This book exhibits the usual merits and demerits of a symposium report; merits, because hardly by any other means would one get so many aspects of the science of ice treated between the covers of one book: and demerits, because of lack of rigour in refereeing and editing. With regard to the first of these charges, it in fact comes off very well: one can agree with Onsager’s closing remark “there is very little I wanted to miss”. On the other hand, the second usual defect is somewhat enhanced on this occasion by the fact that the language of the symposium, English, is not that of the editors, or of a majority of the contributors. Language apart, the number of occasions when the equation as printed is obviously not quite right is irritatingly many.
Printed in photo-reduced typescript (which, in the reviewer’s experience, always makes for a lower standard of proof-reading) on rather thick paper, this is too cumbersome a volume to carry around for browsing. The diagrams are often so much reduced as to need a magnifier, but admittedly can always then be made out.
Of the 57 papers, the largest block, 12, deal with electrical properties: and since the next largest group, 9 dealing with diffusion and relaxation phenomena, include several on dielectric relaxation, electrical properties is, quite rightly, the dominant theme. Gränicher’s introductory review to the whole symposium makes an excellent and much-needed guide through the conflicting evidence of this body of work, which still gives one the same impression as it did ten years ago that complete understanding is just around the corner. In the interim, estimates of the basic parameters have altered, some a little, some a lot, with the acquisition of purer specimens, discovery of relatively very large surface conductivities above –30° C, and recognition of the important difference between various kinds of electrodes. (Mounier and Sixou are perhaps the only workers who achieve an electrode with really well-defined simple properties—the completely blocking electrode realized by inserting sheets of mica between the electrodes and the ice.) In addition, new data on other phenomena such as self-diffusion and proton spin resonance have become available, and the complete scheme of defect behaviour which precisely fits all the facts still evades discovery.
Seven papers deal with mechanical properties, in which, in particular, the Japanese workers add significantly to our detailed knowledge of the creep laws, and dislocation motions and multiplication processes in ice. Higashi is less than enthusiastic for Glen’s theory which connects dislocation motion with the migration of Bjerrum defects. Higashi attaches importance to the equality in activation energies of creep and self-diffusion, and deduces that dislocation motion is thermally controlled by the diffusion of molecular vacancies to dislocation jogs. Unfortunately for the strength of this argument, the activation energies of self-diffusion and dielectric relaxation are, within experimental error, the same, though, these processes must involve the motion of different defects—nature has played an unkind trick on us here. Whether his theory gives the whole story or not, Glen has clearly shown that dislocation motion and molecular reorientation are connected with each other: and an implication is that the man who is only concerned with the creep rate of ice cannot say that discrepancies in the proton spin-lattice relaxation time are no concern of his. All aspects of the science of ice are interrelated.
There are groups of papers on structure (H2O has more crystalline phases, 9, than any other substance we know: even more if an alleged ferroelectric transition at about 100 K is genuine) and on lattice dynamics and the infra-red spectrum. Whalley contributes important papers on both these topics.
Of many less connected points which emerge in this wide-ranging symposium, space permits mention of only a few. Mogenson finds no change in positron annihilation lifetime when ice melts, but a discontinuous change in water, between 2°C and 4°C. Kvajić, Brajović and Pounder find more impurity incorporation (from 5 ppm radioactive Cs2CO3 in distilled water) when the ice grows upwards than downwards—whereas a number of experimenters are careful to grow their crystals upwards, reasonably expecting that to produce a purer product. Dantl and others find ageing effects in ice: the density of freshly formed ice can be as much as 0.3% higher than after ageing.
Hobbs and Ketcham claim to produce the first direct experimental determinations of the surface free energies solid–liquid, solid–vapour, and grain-boundary for ice: however, their analysis is invalid—they cannot consistently make the approximation of isotropy and the assumption that a planar solid surface can be in equilibrium under a sessile drop.
Glen, in a general review of (field) glaciology grossly misuses the word “dilatant”. It is too useful a word to throw away in the original, and etymologically sensible, meaning for which Osborne Reynolds coined it: but to use it in the reverse sense of a corrupted meaning is really too bad! This paper apart, there is nothing directly about field glaciology in the symposium. Papers bearing on meteorological phenomena continue to give the result that whereas there are observable processes producing charge separations which may account for thunderstorms, theoretical mechanisms which can explain the observed magnitudes of these charge separations are still lacking.
The report can be recommended as a massive source of both experimental and theoretical information, but a careful examination for conflicting results is needed.
