NA rURE 



S^Dec. 20, li 



"As the earlier disturbances, on August 27 and 28, extend 

 over several hours, it became necessary to fix on certain 

 sufficieiitly well defined points in the curves representing the 

 barometric pressure, from which to measure the epochs of the 

 passage of successive disturbances. The first and second of the 

 series are, in almost all the curves, well defined and generally 

 similar in form, commencing with a distinctiise, which is again 

 followed by a diNtinct fall, the fall being shorter than the ri-.e. 

 These features are followed by a less definite rise succeeded liy a 

 shallow fall, after which there is again a rise, which gradually 

 passes into the more regular trace. 



" The third Mv\ fourth of the disturbances can be traced in all 

 the curves, but they no longer exhibit the same characters, and 

 are usually nothing more than a sudden sharply defined rise, 

 though in front of some of these there is a more or less distinct 

 trace of a hollow. 



"The fifth and sixth of the series become less distinct and are 

 lost at several stations, being usually rises ; while a seventh 

 faint disturbance, as a shallow hollow, can be traced in a few of 

 the curves, after which nothing can be distinguished. 



" By a comparison of the time intervals between the first and 

 third, the third and fifth, and the fifth and seventh disturbances, 

 and assuming (which the facts seem to justify) that the velocity 

 ■of the wave has remained unchanged in its passage from east to 

 west, it would appear that the first well-defined rise in the first 

 of the series corresponds to the rises which are prominent in 

 those succeeding it. And the same conclusion has been drawn 

 from an examination of the second and fourih compared with 

 the fourth and sixth of the .series. 



"Adopting these conclusions, the times of the successive pas- 

 sages of the initial ri-e have been measured from the curves, 

 suitable allowance having been made where the rise was difficult 

 to trace, or, as sometimes happened, a hollow appeared corre- 

 sponding in position with the hollows in the earlier form of 

 the disturbances. There is, of course, some doubt attaching to 

 these measurements, but their general consistency seems to indi- 

 cate that they mry be accepted as fairly representing the facts 

 under discussion. 



" The following table gives the results of these estimates of the 

 times at which the successive waves passed the several stations, 

 reckoned from midnight of Aug. 26, in Greenwich mean time': — 



Toronto 



Valencia 



Coimbra 



Armagh 



Falmouth 



•Glasgow 



Stonyhurst .... 

 Aberdeen . . . . 



Kew 



Greenwich 



Paris 



Bru sels 



St, Petersburg 



Times of passage of wa 



I. II. III. IV, 



m. h. m.!h. m. 

 io6r 30I 

 5o|62 5,87 55 

 30 62 401 

 40.62 15187 45 

 25162 15 



96 10 



124 25 



201124 30 

 97 45!i24 30 



62 20;S7 35197 30 



62 25187 40 

 62 30J87 20 

 62 30( 



0,62 50I 

 0562 55 86 45 98 40! 

 30163 50.84 4o| 



97 30124 5 



98 30 



98 0|I24 5 



" From these figures are deduced the intervals between the 

 succe-sive pass.ages of the waves from east to west, and from 

 west to east, respectively, or of the times of travelling round (he 

 earth, which are .shown in the next table, for all stations 

 excepting Toronto. 



" From the re-ults thus obtained it would follow that the wave 

 travelled round the earth from east to west in 36h. S7m., being 

 at the rate of '1026 hour for one degree of a great circle of the 

 earth, and from west to east in 35h. 17m., being at the rate of 

 ■098 h >ur for one degree. From the velocities thus determined 

 the probable time of the origin of the wave has been calculated 

 from the known distance of e.ich place from Krakatoa, the time 

 occupied in the passage of the wave from Krakatoa to the place 

 of observation, ar.d the observed time of ihe passage of the waves, 



" The mean value thus obtained from the waves moving from 

 east to west for the time of the origin of the disturbance at 



Intervals occupied in travelling round the earth. 



Valencia .. 

 Coimbra .. 



Armagh 



Falmouth .. 

 Glasgow .. 

 Stonyhurst 

 Aberdeen ., 



Kew 



Greenwich 



Paris 



Brussels 



St.Petersbrf 



Mean ex- ' 

 eluding 

 Toronto 



h. m. h. m. 



36 57 35 35 



36 40 I 35 45 



37 o|35 30 

 37 2 I 35 15 

 37 3:35 20 



36 55 i 35 35 



37 0,35 25 

 36 57(35 15 



36 45 



37 5 

 36 43 



36 57 



35 20 

 35 10 

 35 10 



35 24 



h, m. 

 34 5 



34 5 



35 30 

 35 10 



35 5 



36 o 



35 30 



3S 9 35 17 



Krakatoa is 2"52h. Greenwich mean time, or 9'53li. local time, 

 that is 9h. 32m. a.m. of August 27.' 



" In like manner the waves travelling from west to east gave 

 results which were exhibited in another table. 



" The mean value of the time of the origin of the disturbance 

 obtained from the waves moving from west to east is therefore 

 2'2oh. Greenwich mean time, or 9"2lh. local time, that is, gh. 

 13m. local time. 



" The mean between the two values obtained from the waves 

 travelling against the earth's motion of revolution, and those 

 travelling with it is 2h. 24m. Greenwich mean time, orgh. 24111, 

 local time, August 27, 



"The velocity of the waves in miles w-ill be for those which 

 travel from east to west 674 miles per hour, and for those passing 

 from west to east 706 miles per hour. The velocity of sound is 

 for a temperature of 50° F. 757 miles an hour, a id for 80° F. 

 781 miles an hour. With a temperature as low as zero F. the 

 velocity w^ll only be reduced to 723 miles an hour, which is still 

 considerably in excess of the greater of the observed velocities. 

 The excess of the velocity of the waves which travelled in the 

 same direction as the earth's motion of revolution, that is, frim 

 west to east, over that of those which pas ed in the opposite 

 direction, is about 32 miles an hour, which might be accounted 

 for by the circumstance that the winds along the paths of the 

 waves would, on the whole, be from the west, which wouldcau.se 

 an increase in the velocity of the one set, and a diminution in that 

 of the other, so that the observed difference of 32 miles would 

 correspond to an average westerly »ind of 16 miles an hour, 

 which is not improbable. 



" It should be observed that the path of the wave which passed 

 Toronto approached very near to the North and South Poles, and 

 that the velocity in both directions appeared to be somewhat less 

 than in the w aves w hich pa-sed over Central Europe, The wave 

 w hich passed noi thw ards over Asia travelled at the rate of about 

 660 miles an hour, or about 15 mile, an hour slower than the 

 wave which passed over Great Britain from east to west. This re- 

 duction of velocity seems to be within the limits of what might 

 be due to the low teinperature of the regions. 



" The wave travelling from east to west having been per- 

 ceptible on the barometer traces at several of the stations until 

 about 122 hours after its ori:jin, and its velocity having been 674 

 miles an hour, it had travelled before its extinction more than 

 82,200 miles, and had passed 3i times rnind the entire circuit 

 of the earth, 



" It is further worthy of notice that during August 30 

 and 31 and September I, a very severe cyclonic storm was 

 crossing the North Atlantic, and that the wave coming from the 

 wcstw ard early on the 31st, No. VI. of the series, must have 

 passed on in front of the cyclone, and that its next transit would 

 have carried it into the very centre of the cyclone near the British 

 Isles on the afternoon of September i. This perhaps ac- 



* At these stations the fifth transit can 



' It has not been th .ught necessary I 

 the separate values deduced from the s 

 from the mean by in no case more than a 



