March 1, 1889.] 



♦ KNOWL.EDGE ♦ 



99 



lava through the escape of liquid matter was unequalled : 

 an area of 12,000 by 3,01)0 feet sank down nearly 400 feet. 

 The floor rises and sinks in a remarkably level manner, 

 and it is highly probable that it is supported by the column 

 of liquid lava beneath. When a discharge has taken place 

 the floor falls ; then slowly rises to or above its former level, 

 continuing to rise till another eruption occurs. There is 

 scarcely any visible outflow of lava from Kilauea, in striking 

 contrast with the immense discharge from Mount Loa. 



Volcanic action in a crater where there are liquid lavas is 

 to a great extent a kind of boiling process, and the confined 

 vapours of the enlarging bubbles forcing an escape, are the 

 chief projectile agencies through which the more teriific 

 explosive phenomena are produced, as well as the milder 

 ones cf the Kihxuea type. Increased activity means an aug- 

 mented and more rapid escape of vapours through the viscid 

 lavas, and also greater heat, and extension by fusion of the 

 liquid lavas below the surface. The cinder cones made 

 about the vents in quieter times may be destroyed or swal- 

 lowed up by the great projectile eruptions vphen the force 

 from the rising vapours becomes greater than the mountain 

 can bear. Its sides then break, the liquid lavas pour from 

 the fissures, and the top of the cone frequently sinks into 

 the abyss. 



It is not our purpose to enter on the difficult question of 

 the ultimate origin of the volcanic fires. It may be sufficient 

 to mention the older theory of the original igneous fusion 

 of the earth (still accepted by many geologists) ; and Mr. 

 Mallet's hypothesis that the heat his been produced by 

 lateral pressure, caused by shrinking of the earth's ci'ust.* 

 Whatever may be the cause of the lieat, it is known that a 

 mass at a high temperature may be kept solid by great 

 pressure. When the pressure is relieved — either by slow 

 denudation or by the opening of fissures — it will become 

 liquid, and may rise to the surface. But this rise would be 

 comparatively glow, and would certainly not cause eruptive 

 phenomena, which it is almost certain are due to the agency 

 of water, as was urged by G Poulett Scrope as far back as 

 1825.| The mode of its action has been recently discussed 

 in an elaborate memoir by Professor J. Prestwich J 



The slowness with which lava rises from deep-seated 

 action alone is well shown in the case of Kilauea, where 

 the epochs of high lava-level are separated by years. Three 

 great eruptions in Kilauea were eight years apart, and in 

 the intei'val the lift (as shown by the movement of the floor 

 of the crater) was only 400 feet. 



The sea is frequently a.ssumed to be the source of the 

 water of volcanoes, especially as they are commonly situated 

 either in oceanic islands or near the seashore. We must 

 remember, however, that this nearness to the ocean is pro- 

 bably in great part a result of the way in which rocks which 

 undergo compression, so as to i-ise in island chains and 

 ranges of mountains, are constantly developing new fractures 

 ])arallel to their direction of upheaval. Professor Dana has 

 pointed out that borings near the seashore generally yield 

 ''resh water, and that in Ilaw.aii eruptions appear somewhat 

 more frequent in the wet sea.son. It seems probable that in 

 this case the explosive force is caused by fresh water from 

 the melting snows and very frequent rains, especially as the 

 mountains are extremely cavernous and full of fissures. In 

 the lavas of Vesuvius, however, there are salts which are 

 probably of marine origin. The far greater eruptive force 

 shown by Mount Loa than by Kilauea appears to result 

 from the greater amount of snow upon the summit, as the 

 characters of the lava are identical. 



* " Roy. See. Phil. Trans.." vol. cxiii. 

 t "Considerations on Volcanoes," 1825. 

 ^ " Proc. Rny. Soo," vol. xli. 



In what manner does water act in producing eruptive 

 discharges 1 In all probability by being introduced into the 

 conduit of slowly rising lava. The probable processes are 

 thus briefly summarised by Profe.ssor Dana : (1) Absorption 

 of subterranean moisture from the sides of the conduit. 

 (2) A rise of the lavas thus supplied. (3) After reaching a 

 level where the pressure is sufficiently diminished, great 

 vesiculation of the lava from the expansive force of the 

 gases into which the water is resolved. (4) Further union 

 of the gas particle,? into bubbles (where the vapours are 

 sufficiently abundant) exerting the greater expansive force 

 by which explosive results are produced. The great vesi- 

 culation which occurs is shown in the following figure of 

 the lace-like scoria from Kilauea. The liquid lava when 

 ejected is also drawn into capillary glass (called " Pele' 



Knlarcjkd h'Kii'Kr.s oF ScoitiA Cells and "Pelk's Hair" 

 The latter contain microscopic crystals and elongated air-bubbles. 



hair" by the natives), which is often formed in great 

 abundance. 



On comparing these phenomena with those shown by 

 other volcanoes there is found a great general similarity. 

 The volcanic processes in Vesuvius, for instance, are in the 

 main as in Kilauea: (1) filling; (2) discharging ; (3) col- 

 lapsing — alternating in its conditions between a volcano 

 with a deep crater, and one with a top plain. The crater- 

 pit of Vesuvius is sometimes 100 feet deep; thLs .slowly 

 tills by small injections within it of lava and cinders. 

 This process generally goes on (witli ocavsional outflows 

 from fissures in the sides of the mountain) till the cavity is 

 full, and a top plain of solidified lavas is formed, a mile or 

 more in circuit, and nothing is left of its crater but the vent 

 of a cinder cone. Smaller eruptions may form a large cone 

 or cones till a grand outburst takes place ; lava fi'agmeuts 

 are thrown upwards to great heights, and lava pours frour 

 opening tissur'es, till the great cone collapses and falls, as rir 

 1S06, and still more in 1822. In July, 1834, the summit 

 of Vesuvius was a plain, with a small cinder cone, but at 

 the great eruption in August of that yciir the plain had 

 disap[)eared, leaving a deep crater. This took place ag.ain 

 in 1839 and 1850. 



The most striking difference between Vesuvius and the 

 llawtiiian volcanoes is the ioimensely greater eruptive and 

 projectile force in the former. The Tanxwera erirption in 

 New Zealand irr 1886, and that of Krakatoa in 1883, are 

 examples of eruptions of the more explosive kind. The 

 destructive outbr'eak of Tarawer-a lasted only six hours, 

 but its violence wjis so gr-eat that du.st was ejected to a 

 height of more than 44,000 feet, and an area more than 

 fifty miles across was buried beneath volcjinic mud. At 

 Krakatoa the discharge was equally sudden and still more 

 terrible and destructive. Dust was thrown to a height of 

 probably 50,000 feet at least. 



What is the cau.se of this great drflerence in eruptive 

 force '? Professor Dana considers that in Krakatoa and 

 Taiawera the water- n)ust have jrained access to the interior 



