Sodium chloride seems to diminish the swelling caused by all acids, but in presence of large excess of sodium chloride, most of the acid in combination with the gelatine will probably be hydrochloric, whatever the acid used to originally produce the swelling. A curious fact observed by the author, is that absolute alcohol, which so effectually dehydrates neutral gelatine, is almost powerless to remove either water or acid from gelatine swollen by hydrochloric acid. HCl is freely soluble even in absolute alcohol, but H- and Cl-ions can only exist in it to a very small extent, so that we may conclude that the acid which causes the swelling and retains the water of the jelly exists either in actual combination with the gelatine, or in an ionised condition. [57] The acid retained by the gelatine, as measured by deducting from the total contained in the jelly, a quantity equivalent to the volume of solution absorbed, at first rises rapidly to a maximum, then slightly diminishes and remains practically constant. On the theory suggested, it is evident, however, that the absorbed solution must be more dilute than that outside, and the actual combined acid greater than that shown by the above calculation. The “combined” acid, as determined by indicators, shows slight but continuous increase. It is acid to phenolphthalein, but neutral to methyl orange. Solutions of caustic alkalies are in most respects analogous in their swelling action to those of strong acids. A portion of the alkali is in some way fixed by the gelatine, while another portion is simply absorbed as solution. A maximum swelling effect is also noticed with dilute solutions, which is diminished as the concentration increases. Swelling by alkalies is not diminished by chlorides so far as has been observed, and especially it may be noted that the swelling produced by caustic soda is not diminished by sodium chloride. On the theory which has been suggested there is no reason why alkaline swelling should be reduced by chlorides, since the swelling agent has no Cl-ion, but it is somewhat singular that the sodium salt, having a common Na-ion should produce no repression of the swelling by caustic soda. In the present state of our knowledge no definite explanation can be given, but it is quite possible that the swelling in this case is not produced by the sodium-ion but by some more complex one, or even by the hydroxyl-ion, like most of the characteristic reactions of alkalies. Apparently the gelatin-alkali compound is still strongly alkaline, affecting phenolphthalein indicator like uncombined alkali--an effect which is known to be due to the presence of free HO-ions.