2 PRINCIPLES OF STRUCTURE 4I 



As Fig. 38c shows, the cohesive forces acting on a molecule at the 

 surface do not cancel each other. The particles are therefore attracted 

 by the bulk of the liquid. It will yield to this attraction as far as possible 

 and to some extent decrease its distance from the deeper-lying mole- 

 cules. This results in an increase in density, of which a rough outline 

 is given in Fig. 38d. In this way a surface "skin" is formed, which on 

 its inner side merges into the area of the homogeneous liquid. 



The surface skin possesses a certain firmness because its molecules 





^-y' • • • 



• • • 



a 



^J b) cj d) 



Fig. 38. Inhomogeneity of the phase boundary liquid/gas. Cohesive forces a) symmetrical, 



b) asymmetrical, c) directed inwards; ii) scheme of the inhomogeneous arrangement of 



molecules (greatly exaggerated, as the compressibility of liquids is very small). 



cannot move as freely as in the ideal liquid. This firmness can be de- 

 termined by stretching a lamella of the liquid suspended in a frame 

 by means of a movable bar, and by measuring the weight needed to 

 break the film. This weight is independent of the thickness of the 

 lamella, but is a Hnear function of the length 1 of the bar, since a 

 lamella which is twice as broad can carry twice the weight. The firm- 

 ness of the surface, therefore, refers to the unit of length, i cm, and 

 the force which is capable of rending a lamella surface i cm wide is 

 called the surface tension a of the liquid. As both the surface in front 

 and that at the back ol the lamella must be broken, the force p = 2 ct 1 



(Fig- 39)- 



Instead of the more accurate methods of surface tension measure- 

 ments with the aid of capillary rise or stalagmometry (Hober, 1922, 

 p. 154), the much more primitive breaking method has been mentioned 

 here, because the definition of surface tension is founded on it and it 

 demonstrates in a simple way its dimension as force/cm. Surface 

 tension, therefore, is not tension in the ordinary sense, for otherwise 

 its dimension would have been force/cm^. The difference between these 

 two quantities can be seen from the scheme given in Fig. 40. In order 



