244 STBTICTTJBE OF GLAKDS. 



tube, and this be plunged into a larger vessel, c, d, containing 

 water, the tube being immersed till the solution, a, b, is at the 

 same level, e, e, as the water in the outer vessel, c, d. After a 

 little time it will be found that the fluid in , a has risen, 

 and got above the level, e, e, to b, for example, and that it is 

 continuing to rise, and will go on rising until the two fluids, 

 on the opposite sides of the bladder, are of the same density, 

 so that, if the tube, a, a, be not of sufficient length, the 

 fluid may even run over, having filled it completely. If 

 the tube, a, a, instead of containing a saline solution, contain 

 water, and the recipient, c, d, instead of water, contain a 

 saline solution, things being disposed as before, the fluid in 

 a, a, far from rising, will begin to fall, and instead of fall- 

 ing in c, d, it will begin to rise. When the tube and 

 the recipient contain solutions of different salts respec- 

 tively, but as nearly as may be of the same density, the level 

 of the fluid in neither will be altered perceptibly ; but, after 

 a certain time, the two salts will be discovered mingled to- 

 gether in both the tube and the recipient, or in the fluid on 

 both sides of the bladder. If the densities of the two saline 

 solutions have been different, the surface of that which is the 

 more dense will rise, that which is less dense will fall ; but it 

 will be found, nevertheless, that from the solution of greatest 

 density a portion will have passed into that of least density ; 

 the penetration has not therefore been all one way, but reci- 

 procally from each to the other, only in greatest measure from 

 the less to the more dense fluid. This phenomenon does not 

 take place only when moist animal membranes are the inter- 

 media between the two heterogeneous but miscible fluids ; it 

 also occurs when the interposed body is of an inorganic nature, 

 but thin and porous, and possessed of strength enough to sup- 

 port the increasing column of the denser fluid, such as thin 

 slices of slate, earthenware, &c. In general it may be said that 

 the power producing the phenomenon in question belongs to all 

 bodies which can absorb and retain a fluid in extremely delicate 

 pores."* The blood-vessels, especially the capillary vessels, 

 share this property of permeability to liquids ; hence, while 

 the circulation goes on, portions of the circulating fluid, espe- 

 cially its watery parts, escape through the walls of the vessels, 

 and pass off at the surface. This superficial loss, termed exha- 

 * Chimie, 4te. Aufl. B. ix. S. 161. 



