Infrared measurements have been made of the influence of pressures between 0 and 40 kbar on the stretching and bending frequencies of N-H groups in a range of crystalline compounds which have N-H. . .N, N-H. . .O, N-H. . .S or N-H. . .Cl hydrogen bonds. The frequencies have been found to change in accordance with rules that were derived and tested in Parts I and I1 for other kinds of hydrogen-bond systems.

The electrical conductivity of water, compressed by explosively driven shock waves, has been measured in the experimentally difficult range of relatively low pressures (20-100 kbars). The results link the earlier high-pressure data of David and Hamann and of Yuknavech with the results of static experiments at much lower pressures. They confirm that the high conductivity of shocked water arises from enhancement of the normal self-ionization of water. In addition, experiments have been carried out to determine the efTects of changing the initial temperature of the water, and to compare the behaviour of shocked D20 with that of H20.

A simple formula is proposed to describe the pressure dependence of the variable Φ in El?yanov and Gonikberg's linear free energy relationship for ionization reactions in solution at high pressure. ��� The expression, given in equations (10) and (12), provides a good description of the influence of pressure on ionization equilibria in aqueous solutions. It permits El'yanov's general linear relationships between Φ and ionization free energies, enthalpies and entropies, pH and Hammett's p parameter, to be expressed in terms of the pressure in convenient analytical forms. ��� The formula is shown to be consistent with the simple electrostatic theory of ion hydration, allowing for the effect of pressure on the dielectric constant of water. Combined with the theory, it provides a general means of predicting ionization constants over a wide range of pressures and temperatures simply from knowledge of the changes in molar volume, enthalpy and entropy which accompany the reactions at atmospheric pressure.

Measurements have been made of the influence of pressure on the stretching and bending frequencies of hydrogen bonds in a further series of 15 solid compounds. Two exceptions have been found to the general rule that an increase of pressure decreases the X-H stretching frequencies of weak and medium-strength X-H. 3 .Y bonds. Possible explanations are offered of the exceptions.

Rate constants between 1 and 3000 atm have been measured for the unimolecular solvolyses of tert.-butyl chloride and benzotrichloride. They show that the effects of pressure on the velocities of liquid phase ionic reactions can be explained in terms of the solvation forces involved. Stearn and Eyring's theory of the influence of pressure is shown to be inadequate.The basic ionization constants of ammonia and methylamine have been measured at 1, 10o0, 2000 and 3000 atm. They confirm the suggestion that an increase in solvention interaction is responsible for the change in basic strength with pressure. The effect of pressure on ionic solvation free energies can be worked out quantitatively from Born's formula.The most extensive and accurate observations of the influence of pressure on liquid-phase chemical reactions were made by Fawcett, Gibson, Perrin and Williams 1-5 a t the Research Department of I.C.I. (Alkali) Ltd., Northwich. These authors measured the rate constants k of a number of simple organic reactions at pressures up to 12,000 atm, and Perrin summarized their work in a paper read at the Favaday Society Discussion on Reaction Kinetics in 1937.5In his review Perrin grouped the reactions into three kinetic classes : (i) " slow " reactions, with small collision factors, which were greatly accelerated by pressure ; (ii) " normal " reactions, with normal collision factors, which were slightly accelerated by pressure ; (iii) a unimolecular decomposition, which was retarded by pressure.