78 



NATURE 



[May 28, 1908 



still occasionally found who, in defiance of all, the 

 rigidly scientific investigations of Brugmann, Osthoff, 

 Henrv, Sweet, Murray, and other philologists, per- 

 sist, bv disregarding phonetic ;.nd other ascertained 

 linguistic principles, in connecting together utterly 

 dissimilar tongues, such as the Indo-European lan- 

 guages, Hebrew, and Basque. The author of the 

 above-named work is a writer of this type. His work 

 bristles with philological impossibilities, and he ap- 

 pears to have no conception of the necessity of ascer- 

 taining, before comparison of one language with 

 another, the laws which govern the sound changes of 

 the languages compared and of the immediate groups 

 to which they belong. The Hebrew word Satan 

 he thinks is cognate with the Basque Tusuria " by 

 transposition," and the work abounds in similar equa- 

 tions. The volume is unworthy of serious attention, 

 and its only interest arises from its being one of 

 those strange works that spring from the union of a 

 certain kind of learned industry with misdirected in- 

 genuity. 



LETTERS TO THE EDITOR. 



[The Editor does not liold himself responsible for opinions 

 expressed by his correspondents. Neither can he ttndertalze 

 to return, or to correspond with the writers of, rejected 

 tnanuscripts intended for this or any other part of Nature. 

 No notice is taken of anonymous communications.] 



Fellowship of the Royal Society. 



It is well known that under existing regulations the 

 number of new fellows elected to the Royal Society every 

 year is only fifteen. In this way the total number of 

 fellows is kept at about 450. In the early days when this 

 arrangement was made the limited annual number was 

 doubtless sufficient to ensure the election of all the scientific 

 men who really merited the honour, but since those days 

 the scientific world has been growing larger and larger, 

 and at the same time the general standard of work in all 

 branches has become higher. 



So long as the annual nuinber of candidates was not 

 more than forty or forty-five the selection of fifteen was 

 not very difficult, and no man who had really done good 

 work had to wait more than two or three years before 

 election. Now, however, the annua! number of candidates 

 has increased to eighty or ninety, and this year it is said 

 there were nearly 100 candidates. 



Is it not high time, then, that the Royal Society took 

 definite steps to make some change which would meet the 

 requirements of the changed circumstances? Many of the 

 older members of the society are well aware that the pre- 

 sent state of affairs is unsatisfactory, and some have ex- 

 pressed their sentiments, but nothing has yet been done. 



A simple plan would, of course, be to elect thirty new 

 fellows every year instead of fifteen, but one can see 

 objections to this plan. Has it ever been suggested that 

 the Society should create an associateship and elect 

 associates as well as fellows? The number of fellows 

 might remain as it is, but if a limited body of associates 

 was created, say fifty to begin with, and was increased 

 by the election of twelve or fifteen every year, the pressure 

 would be relieved, and 1 should think A.R.S. would be 

 preferable to a long-deferred F.R.S. Subsequent elections 

 of fellows could then be made from the associates, and 

 .this double election would give better assurance than now 

 exists that none but the best men of the year were admitted 

 to the fellowship. Enquirer. 



Earthquakes and Earthshakes. 



Some of the memoirs, professedly seismological, which 

 have appeared during the last year or two indicate that 

 confusion has arisen from the use of the word earthquake 

 in two distinct and independent senses. .\s this confusion 

 seems likely to increase unless a modification of our 

 nomenclature is adopted, the introduction of a new term 

 appears to be requisite, however much this may be de- 

 precated on other grounds. 



NO. 2013, VOL. 78] 



In the generality of cases, the phenomenon represented 

 by the word earthquake consists of a vibratory motion of 

 the ground, of the nature of a wave motion, propagated 

 outwards from a more or less extensive origin or focus. 

 In some cases this disturbance may lead to damage or 

 destruction of buildings, or even to displacement of the 

 surface layers of the earth ; but these are secondary results 

 of the molecular displacements involved in the propagation 

 of the wave motion, and, apart from them, the earth, 

 after the earthquake has passed, resumes the same position 

 and condition as before. 



Occasionally, however, the word is applied to a dis- 

 turbance of a wholly different kind, resulting in the forma- 

 tion of fractures and displacements of the solid rock, 

 displacements which are molar and permanent, in the sense 

 that the masses affected by them do not return to their 

 original position after the earthquake has passed. 



As the first was the sense in which the word is in- 

 variably used in Robert Mallet's classical researches, as it 

 is that which has been sanctioned by long-continued usage, 

 and as the proportion of records and observations, which 

 do not apply to this phenomenon, is probably less than 

 one in a thousand, I suggest that the word earthquake 

 should continue to be used in this sense, and that for the 

 other sense, in which it is sometimes used, the word 

 earthshake should be substituted. Using the words in 

 this way, we may say that earthquakes, or at any rate 

 severe earthquakes, are frequently, if not invariably, 

 caused by rupture of the earth's crust and the formation 

 of fractures or faults in the solid rock, but these fractures, 

 which are the primary cause of the earthquake, are only 

 the secondary result of the earthshake, the action of which 

 arises at a greater depth, and the ultimate cause of which 

 lies beyond our present ken. The distinction is an 

 important one, and the importance may be greater than 

 will be acknowledged immediately, for some recent studies 

 made by me have indicated a possibility that the earth- 

 shake has sometimes a greater extent than the earth- 

 quake ; in other words, that the area over which permanent 

 displacements of the earth's surface have taken place may 

 be greater than the seismic area, or the area over which 

 the shock was felt. 



Incidentally, it may be mentioned that the w-hole of 

 Prof. See's recent publications on the cause of earthquakes, 

 and the greater part of those by Prof. Hobbs, deal with 

 earthshakes and not \vith earthquakes as here defined. 

 This is natural, for only the permanent changes, resulting 

 from the earthshake, are of importance to the cosmogonist 

 or the geologist ; the transient displacements produced by 

 the earthquake concern them, directly, but little, if at all. 



R. D. Oldham. 



Classification and Mathematics. 



If mathematics is to be regarded as the science of 

 classification, a view apparently taken in many recent 

 works, it may be worth w'hile to consider whether mathe- 

 matical teaching should not begin with the use of models 

 of classifications in genera! rather than with the special 

 classifications in connection with which terms like straight 

 line, rotation, product, power, &c., were originally intro- 

 duced. 



By a model of a classification is meant, for example, a 

 set of things which can be classified" by one respect as 

 colour, and cross-classified by another as shape. Similarly, 

 models can be made having three or four or more differ- 

 entiations, in which any two differentiations have the rela- 1 

 tion of classification and cross-classification. If each I 

 differentiation is supposed to be ordered, we have then spaces ' 

 of two, three, or four dimensions, of which the classified 

 things form the points. By motion of a point in the 

 space is meant its change in those properties which have 

 been used in the classification. Consideration of the mean- 

 ing of extension, rotation, and right angle shows the 

 possibility of using the motion of extended bodies to con- 

 struct a classification of the points of a space, even when 

 we are unable to recognise the differentiations themselves 

 of the space. This is the case met with in ordinary 

 geometry. 



.^s the foundation of geometry lies in the idea of ordered 

 classification, so that of algebra lies in the conception of 



