ULTRA-FILTRATION 75 



Evidence tending in the same direction and towards the same 

 limits of size is afforded by experiments initiated by the classical 

 series of ultrafiltrations of Bechhold. Membranes of known 

 permeability are prepared, i.e. the diameter of the pores is known, 

 and the colloidal solution is filtered through these by pressure. 

 A series of filters is tried till one is obtained which has the smallest 

 pores which will allow the colloid to pass through. Obviously 

 the particles must be smaller than the pores, and probably, though 

 not necessarily, they are larger than the next filter in the series. 

 The sizes of particles obtained in this way are in reasonable 

 agreement with the values obtained from ultramicroscopic 

 calculations. 



The little dots of light seen under the ultramicroscope are not 

 at rest. They dart about hither and thither in a seemingly 

 inexplicable way. According to the kinetic theory of matter, a 

 fluid was assumed to be made up of molecules in a state of very 

 rapid motion and having a mean free path intermediate between 

 that of a solid and that of a gas. The colloidal particles in the 

 liquid are hustled into motion by continuous collision with the 

 rapidly moving molecules of the liquid. If the particles have a 

 natural period of vibration which is a multiple of that of the water 

 molecules their amplitude of vibration will be increased (e.g. by 

 suitably timing blows on a pendulum its excursion can be increased 

 to a considerable extent. Each blow need be very slight). 



This motion of the particles, while a very striking feature in 

 the field of vision of the ultramicroscope, is not characteristic of 

 colloidal solutions. Particles sufficiently small to be influenced 

 by the high velocity bombardment of the molecules or ions of the 

 solvent may still be well within the limits of visibility under an 

 ordinary microscope. This movement owes its name to its 

 discoverer Brown, a botanist, who described the peculiar oscilla- 

 tion of pollen grains suspended in water in 1827. This Brownian 

 movement may be seen by means of an ordinary microscope in a 

 solution of the water-colour gamboge especially when the dia- 

 phragm of the microscope is almost closed. The rate of move- 

 ment is independent of the chemical nature of the particles, but 

 depends on three factors, viz. (a) the size of the particle, (b) the 

 temperature, and (c) the viscosity of the dispersion medium. The 

 rate is increased by decrease in the mass of the particle, by increase 

 in temperature or by decrease in the viscosity of the medium. 

 The movement persists, never changing, once equilibrium has 

 set in, for all time. It has been observed in granite and in other 



