July 25, 1907] 



NATURE 



291 



be taken without sensible error as constant through- 

 out small distances near P. The same method ap- 

 plies mutatis mutandis to temperature. 



The chapter on the diffusion of gases from the 

 point of view of thermodynamics requires more ex- 

 planation than the author has directly given. He 

 says (p. 125) :— 



" When two gases at equal temperature and pres- 

 sure mix by diffusion, the gain of entropy is the same 

 as would occur if each were to expand by escaping 

 into vacuum till it occupied the volume of the mix- 

 ture." 



To this Advocatus Diaboli would say, If instead of 

 two gases you have two quantities of the same gas, 

 oxvgen, caeteris paribus, the whole system remains 

 throughout in the same physical state, and, therefore 

 (art. 86 [2]), there is no gain of entropy. What 

 difference can it make that one volume of oxygen is 

 replaced by nitrogen ? 



I think Prof. Bryan would justify his statements 

 thus : — He asserts, art. 124 (a), " as two gases at 

 equal pressure and temperature in general tend to mix 

 by diffusion and not to separate, the process of diffu- 

 sion is irreversible." And he implies (b) that every 

 irreversible process necessarily involves increase of 

 entropv. If these principles (a) and (6) be granted, 

 125 is probablv justified. But they are both very ques- 

 tionable. 



It is not possible witliin the limits of this notice 

 adequately to discuss either (a) or (b). I would, how- 

 ever, point out that in diffusion, as in all motions of 

 gases, if at any instant the velocities of all the mole- 

 cules were reversed the system, if isolated, would 

 retrace its course. Does not this fact make a broad 

 distinction between diffusion of gases and irreversible 

 processes usually admitted as such? 



S. H. Bl'rbl'ry. 



VOLCANOES. 



1 Vutcnui Attivi della Terra. Morfologia — Diiiamismo 

 — Prodotti Distribiizione Geografica-Cause. Bv Ci. 

 Mercalli. Pp. viii + 421 ; illustrated. (Milano : Ulrico 

 Hoepli, 1907.) Price 10 Lire. 



THLS history of the study of volcanoes may be 

 divided into three periods ; the earliest is covered 

 by the fragmentary remains of the writings of classical 

 philosophers and the sporadic records of great erup- 

 tions of Vesuvius and Etna during the Middle Ages ; 

 the second commenced with the eruption of ^'esuvius 

 in 163 1, which gave rise to over 200 publications, and 

 from this date on we have a fairly complete record of 

 the activity of Vesuvius and Etna ; in the third 

 period, observation became systematised, and vul- 

 canology, as a science, may be said to date from 

 Spallanzani's study of Stromboli in 1788. In the nine- 

 teenth century the science expanded its boundaries, 

 volcanoes in other parts of the world besides Italy 

 began to be studied, experimental methods were ap- 

 plied to elucidating the mechanism of eruptions and 

 the formation of volcanic rocks, and the microscope 

 to the investigation of their composition and structure. 

 As a consequence of this expansion of the science 

 it has come to pass that we have had to look, not 

 NO. 1969, VOL. 76 I 



to Italy, but to other countries, and especially to Eng- 

 land, for a general handbook ; Prof. Mercalli has recti- 

 fied this, and the country where the study of volcanoes,, 

 and the science of vulcanology, took their birth has pro- 

 duced the best and most complete guide to their pur- 

 suit. In the compass of a moderate sized book, we 

 have a remarkably complete, well-balanced review of 

 the subject, which commences with the final result of 

 volcanic activitv, in an account of the rocks produced, 

 and works back through the forms of volcanoes, their 

 dynamics, and distribution, to the cause of volcanic 

 activity. 



The longest and most generally interesting chapter 

 in the book is doubtless that dealing with the dynamics 

 of volcanoes. Fissure eruptions and the outflow of 

 lava without the formation of a volcanic cone are 

 recognised, and in the classification of volcanic explo- 

 sions we come across a third type — in addition to the 

 familiar vulcanian and strombolian types — in what are 

 termed plinian eruptions. This name is applied to 

 the violent explosive eruptions, like that of Vesuvius in 

 79 A.D., of Bandaisan and of Krakatoa, which follow 

 prolonged periods of repose, are of extraordinary 

 violence, are accompanied by comparatively little or 

 no outpouring of lava, while causing the ejection 

 of large volumes of previously solidified material, 

 and are succeeded by another period of repose. 

 The eruptions of Pel^e and St. Vincent in 1902 are 

 regarded as differing in degree only, not in kind, from 

 other known eruptions ; the celebrated spine of Pel^e, 

 which was thrust up to 1000 feet above the crater, was 

 an extreme case of extrusion of solidified lava, and 

 the " black cloud " an extreme case of the avalanches 

 of incandescent ashes which are a not uncommon 

 accompaniment of great eruptions. 



In dealing with the cause of volcanic activity, Prof. 

 Mercalli favours the view, first propounded by Seneca, 

 that it is produced by the access of sea water to highly- 

 heated material in the interior of the earth, resulting 

 in the production of high-pressure steam ; but here, as 

 elsewhere throughout the book, the theory is not 

 pressed, and alternative explanations are fairly stated. 

 A word, too, may be said for the illustrations, which 

 are numerous and excellent. 



OUR BOOK SHELF. 



Shaft Sinking in Difficult Cases. By J. Riemer ; 



translated from the German by J. W. Brough. Pp. 



xii4-i22; with 18 illustrations and 19 folding plates. 



(London : Charles Griffin and Co., Ltd., 1907.) 



Price 10s. 6d. net. 

 Mr. Riemer is one of the leading German authorities 

 on sinking, and a translation of his valuable treatise 

 forms an addition to English technical literature that 

 is specially welcome in view of the fact that shaft 

 sinking, the most complicated of all mining problems, 

 is necessarily dealt with in a brief manner in the stan- 

 dard works on coal-mining. The volume is confined 

 to a description of means that have to be resorted to 

 when ordinary methods of sinking cannot be applied 

 on account of excessive influ.x of water, the means 

 described being shaft sinking by hand, boring shafts^ 

 the freezing method of sinking, and the sinking-druir. 

 method. 



The particulars given relate exclusivelv to recent 



