May 2, 1907] 



NA TURE 



19 



that these sounds are closely connected with ordinary 

 earthquake-sounds, and Prof. Hobbs finds that the 

 Calabrian villages from which brontidi are reported are 

 also those which have suffered most from disastrous earth- 

 quakes, and that they are ranged along the more prominent 

 seismotectonic lines of the district. 



In great detail Prof. Hobbs studies, not only the places 

 damaged by the important earthquakes of 163.8, 16513, 

 1783, 1S94, and 1905, but also those at which numerous 

 slight shocks were observed, for the latter, owing to their 

 small disturbed areas, seem to be the most useful indices 

 of the course of seismotectonic lines. The positions of 

 more than 300 such lines in Calabria and north-eastern 

 Sicily are estimated and drawn upon a series of maps, as 

 well as the bearings of joint-planes, the trend of the 

 volcanotectonic lines, and the distribution of brontidi. 



It will be seen from this brief abstract that Prof. 

 Hobbs's memoirs possess considerable interest. They are 

 the result of extensive reading, and contain many useful 

 references. But his wide generalisations seem to me to 

 be based on insecure principles and insufficient data. Iso- 

 seismal lines, it is well known, are elongated in the direc- 

 tion of the originating faults, but the positions of a few- 

 places at which shocks are felt cannot determine a line 

 of fracture. For instance, one of the British seismo- 

 tectonic lines is located by the positions of four places, 

 two of which are more than 200 miles apart. The seismo- 

 tectonic lines revealed by the New England earthquake of 

 1870 are based on the positions of about a score of places 

 distributed over an area reaching from Quebec to Xew- 

 haven, and on about a dozen apparent directions of the 

 shock. When observed in houses, such directions are 

 almost invariably perpendicular to the principal walls, but 

 Prof. Hobbs assumes that they indicate that the shocks 

 were transmitted along parallel seismotectonic lines. In 

 Calabria, on the other hand, the damaged villages are so 

 numerous that it would be strange if many of them were 

 not collinear. Several of the seismotectonic lines plotted 

 by Prof. Hobbs no doubt correspond with lines of fracture, 

 but the existence of a very large number of his lines must, 

 I think, bo regarded as doubtful. Industrious as he has 

 been in the collection of materials, he has tried within 

 little more than a year to achieve results which the long- 

 ■continued labours of many men might fail to establish. 



C. Davison. 



HYDRATES IN AQUEOUS SOLUTION. 

 A RECORD of researches which have been carried out 

 ■^ by Prof. H. C. Jones with his students and confreres 

 has recently been published by the Carnegie Institution.' 

 The investigations which have been undertaken were to 

 elucidate an observation made by Jones and Ota when 

 studying the freezing points of solutions of double salts in 

 order to ascertain whether in solution they remained as 

 constituent molecules or were broken down. They found 

 that concentrated solutions gave abnormally low freezing 

 ■points, the* molecular lowering of freezing point passing 

 through a well-defined minimum as the concentration 

 changed. Now according to the ionic theory as then ex- 

 pressed, the molecular lowering should decrease continu- 

 ously as the concentration of the solution increased. 



A very large number of solutions of salts, acids, and 

 bases, and neutral organic substances have now been 

 studied, and as a result it has been found that this 

 ■excessive depression as the concentration increases is a 

 general property of solutions. In order to explain this 

 digression from the generally accepted rendering of the 

 ionic theory, Jones postulates that " in solution a part of 

 the solvent is cojnbined with the dissolved substance and 

 no longer plays the role of solvent, at least so far as the 

 freezing point method is concerned." 



By a determination of the freezing point, conductivity, 

 and specific gravity of the solutions, it has been found 

 ■possible to calculate approximately the total amount of 

 water held in combination by the dissolved substance, and 

 consequently the approximate amount combined with one 

 molecule of the compound or of the ions resulting from it. 



Pp. viii-H264. 



1 " Hydrates in Aqueous Soluli 

 '(Washington : Carnegie Institutft 



" By Harry C. Jo 



1007.) 



The theory proposed here differs from that suggested bv 

 Mendeli5eff, who considered that such substances as 

 sulphuric acid and calcium chloride form a few definite 

 compounds with the water in which they are dissolved. 

 But the present theory supposes that combination between 

 the dissolved substance and water to be a general pheno- 

 menon. The compound forms, say, for example, calcium 

 chloride, a complete series of hydrates extending froin a 

 few molecules of water to at least thirty, all the inter- 

 mediate stages being represented. 



The memoir commences with an introduction, in which 

 the earlier work is reviewed and the freezing-point and 

 conductivity apparatus used by the author are described. 

 Then follows part i., dealing with the evidence for the 

 existence of hydrates in aqueous solution and the approxi- 

 mate composition of the hydrates formed by a large 

 nuiiiber of electrolytes. The work here described was 

 carried out by Getman and Bassett. Attention is directed 

 to the effect of temperature on water of crystallisation, as 

 bearing on the theory of hydrates in solution. It is shown 

 that salts which on crystallisation contain water of 

 crystallisation are able to combine when in solution at 

 ordinary temperatures with a much larger quantity of 

 water than they are able to bring with them out of solu- 

 tion on crystallisation. The results obtained are illustrated 

 in many cases graphically by curves and in other cases by 

 tables. 



Part ii. is chiefly the work of Uhler, and deals with 

 spectroscopic investigations. The spectrographic photo- 

 graphs which are given have been magnificently repro- 

 duced, and form quite a feature of the book. The 

 colour changes produced, for example, by the addition of 

 different salts to cobalt salts have been investigated 

 quantitatively. That is to say, the absorption spectra of 

 the substances, separately and when mixed in known 

 quantities, have been observed with a direct-reading 

 spectroscope, and thus the wave-lengths and absorption 

 bands obtained. The special spectrograph which has been 

 used to obtain the photographic record of the absorption 

 bands is also described. The final section deals with non- 

 aqueous solutions, the solution of substances in methyl 

 and ethyl alcohol having been studied. The results seem 

 to indicate that some substances at least, such as lithium 

 chloride, bromide, and nitrate, combine to some ex- 

 tent with the solvent. However, this portion of the 

 work is yet in its initial stage, and much yet remains to 

 be done. We understand that the author is extending the 

 work in this direction. 



.Altogether, the memoir is an extremely valuable con- 

 tribution to the study of the subject, more especially in 

 coimection with concentrated solutions. It has often been 

 urged, and with a considerable amount of truth, that the 

 ionic theory is simply a specialised hypothesis, which is 

 true only of dilute solutions. Prof. Jones has gone far 

 to remove this reproach by broadening the basis of the 

 theory and enlarging its scope. The publishers, the 

 Carnegie Institution, must also be congratulated upon the 

 splendid way in which the letterpress and diagrams have 

 been got up. F. M. P. 



PRODUCTION AND DECAY OF MEDIEVAL 



STAINED GLASS.'^ 

 T^'HE earliest direct evidence as to the methods of 

 rnediaeval glass-painting is contained in the treatise of 

 Theophilus (" Diversarum .Artium Schedula "), which dates 

 back in all probability to the latter half of the twelfth 

 century ; here one finds detailed instructions for the making 

 of the glass as well as for its formation into the flat sheets 

 or " tables " in which it is required by the glass-painter. 



This treatise makes it clear that at that time such 

 window glass was for the most part made by what is 

 generally known as the " muff " process. The process re- 

 ferred to is one of the three known methods of making 

 window glass, namely : — 



(i) Cast or plate glass, made by pouring molten glass 

 on to a flat stone or metal slab. 



(2) Muff or cylinder glass, in which the glass is worked 



NO. 1957, vor. 76] 



