352 



KNOWLEDGE. 



September, 1911. 



There are five conceivable methods whereby heat may be 

 transferred from the magma to the vent : — 



i. Explosive removal of material from the upper part of 

 the vent, followed by uprise of magma. 



ii. Simple outflow of magma at the crater-lip. 



iii. Thermal convection in the la\a cohimn. 



iv. ■■ Two-phase convection." 



\'. The uprise of super-heated juvenile gas through the 

 lava column. 



.\t Kilauea, on the observation of which this theory is partlv 

 based, the first and second methods are inoperative ; the third 

 is known to be inefficient in heat-transfer ; but the fourth, a 

 process conceived to be due to the uprise of comparatively- 

 light, hot, gas-permeated magma in the lava-column, and its 

 return along the margins of the vent as comparatively heavy, 

 cooled, gas-free magma. — may bring a considerable quantitv 

 of heat to the surface. This process is believed to be the 

 cause of the lava-currents in the crater of Kilauea. The fifth 

 method may be eflicient as a heat-bringer at some vents, but 

 not at Kilauea. 



A largely overlooked source of heat, according to Daly, is 

 that set free by chemical reaction between the constituents of 

 the hot and active magmatic fluids. Many striking figures are 

 given illustrating the great amount of heat evolved by the 

 reactions between the common elements of these fluids. 



The process outlined above is known as the " gas-flu.\ing 

 hypothesis," and is considered analogous to that of a gas- 

 blowpipe. 



Some explosive types, such as the Ries cauldron and 

 Bandaisan, are considered due to the contact of hot magmatic 

 material with vadose waters circulating within the rocks. The 

 explosions here are non-volcanic : but there may be all grada- 

 tions between this type, and those, such as the Hawaian 

 volcanos, characterised by quiet magmatic extrusion without 

 explosion. In ordinary volcanoes, there are great variations 

 in the proportions of magmatic and vadose fluids involved, 

 and consequently great variety in modes of eruption. 



With regard to the now much-discussed r61e of steam in 

 volcanic action, Daly says : " Though the rise of hot magma 

 into rocks charged with vadose or connate water does often 

 cause explosion, the steam pressure produced by such vola- 

 tilised water can no more be regarded as the cause of vulcanism 

 than is the boiling of a kettle the cause of the heat in the 

 stove." 



.4. BRUX. — Whilst Daly emphasises the adventitious nature 

 of the intervention of water in volcanic action, it is to Brun. 

 of Geneva, that we are indebted for what appears to be the 

 overthrow of the old axiom that paroxysmal eruption is due to 

 the explosive violence of steam. In the Geological Magazine 

 for June and July, Mr. E. B. Bailey, of the Geological Survey, 

 gives an illuminating review of the new book, " Recherches sur 

 r Exhalaison Volcanique " by this original and courageous 

 worker. One of the most valuable features of Brun's work is 

 the mass of new and exact experimental data he has accumu- 

 lated in respect to vulcanicity. Not only has he measured the 

 temperature constants of many minerals found in lavas, but 

 new work has been done in the methods of collection, extrac- 

 tion, and analysis of the various volcanic gases, both in the 

 field and in the laboratory. Experimental work on the 

 behaviour of rocks during heating has resulted in their classifi- 

 cation as " active " or " dead." Active rocks, typified by 

 recent lavas, expand and liberate gas at such a rate when 

 heated that the molten material fumes over the edge of the 

 crucible, like a miniature lava flow. Active acid rocks are 

 more violent than basic and give rise to veritable explosions. 

 Dead rocks, among which are schist, granite and gabbro, give 

 off gas during heating, but at a higher temperature melt quietly 

 without much expansion or violence. The temperature at 

 which gas is emitted in active rocks so rapidly as to cause 

 sudden expansion and explosion is called the " explosion 

 temperature." The maximum temperature possible at a 

 volcano is fi.xed by the explosion temperature of its magma. 



The principal gases liberated at tne explosion temperature 

 are chlorine, hydrochloric acid, and oxides of carbon ; the 

 solids evoKed are chlorides of the alkali metals and 

 ammonium ; sulphur occurs but is always in small quantity. 

 Such water as is contained in ttie rock is always given ofl' 

 before the explosion temperature is reached. These con- 

 stituents, with the exception of carbon monoxide, are the 

 same as those actually emitted at volcanoes. 



Brun's main thesis is that paroxysmal eruptions are an- 

 hydrous, and that the aqueous character of fumaroles is due 

 to the contact of volcanic heat with superficial waters. This 

 view is supported by many striking observations. It is shown, 

 for example, that near the crater of Vesuvius, the ashes fall 

 quite dry. but are extremely hygroscopic, owing to the presence 

 of chlorides of iron and magnesium. Morever. the ash which 

 falls is white ; whereas if it had been exposed to the action of 

 water vapour at a high temperature, it would redden immedi- 

 ately, owing to its content of ferrous chloride. In the crater 

 itself, such deliquescing salts as Fe-j CI,;, Fe Clj, Mg CU and 

 A\, CI,-., may be collected dry and undecomposed, whilst hot 

 water-vapour would immediately reduce them to oxides. 



Further evidence is adduced from the study of the white 

 clouds which hang over volcanoes. These are generally- 

 regarded as water-vapour ; but Brun shows that they are 

 persistent, and insoluble in the atmosphere as they drift away 

 from the volcanic focus, and are therefore composed of solid 

 particles. .At Kilauea. Brun took a series of dew point read- 

 ings in the great white cloud as it drifted across the crater-lip. 

 His results show in every case a lower dewpoint for the air 

 within than for the air outside the cloud. The lowered dew- 

 point is believed to be caused by the dilution of the air with 

 anhydrous gases carrying hygroscopic solids in suspension. 

 On the contrary, a markedly elevated dewpoint was obtained 

 at the peripheral fumaroles, as was, indeed, to be expected. 



It seems, therefore, that Brun has, at least, established the 

 anhydricity of volcanic exhalations ; and gre.at probability 

 attaches to his view that water is not the agent to which 

 paroxysmal eruptions must be attributed. 



METEOROLOGY. 



By John A. Curtis. F.K.Met.Soc. 



The weather of the week ended July 22nd. as summarised 

 in the Weekly Weather Reports of the Meteorological Office, 

 was very fine over the greater part of England, but slight rain 

 fell in the North and West and in Ireland. In Scotland there 

 was frequent rain, though it was seldom heavy. Thunder- 

 storms were reported during the middle of the week. 



Temperature was again above the average, except in 

 Scotland, where it was either normal or very slightly below. 

 Very high maxima were recorded in the South and East of 

 England and in the Midlands, the highest being 96° at 

 Greenwich, on the 22nd, with 94" at Margate and 93° at 

 Raunds. In Ireland the highest reading reported was 79°, 

 while in Scotland North and West the temperature did not 

 rise above 71°. .-At individual stations in Scotland and Ireland 

 the maximum did not reach 70°, and at Balta Sound, Shetlands, 

 it was only 59°. 



The minima were above 40" everywhere except in Scotland, 

 where, at Balmoral, the temperature in the screen fell to 32°, 

 and on the grass to 30'. In the English Channel the minimum 

 was 55". 



Rainfall was below the average except in Scotland N. and 

 E. The defect was very great in places, and the drought began 

 to be severely felt. In England S.E. the week was rainless. 



In Scotland, however, there was much rain, the total for the 

 week exceeding 2-0 inches atGlencarron and Fort \\'illiam. 



Bright sunshine varied a good deal. In most parts of 

 England it was again above the average, but in Ireland, 

 Scotland and the Northern parts of England it was in defect. 

 In Ireland N. the average dailv duration was less than three 



