JrxE 1, 1899.] 



KNOWLEDGE. 



123 



out the disturbed area. The chief variation would be in 

 its duration, which would be greatest at places in the 

 longer axis of the isoseismals, and least at those in the 

 shorter axis. If the initial impulse were greatest at the 

 north-west end of the focus, the shock would have been 

 strongest towards the end at places to the south-east of 

 the focus, and towards the beginning at places to the 

 north-west. Now, this is what actually occurred — at least 

 according to the majority of the observers who made notes 

 on this point. As a rule, however, there was a break in 

 the shock. In the south-east region, a series of rapid 

 vibrations was first felt ; then came an interval of two or 

 three seconds, during which no motion was perceptible ; 

 and this was followed by a longer series of vibrations, 

 stronger and slower than the first, and compared by many 

 observers to the rolling of a boat when crossing the wake 

 of a steamer, or to the oscillations of a carriage resting on 

 good springs. Some persons at a distance from the centre 

 describe the rocking as gentle and not unpleasant, but near 

 the centre " the vibrations were so fierce and quick that 

 they resembled the beats of an engine going at fifty miles 

 an hour." In the north-west region the order of events 

 was reversed, the first series being longer, more intense, 

 and consisting of slower vibrations. 



It is clear, from the variable nature of the double shock, 

 that the focus cannot have been continuous throughout 

 its length. The waves proceeding from a single focus 

 cannot have been divided by reilexion or refraction at the 

 bounding surfaces of different strata. Neither could the 

 two series correspond to the longitudinal and transverse 

 vibrations from one initial impulse, nor to successive 

 impulses at the same focus. We must therefore conclude 

 that there were two detached foci arranged along a north- 

 west and south-east line. 



An interesting question now arises as to whether the 

 impulses at the two foci were simultaneous or successive. 

 If they were simultaneous, then, at places along a straight 

 line bisecting the line joining the two foci at right angles, 

 the two series of vibrations would arrive together and only 

 a single shock would be observed. Moreover, the boundary 

 between the north-west and south-east regions in which 

 the two parts of the shock were complementary would 

 coincide with the same straight line. But this is not the 

 case. The boundary is clearly a curved line concave 

 towards the south-east, and the two series coalesce and 

 form a single series at places within a hyperbolic band ''' 

 which is traversed by the same boundary line. As the 

 distance of any point on this band from the north-west 

 focus is greater than its distance from the south-east focus, 

 it follows that the north-west focus must have been in 

 action a few seconds before the south-east focus. 



Evidence leading to the same conclusion is afforded by 

 the sound - phenomena of the earthquake. The deep 

 rumbling sound which nearly always accompanies earth- 

 quakes was heard over about two-thirds of the disturbed 

 area. It was most often compared to the noise made by 

 heavy waggons, traction engines, or railway trains passing; 

 sometimes to thunder, wind, the crash of a falling roof or 

 other heavy bodies, or to explosions and the firing of heavy 

 guns. The most marked feature of the sound was its 

 extraordinary depth. It was, in fact, so deep, that it was 

 inaudible to many observers. The records, however, show 

 that persons differ widely in their powers of hearing deep 

 sounds. To some it appeared very loud, like the rumbling 

 of a traction engine heavily laden ; whUe others, in the 

 same places and equally on the alert, heard no sound at 



• The broken line on the map passes along the middle of the 

 hyperbolic band. 



all. The audibility of the sound-vibrations must, of course, 

 decrease with the increasing distance from the origin ; but 

 in the Hereford earthquake it did not depend only on the 

 distance. For instance, in Herefordshire, the percentage 

 of observers who heard the sound was eighty-seven, in' 

 Berkshire it was forty-four, while in the still more distant 

 county of Lincoln it was sixty. If we consider the per- 

 centage in each county to correspond to the centre of the 

 county, we can draw a series of curves of equal sound- 

 audibility, which I have called Uacoustic lines: the meaning 

 of the line marked " 60," say, being that sixty per cent, of 

 the observers, within a small area having its centre on the 

 curve, heard the earthquake sound. Now, these Unes are 

 irregular in form, and by no means concentric with the 

 isoseismal lines. They are much elongated in two direc- 

 tions, one a little east of north-east, and the other a little 

 south of south-west ; and it is remarkable that a line 

 drawn through the points of each curve which are furthest 

 from the centre is concave towards the south-east, and 

 coincides very nearly with the hyperbolic band along which 

 the two series of vibrations were superposed. The explana- 

 tion is evidently that at places along the line of greatest 

 elongation of the curves the sound-vibrations from the two 

 foci arrived simultaneously, and were audible to a larger 

 number of observers' ; and the concavity of this line to the 

 south-east confirms the conclusion already arrived at with 

 regard to the precedence in action of the north-west focus. 



Most earthquakes which occur in non-volcanic districts 

 appear to have their origin in the intermittent growth of 

 faults. The mass of rock on one side of a fault slips, in 

 extreme cases, through several feet; in most cases, perhaps, 

 through a small fraction of an inch. But the area over 

 which the slip takes place may be several miles in length, 

 and the weight of the displaced rock-mass may be very 

 great. The friction resulting from the sudden movement 

 may thus be the source of the waves which, when they 

 reach the surface, are known to us as an earthquake shock. 



Now, since the surface of the fault is inclined to the 

 horizon, the relative positions of the isoseismal lines will 

 be different on the two sides of the fault, and ought, there- 

 fore, to throw some light on the direction in which the 

 fault-surface "hades" or slopes. Keferring to the map, 

 it will be seen that the distance between the isoseismals is 

 greater on the north-east than on the south-west side for 

 the inner lines, and less for the outer ones. The inference 

 from this is that the originating fault of the Hereford 

 earthquake hades to the north-east. The fault-line must 

 therefore be approximately parallel to the longer axis of 

 the isoseismal 8, and lie a short distance to the south- 

 west of this axis. Its exact position cannot be determined, 

 but it probably passes through a point about one mile 

 south-west of Hereford. 



There is no fault on the Geological Survey map of the 

 district to correspond with that suggested by the seismic 

 evidence. But this is not surprising, for the area in which 

 it lies is one covered by Old Red Sandstone, in which the 

 detection of faults is difficult. A short distance to the 

 south-east, however, there is a small triangular area of 

 Silurian rocks, known as the May Hill anticlinal, the 

 north-east side of which is bounded by a fault which, if 

 produced, would coincide very nearly with the earthquake- 

 fault. It should be noticed that the course of this fault, 

 as traced on the Survey map, ends abruptly on the margin 

 of the Old Bed Sandstone, possibly on account of the diffi- ■ 

 culty of following it any further ; and it appears to me 



* The fame reasoning apparently leads to the conclusion that the 

 isoseismal lines should have been distorted in a similar manner; but 

 there was less inequality between the sound-vibrations from the two 

 foci than between the larger vibrations which formed the true shock. 



