522 RICE ART. L 



term "surface tension" for the quantity denoted by a, with 

 unfortunate results for the real understanding of certain 

 phenomena by students reading elementary accounts of capil- 

 lary rise, for example. In consequence vague notions are preva- 

 lent that in some way a tight skin of water holds up the elevated 

 column in the capillary tube and "pins it" to the inner wall, 

 or, on the other hand, that a tight skin of mercury holds the 

 mercury in a capillary tube down below the general level in the 

 vessel outside. In the case of a spherical membrane under ten- 

 sion enclosing one body of gas and surrounded by another, both 

 pressures are available for observation, the inside as well as the 

 outside. In the present instance the intrinsic pressure of the 

 liquid is not open to observation, nor its cohesion; but we can 

 infer from the result (9) that the internal pressure just within a 

 spherical mass of liquid, subject to a definite external pressure, 

 is greater than it would be under a plane surface, subject to the 

 same external pressure, by the amount 2(t/R. In short the 

 liquid in the sphere is a little more compressed than that under 

 the plane surface, but tliis extra compression is not due to a 

 "surface membrane" in tension, but to a small change in the 

 inward attraction on the membrane due to the curvature. 

 Indeed the elevations and depressions observed in capillary 

 tubes are easily seen to arise indirectly from this cause. In the 

 first instance, the curvature at the surface of water in a capillary 

 tube dipping into a beaker of this liquid is caused by the strong 

 molecular attraction of glass on water as compared to the 

 attraction between the molecules of water (water "wets" glass 

 and adheres powerfully to it). This concave curvature can 

 only exist if the internal pressure just at the surface is less than 

 the external pressure; this external pressure is practically the 

 same as exists on the plane surface of the water in the beaker. 

 Thus the internal pressure just under the curved surface in the 

 tube is less than that under the plane surface in the beaker, and 

 this cannot be so unless the level in the tube is higher than in 

 the beaker; in short the column in the capillary tube is pushed up, 

 not pulled up. For a liquid like mercury which adheres 

 scarcely at all to glass, the absence of molecular attraction by 

 the glass necessitates a convex curvature in the capillary tube, 



