586 



SCIENCE. 



[Vol. IX., No. 228 



came througli circular letters, newspaper reports, 

 etc., with which untrained and not very trustworthy 

 observers have much to do. The third zone con- 

 sists of all that is left of the disturbed area : over it 

 the effects were transient, and all evidence rests on 

 human testimony, unsupported by that of material 

 objects. 



Thus it would seem, that, in the construction of 

 the map, isoseismal lines would be drawn with three 

 different degrees of confidence, and that they must 

 be drawn more freely, and with less attention to de- 

 tail, as they are farther removed from the epicentral 

 tract. Local variations in intensity-estimates should 

 have less weight, and the lines would approximate 

 more nearly to smooth curves. On the map as 

 drawn by Messrs. Button and Hayden, this order of 

 things appears to be reversed : the smoothest, most 

 regular curves are those immediately surrounding 

 the epi centrum, while they become more irregular 

 as the distance from that point increases. 



In work of this kind, irregular and sinuous lines 

 imply numerous and reliable observations, while 

 those more regularly and uniformly curved will gen- 

 erally be drawn for areas over which observations 

 are few, and not of sufficient weight to show more 

 than the general trend of the line. For these reasons 

 it appears to me that the map is faulty, in that 

 too much weight has been given to individual ob- 

 servations at great distances from the epicentrum ; 

 that the sinuosities and irregularities in the lines, 

 particularly those of the Mississippi valley, do not 

 represent any thing real; that they should be 

 smoothed out ; and that it is doubtful if stifficient 

 evidence exists for the construction of the two iso- 

 lated areas, surrounded by closed curves, which 

 appear in northern Illinois and in southern Indiana 

 and Illinois. I venture to suggest, in regard to the 

 latter 'area of silence,' that its existence may to some 

 extent be due to the fact that information concerning 

 that area was collected several months after the 

 occurrence of the earthquake. 



One of the most interesting features of this paper 

 is the method employed in determining the depth of 

 the seismic centre. Under certain restrictions, no 

 criticism can be made upon the analysis of the prob- 

 lem ; but in its practical application it is, in my 

 judgment, open to serious objection. It is not easy 

 to decide what is the best measure of the ' intensity ' 

 of an earthquake. A simple expression for it, and 

 that accepted by Messrs. Button and Hayden, is 

 * the energy per unit area of wave-front.' 



This definition once adopted, their analytical and 

 graphic treatment of the problem is elegant and 

 satisfactory ; but in the application of the method to 

 the Charleston earthquake, or to any other, it is im- 

 portant to ask whether any means exists for deter- 

 mining the ' intensity ' as defined above. While it is 

 true that the disapjDearance of the consonant a from 

 the abscissa of the points of inflection renders it inde- 

 pendent of the absolute intensity, it miist not be for- 

 gotten that in determining relative intensities the 

 thing to be kept in mind is ' the energy jDer unit 

 area of wave-front.' As far as can be seen from the 

 contents of the paper, the result depends on the tin- 

 justifiable assumption that surface destruction is pro- 

 portional to this. It is a well-established fact that 

 the destructive effects of a motion are not pro- 

 portional to the energy involved, and in earthquakes 

 many things combine to produce what is ordinarily 

 called the ' intensity' of the shock, or, perhaps bet- 



ter, its ' destructiveness.' An exact expression for 

 this is extremely desirable, and it seems to me that 

 Professor Milne has approximated to it pretty close- 

 ly in adopting, as he has, the ' maximum acceleration 

 of the earth-particle in a horizontal plane.' There 

 can be little doubt that horizontal movement is more 

 effective in overturning and destroying buildings, 

 chimneys, etc., than vertical; yet the fact, if it be a 

 fact, finds no expression in the method of Messrs. 

 Button and Hayden. Their formula and curve de- 

 mand the maximum intensity at the epicentrum ; 

 and this is correct, according to their definition 

 of intensity. 



But does the greatest destruction take place at tbe 

 epicentrum, or is it to be found in a zone whose ra- 

 dius depends on the depth of the seismic centre ? I 

 would not venture to place my own judgment, based 

 upon a hasty examination along a single line, against 

 that of an observer who has gone more leisurely over 

 the field ; but, as I can nowhere discover in the 

 paper a distinct statement as to where the most de- 

 structive effects were observed, I may remark that it 

 appeared to me that there was much less destruction 

 in the neighborhood of the epicentrum, where the 

 vertical component of the motion seemed to have 

 predominated, than in and about the city of Charles- 

 ton. 



Of course, it is possible that from a study of the 

 surface disturbance the relative amount of energy 

 per unit area of wave-front at different points may 

 have been worked out, and the point of inflection 

 found from these results; but it would be an ex- 

 tremely complex problem, and, in addition to diffi- 

 culties already suggested, it is complicated by the 

 fact that the normal motion of the particle must be 

 changed as the wave emerges from the earth : this, 

 indeed, would stand in the way of getting just what 

 is desired from perfect instrumental records, as, 

 at best, they can only reveal surface movements. 



I am unable to agree with the conclusion of 

 Messrs. Button and Hayden expressed in the state- 

 ment that the amplitude of vibration of the earth- 

 particle was in some places not less than ten inches 

 or a foot. So large an amplitude appears to me to be 

 extremely improbable. It is only within a few years 

 that any thing like accurate measures of amplitude 

 have been made ; and it is well known, that, wherever 

 it has been measured, it has been found to be 

 small. 



In the ' general run ' of Japanese earthquakes, the 

 amplitude has been found to be not much greater 

 than a millimetre, and often less. In a few cases it 

 has been several millimetres ; and I believe in one or 

 two, which were nearly ' destructive,' and by which 

 chimneys were overthrown and walls cracked, it has 

 been as high as ten or twelve millimetres. It will 

 be noticed, however, in examining these reports, 

 that, in most of the cases in which large amplitudes 

 are reported, the disturbances were of unusiial 

 length. 



Although, in the construction of their numerous 



Gray 

 ' steady-point ' seismographs, Messrs. Ewing (I want 



Milne 

 to be careful not to put any one of these names 

 first) have well-nigh revolutionized the science of 

 seismology, I am inclined to the o]3inion that in a 

 prolonged disturbance the ' steady point ' is likely to 

 be set in motion, and that a magnification of the 

 amplitude may sometimes result. A very large 



