On Stationary Waves in Flowing Water, 445 



to make, for the surface of the soft carbon might not be in 

 the same state of cleanliness as that of the hard carbon. 



But even suppose we allow that there was better contact 

 with the mercury in the case of the soft carbon than in that 

 of the hard carbon, it does not necessarily follow that the ob- 

 served decrease of resistance consequent on increase of pressure 

 occurred in the carbon of the button. The buttons are, I 

 believe, formed by compressing lampblack mixed ivith gum- 

 water. Must we not suppose, then, that the particles of 

 lampblack are bound to each other by the gum, and separated 

 from each other by the gum, to a greater or less extent ? 

 Would not the diminution of resistance experienced in the 

 body of the button when pressure was applied be due to one 

 or both of the two following causes : — 



1. Diminution of the resistance of the thin coating of gum 

 between the particles of carbon ? 



2. Better surface-contact between one particle of carbon 

 and another? 



The influence of time on the change of resistance might be 

 quite accounted for by supposing cause 1 to be at work*. 



LII. On Stationary Waves in Flowing Water. — Part II. 

 By Sir William Thomson, F.R.S. fyc. 



[Continued from p. 357.] 



Correction in Part IT. — In lines 4-7 of paragraph 3 on page 357, delete 

 the words ", because " ... to ... " canal "; and add the sentence " The 

 explanation of this will he more fully developed in Part III., to be 

 published in December." 



TO find, as promised in Part I., the sum of horizontal 

 pressures on an inequality of the bottom, or on a bar, or 

 on a series of inequalities or bars, consider the horizontal 

 components of momentum of different portions of the water 

 in the following manner. Because the motion is steady, the 

 momentum of the matter at any instant within any fixed 

 volume of space 8 remains constant ; and therefore the rate 

 of delivery of momentum from S by water flowing out on one 

 side above gain of momentum by water flowing into S on the 

 other side must be equal to the total amount of horizontal 

 force acting on the water which at any instant is within S ; 

 the direction of this force being that of the flow when the 

 momentum of the leaving water exceeds that of the entering 

 water. Now let S be the space bounded by the bottom, the 



* See the account of my own experiments on the viscous metals zinc 

 and tin, quoted above. 



