O'Eeilly — On Gaseous Products of Great Eruptions. 23 



at 317 cubic kilometres. This computation, however, rests on hut 

 few data, so that, in my opinion, a quantity of 150 to 200 cubic 

 kilometres will come nearer the truth." 



The quantity thus estimated for the Krakatoa eruption is 

 evidently but a part of the total emission of solid matter : what 

 proportion this heavier part of the ejected ash bears to the finely- 

 comminuted matter, to the presence of which in the atmosphere the 

 continuously recurring glows have been attributed, it would be 

 impossible to say. How much more must be allowed for the still 

 finer matter, which continues suspended at very great altitudes, 

 which evidently encircles the earth, and to the presence of which 

 is attributed the faintly coppery haze visible round the sun's 

 image ever since the eruption, it is still less possible to estimate ; 

 but we can with safety say, that the quantity of vapour and gases 

 emitted must have been in some degree proportional to the total 

 quantity of ejected matter. We are further justified in assuming 

 that the quantity of gas and vapour brought into action was not 

 the minimum strictly necessary to project the totality of this 

 solid matter into space ; therefore any estimate of the quantities 

 of these gases that may be attempted from the data accessible 

 can only be much beneath the truth : indeed this is precisely one 

 of those cases where, wanting any term of comparison, the mind 

 is simply unable to exaggerate, even were there the will to do 

 so. 



Considering, therefore, only the portion of the ashes the 

 volume of which has been estimated, and the data as to the 

 height to which they attained, it is possible to arrive at a term 

 of comparison for the quantity of gases emitted by comparing 

 with the results produced by the use of gunpowder or other 

 explosives. 



In Berthelot's remarkable work, Sur la force de la Poudre, 

 1872, there is a table at p. 190, wherein for each explosive 

 examined by him he gives the amount in volume of gases gene- 

 rated per kg. of consumed explosive matter, and the temperature 

 in calorics attained. By the aid of this table an approximate 

 value for the gases having acted explosively in the case of the 

 Krakatoa eruption can be attempted. Let us consider in the 

 first place the work done in the case of the discharge of a 100 

 ton gun, for which I find in Nature, vol. xxviii. p. 385, the 



