early. Table 1 shows the residuals P(0-C) at 40 selected stations and also the mean residuals in four different groups of these stations, arranged according to their direction from the focus. The assumed values of the epicentre for which figures are given are () 29o.7 N., 66°:7 E., which is very near the middle of the inner region of maximum disturbance marked in his map by Mr. West and (ii) 29o.6 N., 66°.5 E., the epicentre which is found to fit the observations best. The assumed value of t, or epicentral time is 21h 32m 59s G.M.T. It may be recalled that to is not necessarily the actual time of the earthquake ; it is the time “ which makes t-t, at short distances proportional to the distance”. If the focus is at the surface, the actual time of occurrence or the hypocentral time of the earthquake, according to Jeffreys, is about 5.8 secs. earlier than to. TABLE 1.–Comparison of Epicentres. Ph(0-C)-Epc. 29°•7 N., 66°:7 E. (0-C) calculated using geographical co-ordinates P,(0-C)=Epc. 29°16 N., 666”.5 E. and J. B. tables. P,(0-C)—Epe. 29°-6 N., 66°:5 E.--P(0-C) calculated with observed travel times corrected for ellipticity (Bullen's tables) and using Jeffreys' revised tables (1937) of traveltimes. The arcual distances A of the observatories from the epicentres in the above table have been obtained from the geographical coordinates and the calculated times of travel P,(C) and P.(C) taken from Jeffreys-Bullen tables. In the last column, the differences P,'(0-C) have been calculated correcting the observed travel-times for the ellipticity of the earth so as to give results as for the standard sphere (using Bullen’sl “ Tables for reduction of apparent traveltimes of P and S seismic waves" and Jeffreys' revised tables of travel-times. 1937). It will be seen that if we adopt as epicentre 29o.7 N., 66°:7 E., there is a considerable difference between the mean residuals from the westerly and easterly groups of observatories and that this difference practically vanishes if we adopt 29o.6 N., 66°:5 E. No appreciable effect is produced on the mean 1 K. E. Bullen, " Tables for reduction of apparent travel times of P and S seismic residuals by correcting for ellipticity although there are significant differences in individual residuals. From the magnitude of the residuals, it is clear that a correction of-0-5 sec. is necessary for to. The corrected value of t. is 30d 21" 32" 58.5* G.M.T. waves Ne:v Zealand J. Sc. and Tech., Vol. XIX, No. 1, pp. 47–54, (1937). 2 H. Jeffreys, “ Further corrections to P, S and SKS Tables M. N. R. A. S., Geoph. Suppl. 4, No. 3, p. 242, (1937). The frequency distributions of Presiduals in the two cases are shown in figs. 1 and 2. 205 1 -205 -10 10 MEAN P (0-c) Fig. 1.-P residuals : (0-C) with to assumed to be 21h 32m 59s and using J. B. tables. اسلا -205 20$ படILI 10 P (0-0) MEAN Fig. 2.-P residuals : (0-C) with to assumed to be 21h 32m 59s and using J. B. tables with ellipticity correction. Some noteworthy features of the seismograms. The seismograms of this earthquake obtained at the Indian observatories are somewhat complicated. In the Agra Milne-Shaw record there is an impulsive beginning of a phase about 15s after the beginning of the first ep. In the Bombay seismograms, the following phases are recognisable : The phase i* marks the beginning of a series of long period oscillations superposed on the much more rapid oscillations usually characteristic of the preliminary phase at this distance. It perhaps corresponds to the beginning of P, though one would hardly have expected to see direct P, at this distance (12°-1). In the seismograms of Hyderabad and Kodaikanal, one notices phases 16 sec. and 14 sec. respectively after the first incidence of disturbance. All these suggest that the shock was a multiple one, the first impulse being feeble and the second one marked iP, in Table 2, being the one recorded at the more distant stations. In more distant stations. In this connection, reference may be made to the following observations made by Mr. West (loc. cit., p. 212). " At least five to ten seconds before the main shock started, a small tremor was felt which was sufficiently strong to be recognized as an earthquake.” This was at a place about four miles north of Quetta. At Quetta itself, "a sentry on duty on top of the Ammunition Depot noticed a shake |