April 2, 1900.] 



KNOWLEDGE. 



86 



the sound is heard with every earthquake, and by a 

 large proportion of the observ-ers who feel the shock ; 

 and here the sound-area is always large. In weak eartli- 

 quakes, the noise is heard further than the shock is 

 felt; in strong ones, it has been heard as far as 180 

 miles from the epicentre. In Japan, on the other hand, 

 the sound is inaudible at a distance of a few miles from 

 the epicentre. Of the earthquakes which originate be- 

 neath the land, about one-quarter arc accompanied by 

 sound; while this is the case with less than one per 

 cent, of those which have submai-ine foci, although more 

 than nine-tenths of the epicentres were not more than 

 ten miles from the coast. Indeed, so deaf are the 

 Japanese to the earthquake-sound that it is probably 

 hejvrd by them only in the case of those shocks which 

 originate at a very slight depth below the surface of the 

 ground. 



In all countries, however, the sound-area is less than 

 the disturbed area of a strong earthquake ; and in a 

 disastrous earthquake it may occupy only a compara- 

 tively small region in the neighbourhood of the epi- 

 centre. But there is no constant relation between the 

 two ai'eas; for, in moderately strong or weak earth- 

 quakes, they nearly coincide, or the sound-area even 

 overlaps the other on one or more sides ; while, in a 

 very weak earthquake, it overlaps it in all directions. 

 ^[oreovcr, there are some very interesting cases in which 

 the disturbed area ceases altogether to exist, that is, 

 the sound is heard while no shock whatever is felt. 



That such earth-sounds have the same origin as or- 

 dinary earthquakes is highly probable. They are heard 

 in districts where slight shocks are frequent; and some- 

 times a series of earth-sounds is interrupted by a shock 

 accompanied by a precisely similar noise. A great 

 earthquake is always followed by a crowd of after- 

 shocks, among which earth-sounds occur in great num- 

 bers at places near the epicentre. It would therefore 

 seem that earthquakes and earth-sounds may be 

 traced to the same cause, that the chief difference in 

 reality lies in ourselves, in the sense by which we 

 perceive them — in other words, that an earth-sound is 

 merely an earthquake too weak to be felt. 



A point of some importance is the relative position of 

 the sound-area and disturbed area of an earthquake. 

 So far as known, the two areas never have the same 

 centre. Their longer axes are parallel to one another, 

 but the sound-area is always displaced with respect to 

 the other, sometimes in the direction of the longer axis, 

 but generally in that of the shorter axis. In the latter 

 case, moreover, the displacement takes place towards 

 the line of the fault with which the earthquake appear, 

 to be connected, implying that the loudest sound-vibra- 

 (ions do not come from so deep-seated a portion of the 

 fault as the vibrations which constitute the earthquake- 

 shock. 



In old earthquake catalogues, the sound is generally 

 said to precede or accompany the shock, very rarely 

 to follow it ; in Japan, the sound is seldom, if ever, 

 heard after the shock ceases to be felt, but it is nearly 

 always heard before the shock begins. We may fairly 

 infer from this that the fore-sound is louder than the 

 after-sound. More detailed studies of recent British 

 earthquakes show that the beginning of the sound gene- 

 rally precedes that of the shock in all parts of the 

 sound-area; while the end of the sound more fre- 

 quently follows that of the shock than otherwise, even 

 at very great distances from the centre. In weak earth- 

 quakes, the instant when the sound is loudest always 

 coincides with that when the shock is strongest; and 



this is generally, though not always, the case with strong 

 eai'thquakes. The duration of the sound is as a rule 

 obviously greater than that of the shock. 



In order to give definitoness to the explanation of 

 the phenomena described above, I will assume the truth 

 of the theory which ascribes non-volcanic earthquakes 

 to the friction produced by the sliding of one of the 

 rock-masses adjoining a fault over and against the other. 

 The seismic focus in such a case must bo a surface in- 

 clined to the horizon, and the relative displacement 

 of the two rock-masses will be greatest near the centre 

 of the focus and will die away towards the edges. Thus, 

 from all parts of the focus, there must proceed vibra- 

 tions diflering in amplitude and period, the lai'ge and 

 slow vibrations coming from the central' region, and the 

 small and rapid ones from the margins. It is the latter, 

 1 believe, especially those which come from the upper 

 and lateral margins, which are responsible for the earth- 

 quake-sounds. 



It is evident, on this view of their oi-igin, that the 

 sound will become gradually louder until the sliock is 

 felt, and afterwards die away. The intensity of tli(> 

 sound will also increase with that of the shock in dif- 

 ferent earthquakes; but while the marginal vibrations 

 ai-e limited in amplitude and period, those from the 

 central parts of the focus have a wider range, and 

 therefore the intensity of the sound will not be pro- 

 portional to that of the shock. Similarly, in a violent 

 earthquake, the disturbed area will extend far beyond 

 the sound-area ; while, in a weak earthquake, the latter 

 area will overlap the former. In the limit, the central 

 region of the focus will vanish, and the sound will be 

 heard without any accompanying shock. 



The most perceptible sound-vibrations will be those 

 which come from the upper and lateral margins of the 

 focus, and the boundary of the sound-ai'ea, with respect 

 to that of the distm-bed area, must therefore be shifted 

 towards the fault-line, and also in the direction of the 

 fault if one lateral margin be longer horizontally than 

 the other. 



The sound-vibrations from the margin nearest to the 

 observer will be heard before the shock begins, those 

 from the upper margin and the central region during 

 the shock, and those from the furthest margin after the 

 shock ends. Thus, the fore-sound, on account of its 

 nearer origin, will be more generally noticed than the 

 after-sound; and, for the same reason, will be the only 

 sound heard by Japanese observers. The after-sound 

 will be less frequently heard as the distance from the 

 origin increases ; and the duration of the sound, es- 

 pecially at places near the epicentre, will be greater than 

 (hat of the shock. 



Itcttcrs. 



■ ♦ 



[Tlie Editors do not hold themBelvee reBponaible for the opinions or 



Btatemonts of correi^pondente.] 



♦ — ■— 



IS THE UNIVERSE INFINITE? 



TO THE EDITORS OF KNOWLEDGK. 



Sirs, — The letters of some of your correspondents, in- 

 cluding Mr. Inglis, the latest of them, leads me to 

 suggest a fonn of the above problem which will keep us 

 clear of the theoretical difficulties about infinity I 

 would therefore substitute. Does the stellar universe 

 extend to 1,000 times the average distance of a star of 

 the sixth magnitude? 



The answer seems to me to be, Certainly not, unless 

 (1) there is a medium in space which absorbs light, or 



