All dissolved crystalloids do not pass through gelatinous membranes with equal ease, and substances are known, mostly gelatinous precipitates, which do not permit the diffusion of dissolved salts, though they allow water to pass freely. Thin layers of such precipitates form what are called “semipermeable membranes.” The existence of such membranes affords us the possibility of direct measurement of the tendency to diffusion, or as it is generally called the “osmotic”[53] pressure of dissolved bodies. Thus a porous earthenware battery-cell may be immersed in a solution of copper sulphate, and filled with one of potassium ferrocyanide. In this way its pores will be filled with a gelatinous precipitate of copper ferrocyanide, which is pervious to water, but impervious to most dissolved substances. If now the cell be filled with a dilute solution of some crystalloid, say sugar, and its top closed by a perforated cork fitted with a vertical tube, and the cell be plunged in water, the latter will pass into the cell, and the dilute solution will rise in the tube to a height of many feet above the water outside. By substituting a mercury pressure gauge for the vertical tube, exact measures of the pressure in the cell can be made, which is the osmotic pressure of the dissolved substance. At first sight it is paradoxical that the water should flow into the solution, apparently against a heavy pressure, but the explanation is simple. Mention has already been made of the enormous internal pressures of liquids produced by the attractions of their molecules. In the solution a portion of this is borne by the dissolved substance, and the water flows in from the outside till an internal mechanical pressure is produced, equal in amount to the osmotic pressure of the dissolved substance. The resemblance of the phenomena of solution to those of vapour-pressure has already been mentioned, and it is found to be even quantitative, since the measured osmotic pressures are exactly equal in amount to those which the dissolved body would produce if it were in the state of vapour at the same temperature and occupying the same volume as the solution. It acts, in fact, precisely as the “partial pressure” of a vapour. There are several indirect ways of measuring the osmotic pressure of dissolved bodies, as for instance, from the lowering of the freezing point, or the raising of the boiling point of the solution as compared to those of the pure solvent, all of which confirm the direct measurements, and show that in a given volume at the same temperature, the same number of molecules will produce the same osmotic pressure whatever their nature, or conversely, that at the same osmotic pressure and temperature equal volumes of any solution must contain the same number of molecules. The use of these facts in determining molecular weight is obvious.