part 3] august 1895 in northern Italy. 235 



only at Home, where the time (20h. 51m. 10s.) is in accord with 

 the supposition that the record was due to the arrival of the 

 surface-waves. In 1895 the number of instruments capable of 

 recording distant earthquakes was few, and none of them recorded 

 this earthquake, which may be accounted for by the fact that at 

 Rome, about 100 miles from the outer edge of the seismic area, 

 there was only a slight thickening of the record on the two most 

 sensitive of the instruments in action. Had this earthquake 

 occurred some years later it is probable that more records, and 

 possibly more distant records, would have been obtained from the 

 more numerous, and more sensitive, instruments then installed ; 

 but the Rome record is of interest, as showing that the disturbance 

 was one of the type which gives rise to long-distance records. 



There remain, then, the descriptive records, on which to base an 

 estimate of depth of origin, and, in making use of these, the first 

 difficulty encountered is the explanation of the peculiar distribution 

 of the records. One possible hypothesis is, as has been stated, that 

 there was in reality a large area of tolerably uniform intensity of 

 shock, and that the irregular distribution of the records is due to 

 the presence or absence of potential observers. In this case we 

 are dealing with a shock due to the propagation of wave-motion 

 directly from the batbyseism, and this wave-motion, as revealed 

 by the records, was of twofold character: there was first the 

 undulatory movement, only noticed through the effect of the 

 inclination of the surface produced by it, and also a vibratory 

 movement of greater rapidity, giving rise to the sound phenomenon 

 and to tremors which were everywhere of feeble intensity. The 

 actual amount of the acceleration was everywhere small, but the 

 reports do not admit of the formulation of any precise estimate of 

 the variation : at the outside it may have been twice as great in 

 the central part of the seismic area as in the marginal regions. 

 The rate of variation of acceleration of the wave-particle with 

 distance from the origin has not been investigated; but, if it be 

 taken as inversely proportional to the distance from the origin, 

 the depth would come out as something near 50 miles ; while, if 

 the variation is inversely proportional to the square of the dis- 

 tance, the depth might reach double this figure. 



This is not, however, the only possible interpretation of the 

 records, for it may be that the intervening gaps, from which the 

 earthquake was not reported, represent a real absence of noticeable 

 shocks, that the immediate origin of the earthquake was a series 

 of fractures, of comparatively shallow depth, and that the distri- 

 bution of the records represents, at least approximately, the extent 

 of the sensible shock. In some ways, this interpretation is in 

 accord with the peculiar distribution of the records and the absence 

 of any defined area of maximum intensity. The records lie mostly 

 along a line running about north-eastwards from Pisa, and along 

 another running eastwards from Como, the two lines meeting in 

 the district north-west of Venice ; there is also indication, less 



