2/2 



NATURE 



{yuly 2 1, 1887 



appear that the Appalachians offered any sensible barrier 

 to the progress of the deeper waves, but it does 

 appear that they affected in a conspicuous degree the 

 manner in which the energy of the waves was dissipated 

 at the surface. Another minimum area was found in 

 Southern Indiana and Illinois, and also in Southern 

 Alabama and Mississippi. There is a curious circum- 

 stance connected with the minimum area in Indiana and 

 Illinois. On February 6 last, an earthquake of notable 

 force occurred in just this locality. Circulars were sent 

 out at once, and on plotting the isoseismals they showed 

 a singular coincidence in almost exactly filling the vacancy 

 or defects of intensity of the Charleston earthquake. At 

 present there is nothing to indicate whether this coin- 

 cidence is accidental or whether there is some hidden 

 relation. 



Where the waves passed into the newer delta region of 

 the lower Mississippi, the surface intensity of the shocks 

 rapidly declined. This is indicated in the map by the 

 compression of the isoseismals in those localities. We 

 incline to the opinion that this sudden diminution of the 

 intensity is due to the dissipation of the energy of the 

 waves in a very great thickness of feebly elastic, imper- 

 fectly consolidated, superficial deposits. It is a matter of 

 common observation in all great earthquakes that the 

 passage of the principal shocks from rigid and firm rocks 

 into gravels, sands, and clays is, under certain circum- 

 stances, attended with a local increase in the amplitudes 

 of the oscillations and in the apparent local intensity and 

 destructiveness, ind the reason for it is intelligible. But 

 where such looser materials are of very great thickness 

 and great horizontal extent the reverse should be ex- 

 pected. For when a wave passes from a solid and highly 

 elastic medium into a less solid and imperfectly elastic 

 one, the amplitude may be suddenly increased at the 

 instant of entering ; but so rapid is the extinction, that, 

 if the new medium be very extensive, the impulse is soon 

 dissipated. 



Many reports throughout the Central States indicate 

 localities of silence which are not expressed upon the 

 map. The reason for omitting them is that it has been 

 impracticable to secure a sufficient density of observation 

 {i.e. a sufficient number of reports per unit area) to enable 

 us to mark out and define these smaller areas with very 

 gi'eat precision. To do this for the whole country would 

 require some tens of thousands of observations and the 

 expenditure of tens of thousands of dollars to systematize 

 and discuss the data. A map shaded to show the varying 

 intensity by varying the depth of the shading would have a 

 mottled appearance, in which the mottling would be most 

 pronounced in the areas of a little below the mean in- 

 tensity, say between the isoseismals 3 and 5. This fact is 

 of great importance in the interpretation of the isoseis- 

 mals, for the omission to consider it results in giving to 

 the middle isoseismals too high a value. In any isoseis- 

 mal zone, what we should like to ascertain is the mean 

 intensity of the whole area included within that zone. As 

 a matter of fact, the data we possess consist more largely 

 of maximum than of minimum or average intensities, and 

 therefore tend to considerably augment the mean derived 

 intensity above the true mean. This will become appar- 

 ent by an inspection of the map where the zones of 5, 6, 

 and 7 intensity are disproportionately broad, while those 

 of 3 and 4 are disproportionately narrow. We have not 

 attempted to allow for this source of error, though fully 

 aware of it, because we had no means of determining 

 what allowance to make. We have drawn the lines 

 wholly upon the face of the returns, and the investigators 

 who may attempt to utilize our results must grapple with 

 the corrections as best they may. 



Throughout the States of North Carolina, South Caro- 

 lina, Georgia, and North-Eastern Florida, and in general 

 anywhere within about 250 miles of the centre, the energy 

 of the shocks was very great. At Columbia, Augusta, 



Raleigh, Atlanta, and Savannah, the consternation of all 

 the people was universal. The negroes and many of the 

 poor whites were for a week or two not exactly demoral- 

 ized, but intensely moralized, giving themselves to reli- 

 gious exercises of a highly emotional character, the 

 stronger and deeper natures among them being im- 

 pressed with a feeling of awe, the weaker natures with a 

 feeling of terror. And this was general throughout the 

 large region just specified. In all of the large towns 

 within 200 miles of Charleston more or less damage was 

 suffered by houses and other structures. Walls were 

 cracked to such an extent as to necessitate important 

 repairs ; dams were broken, chimneys were overthrown, 

 plastering shaken from ceilings, lamps overturned, water 

 thrown out of tanks, cars set in motion on side tracks, 

 animals filled with terror, fowls shaken from their roosts, 

 loose objects thrown from mantels, chairs and beds moved 

 horizontally upon the floor, pictures banged against the 

 walls, trees visibly swayed and their leaves agitated and 

 rustled as if by a wind. These occurrences were general, 

 and were more strongly marked until they became terrify- 

 ing and disastrous as the centre of the disturbance was 

 approached. At Augusta, 1 10 miles distant from the epi- 

 centrum, the damage to buildings was considerable ; and 

 at the arsenal in that place the commanding officer's 

 residence was so badly cracked and shattered as to 

 necessitate practical reconstruction. In Columbia, 100 

 miles distant, the shock was very injurious to buildings 

 and appalling to the people, but no substantial structures 

 were actually shaken down. In Atlanta, 250 miles dis- 

 tant, there was no worse injury than falling chimneys 

 and some slight cracks in the walls, but the houses were 

 instantly abandoned in great alarm and confusion by their 

 occupants, and many preferred passing the night in the 

 streets to re-entering their dwellings. At Asheville, N.C., 

 230 miles distant, and at Raleigh, 215 miles distant, the 

 shocks were quite as vigorous as at Atlanta. 



Coming nearer the seismic centre we find the intensity 

 increasing on all sides as we approach it. The region 

 immediately about the epicentrum in a great earthquake 

 always discloses phenomena strikingly different from those 

 at a distance from it, and the differences are not merely 

 in degree but also in kind. The phenomena characteristic 

 of the epicentral area cease with something like abrupt- 

 ness as we radiate away from the epicentrum. The 

 central phenomena are those produced by shocks in which 

 the principal component of the motion of the earth is 

 vertical. Proceeding outwards, these predominating 

 vertical motions pass, by a very rapid transition, into 

 movements of which the horizontal component is the 

 greater, and in which the undulatory motion becomes 

 pronounced. The epicentrum, and the zone immediately 

 surrounding it, is the portion of the disturbed tract which 

 merits the closest attention, for it is here that we may find 

 the greatest amount of information concerning the origin 

 and nature of the earthquake. To appreciate this we will 

 venture to offer some theoretical considerations. 



Allusion has already been made to the indefinite 

 character of the data used for estimating the intensity of 

 the shock. There is no unit of intensity which is at 

 present available. In selecting certain effects of an earth- 

 quake to characterize varying degrees of intensity, the 

 most that can be hoped for is a means for discriminating 

 whether the relative energy of a shock is greater or less 

 in one locality than in another. But how much greater 

 and how much less — in conformity with what law — is a 

 problem which remains to be solved. An earthquake im- 

 pulse, however, is a form of energy transmitted as an 

 elastic wave through the deeply-seated rocks, and its pro- 

 pagation and varying intensity are subject to the laws of 

 wave-motion There must be, therefore, some typical law 

 governing the rate at which such a wave diminishes the 

 intensity of its effects as it moves onward. To anticipate 

 the objection that this typical law would apply only to a 



