50 



Professor H. S. Hele-Shaw 



[Feb. 10, 



The model (Fig. 3) represents on a large scale a section of the 

 channel already shown, in which groups of particles of the water are 

 indicated by round balls, lines in the direction of flow of these 

 groups (which for convenience we may call particles) being coloured 

 alternately. When I move these so that the lines are maintained, 

 we imitate " stream-line " motion, and when, at any given poiut of 

 the pipe, the succeeding particles always move at exactly the same 

 velocity, we have what is understood as " steady motion." 



Fig. 1. 



Fir.. 2. 



As long as all the particles move in the straight portion of the 

 channcd, their behaviour is easy enough to understand. But as the 

 channel widens out, it is clear that this model does not give us the 

 pxoper distribution. In the model, the wider portions are not filled 

 up, as they would be, with the natural fluid ; for it must be clearly 

 understood that the stream-lines do not flow on as the balls along 

 these wires, passing through a mass of dead water, but redistribute 

 themselves so that every particle of water takes part in the flow. 



