Oct. ii, 1888] 



NATURE 



5^7 



burst ; for, while it is possible that similar waves were 

 propagated through the atmosphere during great 

 eruptions in former years, these appear to be the only 

 instances recorded of anything of the kind on such a 

 vast scale since the establishment of continuous self- 

 recording barometers. 



That air-waves caused by the sudden expansion of 

 the erupting gases could leave a perceptible record on all 

 the barometer traces as far as the antipodes of Krakatab, 

 is of itself a sufficiently remarkable fact, but that such 

 waves could record their passage back and forwards no 

 less than seven times, is a circumstance which even now, 

 five years after its occurrence, fills us with astonishment. 

 A selection of forty-seven stations has been made, which, 

 as far as possible, represent the habitable world ; and 

 the times of passage of the wave from Krakatab to the 

 antipodes and from the latter back to Krakatab have been 

 deduced by comparing the significant, and in many cases 

 similarly-shaped, notches in the barometer traces. 



Of course, where, as in the present case, the form of the 

 wave itself was complicated, gradually became deformed, 

 and was traceable for no less than 127 hours from its 

 commencement.perfect accuracy in determining the precise 

 moments of passage of the various phases could scarcely 

 be expected. Yet it is evident on the face of it that a 

 very high degree of accuracy has been attained, by which 

 not only can the precise moment of the great outburst 

 be determined by the simple process of calculating back- 

 wards, but also certain variations of velocity be traced 

 in portions of the wave which took different routes over 

 the globe. 



The general pace at which the air-wave spread outwards 

 in concentric circles from Krakatab as a centre, was 700 

 miles per hour, which is slightly less than the velocity of 

 sound at zero Fahrenheit, viz. 723 miles. The entire circuit 

 of the globe and back was thus made in about thirty-six 

 hours. Also, by a careful comparison of times and 

 probable errors, the probable moment of the greatest 

 explosion is calculated to have been 2h. 56m. G.M.T., or 

 9h. 58m. local time, on the morning of August 27.' 



This great explosion appears to have been not only the 

 culminating point of the Krakatab eruption (the pre- 

 ceding minor outbursts appearing as a mere roughening 

 of the barometer scale, or a series of moderate oscillations 

 on that of the gasometer at Batavia), but owing to its 

 surpassing intensity, a feature altogether peculiar to this 

 eruption, and one by which it will always be distinguished 

 from others, such as that of Asama (Japan) an J Skaptar 

 Jokull in 1783, or Tamboro in 1815, which, in respect of 

 the amount of material ejected in the form of lava, and 

 other effects, appear to have equalled if not exceeded it. 



One of the most interesting results of this discussion 

 of the Krakatab air-wave has been the discovery of its 

 variation of speed according as it travelled with or against 

 the earth's rotation. As a general fact it may be said 

 that such variation is plainly traceable to the prevalent drift 

 of the winds. 



Thus in the extra-tropics the wave moving from west 

 to east was accelerated, and that from east to west 

 retarded, by about 14 miles per hour ; while within the 

 tropics the wave which passed through Mauritius and 

 Loanda was affected in a precisely reverse manner, the 

 passage eastwards being retarded, while that westwards 

 was comparatively unaffected, the amount corresponding 

 to an east to west wind of about 10 miles an hour. It is 

 at least curious to notice, that on p 35 of the " Motions 

 of Solids and Fluids," by Prof. Ferrel (Washington, 1882), 

 the value of the due E. to W. component of the trades 

 between 15° N. and S. lat. is given as 10 miles per hour, 

 while the mean of the W. to E. component of the anti- 

 trades for latitude 45° at the earth's surface and a height 

 of 3 miles above it, is exactly 14^ miles per hour. 



1 This differs by only 4 minutes from ioh. 2m., the epoch determined from 

 fewer data by M. Verbeek. 



The greatest general retardation took place in the 

 Southern Ocean, possibly owing to the low temperature of 

 the southern hemisphere in August. All these points are 

 very distinctly shown in the diagrams. 



As regards the actual sounds, the facts are without pre- 

 cedent. The unvarnished record reads like a fairy tale. 

 When we are told that at distances of over 2000 miles 

 from the volcano, the noise was like the firing of heavy 

 guns, and that at numerous points of the Indian Ocean 

 steamers were despatched in search of supposed vessels 

 in distress, we are prepared to accept with less hestitation 

 the numerous other collateral evidences of the enormous 

 explosive energy which generated them. 



The area over which the sounds were heard is roughly 

 estimated at one-thirteenth of the entire surface of the 

 globe. In other words, it was nearly equal to Europe and 

 Africa together, or slightly exceeded that of both Americas. 

 All these details are illustrated by numerous diagrams. 



Part III., by Captain W. J. L. Wharton, R.N., F.R.S., 

 deals with the so-called seismic sea waves generated during 

 the eruption ; one of which not only dealt death and destruc- 

 tion all over the Straits of Sunda, but travelled as far as 

 Cape Horn, and possibly the English Channel. 



It appears that there were two sorts of waves generated 

 — one of long period (two hours), which alone recorded 

 itself on the automatic gauges and travelled to great 

 distances ; and others of much shorter period, which 

 were mostly confined to the immediate vicinity of the 

 volcano. 



The only hypothesis by which the facts can be recon- 

 ciled, according to Captain Wharton, is that at the time of 

 the greatest explosion, at 10 o'clock on August 27, "waves 

 of both characters would be more or less synchronously 

 formed," the longer wave being caused by upheaval, and 

 the shorter ones, which caused the destructive effects in 

 the Straits of Sunda, by the displacement due to ejected 

 masses or fragments of the volcano falling into the sea all 

 round it. 



In proof of upheaval, which appears to be the only 

 probable cause of the longer wave, Captain Wharton cites 

 the generally shallowed condition of the sea immediately 

 surrounding Krakatab, especially on the northern side. 



We cannot, however, help observing that, according to 

 Prof. Judd, the geological evidence is entirely against 

 upheaval throughout the area ; and the formation of the 

 new shoals and islands is attributed by him solely to the 

 piling up on the sea floor of the coarser matter, including 

 the framework of the volcano, which was ejected during 

 the explosive outbursts. It is a remarkable fact, indeed, 

 that during the eruption there was no trace of any local 

 seismic disturbance such as might be supposed to accom- 

 pany an upheaval of the ground. A variety of peculiar 

 effects were witnessed, such as clocks stopped, lamps 

 broken, and houses cracked, but all of these were traceable 

 to air and not earth vibrations. 



The precise cause, therefore, of the long wave will, as 

 Captain Wharton says, " ever remain to a great extent un- 

 certain." One fact, however, remains clear — that both it 

 and its minor predecessors were distinctly connected wit! 

 corresponding explosions from the crater, which recorded 

 themselves in unmistakable language on the gasometer 

 pressure-gauge at Batavia. Whatever the precise proxi- 

 mate cause, therefore-— whether slow upheaval, according to 

 Cap:ain Wharton, or the impact of falling matter, according 

 to Prof. Judd— the action commenced with each explosion. 

 The height of the local manifestation of the great wave 

 at 10 o'clock is estimated to have been 50 feet, though in 

 places where it reached the shore it appears to have run 

 up to 70 feet. 



The terribly destructive effects of these shorter " super- 

 seismic" waves, of which this one appears to have been 

 the greatest, are amply detailed in M. Verbeek's Report, 

 and the accompanying views of the localities visited. 

 They reached the above majestic height only in the 



