260 lECTURE XKI. 



and are possessed also of cohesive powers, which essentially distinguish them 

 from elastic fluids, and which resist any force tending directly to separate the 

 particles from each other, while they admit any lateral motion with perfect 

 facility. In treating of hydrostatics, therefore, we suppose the fluids con- 

 cerned to be of uniform density throughout; and as far as elastic fluids agree 

 with this description, they are subject to the same laws with liquids; on the 

 other hand, all fluids, as far as they are compressible, possess properties simi- 

 lar to those which will hereafter be examined, when we investigate the subject 

 of pneumatic equilibrium. 



The first law of hydrostatics which arrests our attention, is this, that the sur- 

 face of every homogeneous gravitating fluid, when at rest, is horizontal. If any 

 part of the surface were inclined to the horizon, the superficial particles would 

 necessarily tend towards its lowest part, in the same manner as if they moved 

 without friction on the inclined surface of a solid. And if any two portions of 

 the surface of the fluid are separated, as in two branches of a tube or pipe, how- 

 ever they may be situated, the fluid cannot remain at rest, unless the surfaces be 

 in the same level plane: for if we imagine such a tube, containing water, to be 

 made of ice, and to be immersed in a large reservoir of water, and then 

 thawed, the water will make a part of the general contents of the reservoir, and 

 consequently -will remain at rest, if its surfaces are level with that of the re- 

 servoir: and it is obvious that the tube has acquired no new power of sup- 

 porting it from being tliawed: consequently, the water would have remained 

 in equilibrium at the same height in the original state of the solid tube. The 

 experimental proof of this proposition is easy and obvious, and the property 

 affords one of the most usual modes of determining a horizontal surface. But 

 when we compare the heights of fluids occupying tubes of different magni- 

 tudes, it is necessary, if the tubes are small, to apply a slight correction on 

 account of the actions of the tubes on the fluids which they contain, 

 which are more apparent, as their diameters are smaller. The same cause 

 produces also a curvature in each separate surface, which, is always visible 

 at the point of contact with the tube or vesseL (Plate XIX. Fig Q,39-) 



If several separate fluids of different kinds be contained in the same vessel^ 

 they 'vill never remain at rest unless all the surfaces intervening between 



