THE KINETIC THEORY REVIVED. The Dynamical Theory of Gases. By J. H. Jeans. Second edition. Pp. vii + 436. (Cambridge: At the University Press, 1916.) Price 16s. net. More than eleven years have passed since the first edition of this work was reviewed in NATURE (April 27, 1905). Most of the pioneers of the rigorous mathematical theory have passed away, and the attempt to reconcile Boltzmann's minimum theorem with the properties of an aggregate of perfectly reversible units may be said to have been abandoned. On the other hand, the recently developed quantum hypothesis has, to some extent, led us to believe that something more than the equations of reversible dynamics is needed to account for the phenomena of Nature. Equipartition may be characteristic of molecular systems, but the celestial universe shows no tendency towards Maxwell's law, and would probably refuse to obey it even if started according to this distribution. The plan which Prof. Jeans now adopts in his book is probably the best one in the circumstances. The kinetic theory cannot be proved mathematically, neither can the data determined from a calculable mathematical theory be made to serve as more than approximations to the results of experiments. Thus arises a school of slipshod students of physics, who, when they cannot prove a result mathematically, state that it "has been shown experimentally," and if they cannot get their experiments to verify they state that it "may be proved" (from theory). This danger is largely obviated by the division of the earlier chapters into four sections, entitled "Mathematical Theory of a Gas in a Steady State," "Physical Properties of a Gas in a Steady State," "Mathematical Theory of a Gas not in a Steady State," "Physical Phenomena of a Gas not in a Steady State." Among the miscellaneous applications it is interesting to note Prof. Jeans's remarks on the rate of escape of gases from planetary atmospheres. It will be remembered that the late Dr. Johnstone Stoney attempted to account for the loss of gases by the motion of the molecules which describe hyperbolic orbits under the attraction of the primary; and by assuming the absence of a particular gas from a particular member of the system he deduced the absence of other gases from other systems. It was, however, subsequently shown that, under the assumptions made by Dr. Stoney, the gases in question would not escape, and Dr. Stoney advanced the opinion that the methods of the kinetic theory on which his own investigations were based were inapplicable to the problem to which he had applied them. According to Prof. Jeans's views, hydrogen does not at present escape, but it did so when the earth was at a far higher temperature than at present. On the other hand, the brief discussion on our existing knowledgeTHE KINETIC THEORY REVIVED. The Dynamical Theory of Gases. By J. H. Jeans. Second edition. Pp. vii + 436. (Cambridge: At the University Press, 1916.) Price 16s. net. More than eleven years have passed since the first edition of this work was reviewed in NATURE (April 27, 1905). Most of the pioneers of the rigorous mathematical theory have passed away, and the attempt to reconcile Boltzmann's minimum theorem with the properties of an aggregate of perfectly reversible units may be said to have been abandoned. On the other hand, the recently developed quantum hypothesis has, to some extent, led us to believe that something more than the equations of reversible dynamics is needed to account for the phenomena of Nature. Equipartition may be characteristic of molecular systems, but the celestial universe shows no tendency towards Maxwell's law, and would probably refuse to obey it even if started according to this distribution. The plan which Prof. Jeans now adopts in his book is probably the best one in the circumstances. The kinetic theory cannot be proved mathematically, neither can the data determined from a calculable mathematical theory be made to serve as more than approximations to the results of experiments. Thus arises a school of slipshod students of physics, who, when they cannot prove a result mathematically, state that it "has been shown experimentally," and if they cannot get their experiments to verify they state that it "may be proved" (from theory). This danger is largely obviated by the division of the earlier chapters into four sections, entitled "Mathematical Theory of a Gas in a Steady State," "Physical Properties of a Gas in a Steady State," "Mathematical Theory of a Gas not in a Steady State," "Physical Phenomena of a Gas not in a Steady State." Among the miscellaneous applications it is interesting to note Prof. Jeans's remarks on the rate of escape of gases from planetary atmospheres. It will be remembered that the late Dr. Johnstone Stoney attempted to account for the loss of gases by the motion of the molecules which describe hyperbolic orbits under the attraction of the primary; and by assuming the absence of a particular gas from a particular member of the system he deduced the absence of other gases from other systems. It was, however, subsequently shown that, under the assumptions made by Dr. Stoney, the gases in question would not escape, and Dr. Stoney advanced the opinion that the methods of the kinetic theory on which his own investigations were based were inapplicable to the problem to which he had applied them. According to Prof. Jeans's views, hydrogen does not at present escape, but it did so when the earth was at a far higher temperature than at present. On the other hand, the brief discussion on our existing knowledge