While passing along the rift I noticed, projecting through the debris, some immense masses of obsidian and pitchstone that appeared to have ' boiled' up towards the conclusion of the eruption; therefore I resolved to return the same way and secure some specimens. I was fortunate enough to obtain here one piece of obsidian that throws some light upon the formation of the black pumice, it having apparently changed into that substance upon one side, where it had been subjected to intense heat.

Professor Johnstrup expresses an opinion that the whole of the pumice and ashes were ejected from the cone-shaped crater north of the rift, which he accordingly christened the Pimpstens Krater-i.e., Pumice Crater. I am, however, of a different opinion : I believe that from the rift throughout its whole length pumice, obsidian, pitchstone, and ashes were ejected as well as from the crater, the latter marking the termination of the rift, and being builded up into its present cone-shaped form towards the conclusion of the eruption, when the rift was blocked up with the debris of the side of the mountain. I think that this is proved by the presence of the blocks of obsidian and pitchstone scattered about among the pumice, for had those substances been ejected before the last named, they would have been buried under it, and if subsequently, they would have been resting on, not deeply inbedded in it.

It is recorded that the greatest eruption of pumice and ashes took place on the morning of the 29th March, 1875, the eruption being preceded by a sharp shock of earthquake; but according to the letter before alluded to, written by a resident in Reykjavik, and dated the 23rd of that month, a considerable eruption must have taken place prior to that date, for the letter says: ' Ashes, too, had fallen over the north-east coast, so that pasture fields were covered so that the farmers had to take their sheep into the huts and feed them.' Whenever the eruptions took place, it is certain that strong westerly winds must have prevailed, for but a comparatively small quantity of the pumice ejected fell in Askja west and north of the rift and crater, the bulk being borne away to the eastward and scattered over the country in that direction, vast quantities being carried out to sea, as mentioned in the prefatory chapter. Professor Johnstrup states that over 3,000 square miles of country east of the volcano, fortunately chiefly sandy wastes and barren moorland, were covered with pumice by this eruption, the minimum depth near the coast being about two inches, the depth gradually increasing as the volcano was approached. The pasture-land belonging- to several farms east of the Jokulsd was rendered useless for some years to come, and pecuniary assistance was sent from England to the farmers ruined in consequence.

Askja, it is believed, was visited but twice during the eruption-first in February, by Jon of Viftihcer, and later on, in July, by Mr. Watts. As the latter's account of what was taking place at the time of his visit throws some light upon the formation of the bed of the lake, I beg permission to quote a couple of paragraphs. Alluding to this abyss, which he erroneously, it would appear, terms the crater, Mr. Watts says (p. 86 ' Across the Vatna Jokull}) : ' Presently, apparently about a mile away to the north, we could see the rim of the crater, at a great depth beneath us (Mr. Watts was on the summit of the mountain south of Askja), and while we were looking at it, a great crack opened upon the margin, and a huge slice slipped with but little noise into the crater-deep, deep down beyond the range of vision From this it is reasonable to believe that the disrupted mass did not sink at once to its present level, but subsided slowly, possibly continuing to do so as long as lava streamed forth in the Myvatn's Orcefi. There can be no mistake as to this being the c crater ' of Mr. Watts, notwithstanding that it is oval, not ' triangular in shape,' for he mentions its circumference correctly- viz., five miles-and it is the only abyss in Askja of that magnitude; but with reference to its being a crater, Professor Johnstrup states, with the sentence omphasised in italics, ' that there was no evidence that the site of the lake had been a crater.'

The ' Pumice Crater' north of the rift at this time, it appears, did not emit steam, but smoke, for Mr. Watts says further on : 'A shaft, like the mouth of a huge coal-pit, was disclosed to the N.N.E. corner of the valley, but beyond the rim of the crater, from which a column of pitch-black vapour was issuing- boom ! boom ! from its hoarse black throat, was succeeded in a few seconds by a heavy shower of coarse, earthy granules. . . . Suddenly a fearful crash made us stand aghast; it seemed as if half the mountain had tumbled in upon the other side of this horrible valley . . . and huge wide cracks, even where we stood, showed us that our position was not altogether a safe one Under such circumstances exploration must have been next to impossible, and therefore the imperfect description of Askj'a, both in Mr. Watts' paper read before the Royal Geographical Society, and in his book, is not to be wondered at, though the strange hallucinations he labours under as to the size and situation of the volcano are, for on p. 189 of the book quoted he says ' Oskjugja can only be regarded as a lateral crater of the Vatna '! whereas it is distant therefrom at least twenty-five miles.

Respecting the lava deposits in Askja, bared by the subsidence now of the bed of the lake, I beg leave to translate an interesting paragraph from the paper by Professor Johnstrup before alluded to :-

' An excellent insight into the history of Askia's formation is here afforded, the vertical fractured surfaces showing what a multitude of lava-floods must have been deposited in Askia's cauldron shaped valley [kjedelfonnige Dal), one above the other. (Italics sic.) The divisions between these lava-floods are distinctly marked by the layers of red slag-like lava, which time after time has formed the surface of the underlying lava strata ; and I doubt very much if there can be found in any other place in Iceland, except the Almannagja, where, however, the formation is far from being so distinct, such an instructive and grand profile as this. It has more than ordinary interest, owing to the striking resemblance presented by the volcanic deposits here to those widely spread rock -formations (Bjergdannelser) of basalt and dolerite which have been pronounced by most geologists to be of plutonic origin (plutonisk Oprindelse). Had they had an opportunity of viewing this profile, they would certainly have entertained a different opinion.'