITY 83 



particles in pure water do not touch one another at any time, each 

 particle being surrounded by a lifpiid layer. This layer is destroyed 

 by the addition of salts. 



To use a somewhat homely illustration, the colloidal particle 

 may be likened to a morsel of bait dropped into the water of a 

 river estuary. The moment that it reaches the water it is pushed 

 to and fro by a multitude of hungry small fish. The velocity 

 and amplitude of the oscillatory movements of the bait depend 

 principally on the size of the bait and on the energy with which 

 it is attacked. 



{(') Distribution of Particles. If a fine suspension of gamboge or 

 mastic be kept imdisturbed at constant temperature for some time, 

 Perrin found that there was a distribution of the particles under 

 the influence of gravity. At the bottom of the container will be 

 found a denser distribution than at the higher levels. This is 

 exactly similar to the decrease in the density of the atmosphere 

 with height above sea level, and Einstein argued that the distribvi- 

 tion of suspended particles with height should follow the law which 

 governs the density of the atmosphere with height. Perrin 

 proved by experiment that this was true. In one experiment 

 with mastic at four different levels 12/x apart he found 116, 

 146, 170 and 200 particles per unit. For the same levels the 

 following values were calculated : 119, 142, 169, 201. After this 

 adjustment of concentration to level has been reached, no other 

 change seems to take place. While this is true for the compara- 

 tively coarse suspensions used by Perrin, and even for finer 

 suspensions when examined in very thin sheets of liquid, it has 

 definitely been proved to be untrue for colloids, and even for 

 emulsions when the portions of the liquid within about lOO^u, of a 

 surface are neglected. In the body of the liquid gravity seems to 

 play no part in the arrangement of the particles, the concentration 

 being uniform throughout the non-surface portion and remaining 

 so for an indefinite length of time. This stability is due to the 

 electrical properties of the dispersed material. 



Study of the optical properties of colloidal solutions leads one 

 to the conclusion that the individual dispersed particles, although 

 they are too small to be seen by the microscope luider any power, 

 i.e., cause no obstruction to the larger light waves, are still able to 

 cause deviation of the ripples of light. One would, therefore, 

 expect that they would show colligative properties indicative of 

 more sluggish particles than those in simple molecular solution. 

 Briefly, their osmotic pressure, activity of diffusion and power of 

 lowering the vapour pressure of water would be low, and their 

 viscosity and their resistance to the passage of salts through them 



6—2 



