May i, 1 884 J 



NA TURE 



quantity of lava, which was mostly shot as fine dust out of 

 the crater. The porous nature of the ejected substances 

 — pumice-stone was almost the only substance formed — is 

 doubtless to be ascribed to the steam, which was blown 

 with great force through the lava. I must keep the more 

 detailed description of the way in which the eruption 

 proper was prepared and took place for a future time, as 

 drawings are requisite for a right understanding of the 

 matter. I must still observe here, however, that through 

 the Krakatoa eruption our notions about the shape and 

 extent of subterranean regions will probably be much 

 modified. If it may be assumed that there exists a con- 

 nection between our eruptions, the heightened activity of 

 the volcanoes in the Indian Archipelago since that time, 

 and the earthquakes in Australia which took place simul- 

 taneously with and succeeded the eruptions of the last 

 days — in any case a remarkable coincidence — then much 

 larger dimensions will have to be allowed to those regions 

 than the present geologists are accustomed to assign 

 to them. 



Krakatoa is the only point which has been active. 

 There are reports that Sebesi and the Radja Bassa have 

 also shown activity, but this is inaccurate. 



Of the old Krakatoa there is no detailed survey : the 

 English and Dutch sea-charts, both on a small scale, 

 besides a couple of sketches taken by Buijskes in 1849, 

 and by me in 1880, alone give some idea of the surface- 

 formation of the island. From the two sketches may be 

 seen that the island had three tops — the northern, called 

 Perboewatan (in some reports Roewatan), was the lowest 

 of the three, and showed streams of lava on various sides ; 

 this is the point which first began to be active in May 

 18S3, and which probably also ejected pumice-stone in 

 i860. The centre top bore the name of Danan, and was 

 active also in August 1S83. The southern peak, the 

 mountain Rakata properly speaking (which was corrupted 

 into Krakatoa), was by far the highest point of the whole 

 island, and, according to the sea-chart, 822 metres high. 

 This point is also an old crater, but was not active 

 in 1883. 



On May 20, 1883, the Perboewatan began suddenly to 

 show signs of activity ; that nothing was known before 

 then of an approaching eruption must be ascribed to the 

 fact that Krakatoa was uninhabited, and only visited 

 occasionally by Lampong fishermen, who went no further 

 than the coast. Otherwise it would seem inexplicable 

 that no previous signs of it should have been noticed. 

 The eruptions lasted with various degrees of violence, 

 and with intervals, till August 26 ; while latterly also the 

 crater of Mount Danan became active. Though in them- 

 selves not unimportant, these eruptions were insignificant 

 compared to what followed. On August 1 1 trees were 

 still growing on the peak, so that the destruction of veget- 

 able life was then still limited to the immediate neigh- 

 bourhood of the craters. On the 26th the explosions 

 much increased in violence, and they reached their maxi- 

 mum on the 27th, at io a.m. They then diminished in 

 strength, but lasted still the whole night of Monday to 

 Tuesday, till they suddenly ceased on the 28th, at about 

 6 a.m. 



About the eruptions from May till August 26 little of 

 importance has. on the whole, become known ; all I 

 have been able to collect will be mentioned in the 

 detailed report. 



The eruptions of August 26 and 27 were accompanied 

 by violent detonations and air vibrations. During those 

 days almost incessantly a rumbling sound was heard which 

 resembled the noise of thunder at a distance, the explo- 

 sions properly speaking were accompanied by short 

 detonations which can best be compared to heavy cannon 

 shots, but the most violent detonations were still shorter 

 and more rattling, and cannot be compared to any other 

 sound. 



The sounds of the explosions in May were heard in 



a north-west direction at Moeara Doea in Palembang, 

 and at Bintoehan, in the division of Kauer in Benkoelen, 

 respectively 230 and 270 kilometres from Krakatoa ; the 

 transmission of the sound on August 27 surpasses, how- 

 ever, all which is known of the kind. The explosions 

 were heard in Ceylon, in Burmah, at Manilla, at Doreh in 

 the Geelvink Bay (New Guinea), and at Perth on the 

 west coast of Australia, besides all the places which lie 

 closer to Krakatoa than the above-mentioned. If a circle 

 is drawn from Krakatoa with "a radius of 30°, 1,800 geo- 

 graphical miles, or 3,333 kilometres, the circle will go 

 exactly over the furthest points where the sound was 

 heard. The furthest distance between the points east 

 and west where the sound was heard is therefore 6o° 

 (the diameter of the circle) or one-sixth of the whole 

 circumference of the earth. The surface of this circle, 

 or rather of the spherical segment, comprises more than 

 one-fifteenth of the surface of the earth. In historic 

 times no eruption is known of which the sound was 

 transmitted over such an enormous area. At the eruption 

 of Tombora in Sumbawa, in 1815, the radius of the circle 

 within which the sound was heard, was but half the size, 

 namely 15°, the surface being therefore 3^93 times smaller. 



If a circle with the same radius, namely 30 , is drawn 

 round the earth, taking Amsterdam as the centre, the 

 circle would have the following course. The northern 

 point lies 82 north latitude, thus north of Spitzbergen ; from 

 there the circle runs to the middle of Novya Zemlya, thence 

 along the Ural Mountains to Orenburg, Tiflis, Damascus, 

 Jerusalem, Suez ; crosses the Tropic of Cancer at about 

 1 5 east longitude from Greenwich, reaches the most 

 southern point at 22 north latitude in the Desert of 

 Sahara, crosses the Tropic of Cancer once more at 5 west 

 longitude from Greenwich, runs close along Ferro, includes 

 the Canary Islands and Azores, besides the greater part 

 of Greenland, and runs back to the starting point north 

 of Spitzbergen. In various places it was observed that the 

 strongest detonations were heard at different hours, and 

 also that in places in the neighbourhood of Krakatoa little 

 or nothing was heard of the sound, whereas it was heard 

 very distinctly in places further removed. Thus for ex- 

 ample the loudest report was heard at Buitenzorg at a 

 quarter to seven, at Batavia at half-past eight, at Telok 

 Betong at ten o'clock on the morning of the 27th. This was 

 caused principally by the direction of the wind ; it appears 

 clearly from the reports that the sounds were loudest on 

 the side of Krakatoa whither the wind blew, and the fine 

 ash particles were blown. But this does not yet explain 

 the fact that the sound was sometimes better heard in 

 places that were further off than in those that were nearer 

 when those places lay in the same direction from 

 Krakatoa, such as, for example, Anjer, Serang, and Bata- 

 via. This phenomenon is to be ascribed solely to the great 

 quantity of ash particles which were present in the lower 

 atmosphere. If one assumes for example the presence of 

 a thick cloud of ashes between Krakatoa and Anjer, this 

 would act on the sound waves like a thick soft cushion ; 

 along and above such an ash cloud the sound may be 

 propelled very easily to further removed places, for 

 instance Batavia, whereas at Anjer, close behind the ash 

 cloud, no sounds or only faint ones would be heard. 

 Cther explanations, such as by the interference of sound, 

 seem to me less probable though not entirely impos- 

 sible. 



Besides these sound vibrations, very long air waves 

 were formed during the explosions, which did not mani- 

 fest themselves by any sound, but had nevertheless an 

 important effect. The most rapid of these vibrations 

 communicated themselves to the buildings and walls of 

 rooms, so that objects which hung on the walls or from 

 the ceiling were set in motion. Thus at Batavia and at 

 Buitenzorg, a distance of 150 kilometres from Krakatoa, 

 the doors and windows began to rattle, clocks stood still, 

 ornaments on cabinets fell down, and hanging lamps were 



