' But he whose weary step hath traced Iceland's mysterious awful waste,

Whose eye the OdaSahraun hath viewed Can tell thee what is solitude!

Mindful of the past, the timid swain Nurtured near the dread domain,

Hath peopled the caverns of the Hraun With lawless sons from outlaws born ! '

(Adapted from Mrs. Hemans.)

EARLY the next morning we set out for Svartdhot (Black-river-cot), a farm situated on the border of the Oddftahraun, whose owner acted as my guide to Askja in 1878. The cot stands close by the eastern shore of the Svartdvand (Black-river-lake) on the northern bank of the river issuing therefrom. Both river and lake are well stocked with char and trout.

At Ami's request we went a little out of our way to his home Skutustaftir (Skutu's-stead), a parsonage attached to a church at the south end of Myvatn. Here, as no less than five couples were to be married, we somewhat reluctantly consented to stay several hours that Ami, who was more or less closely related to at least half the parties, might witness the ceremony. Space cannot be spared to describe the weddings, or the proceedings after, much as I should like to do so, as I must c speed the pen,' otherwise, reader, you and I will never, figuratively, reach the volcano amidst the Misdeed-lava-desert.

It was six in the evening before we got away, so to make up for lost time, Arni essayed a ' short cut' which resulted in a flounder through a bog for nearly three hours, and, as we contrived afterwards to lose our way on the moorlands west of the mountains south of Myvatn, it was five o'clock the next morning before we reached our destination, which is distant but six hours' ride from Sfcutustafiir by the regular horse track.

We went to bed immediately upon our arrival, and slept until after midday. Four hours later saw us again in the saddle and, with Einar the farmer as guide, on our way to Ashja. We took nothing with us but food for two days, and a spare pony laden with two large knitted woollen sacks, crammed full of hay as fodder for our steeds.

An hour's ride across level moorland, alive with willow grouse (Lagopus subalpina. Nils.), brought us to a considerable river named the Suftrd-South river (?)-a tributary to the Shivering-flood. We had followed a dry watercourse, formerly the channel of a small stream flowing from a tarn. I was told by Einar that for some years from fifteen to twenty of his sheep were drowned annually in this stream, the unfortunate brutes, it appears, when they came to drink, frequently slipped in, and were unable to get out, the sandy banks being concave, where they had been hollowed out by the running water, with the matted roots of the grass and bushes growing above overhanging considerably. One year-1872 I think- Einar lost seventy ! This, he at last considered, was too high a tax to pay; so he finally did what ought to have been done years before, as it was only a matter of a few hours' work, dammed the stream at its outlet from the tarn. What was somewhat extraordinary, the water in the tarn after rising considerably above its former level, suddenly fell to this again, its overflow having found a subterranean outlet. I merely mention this incident to show what a long-suffering, patient, and apathetic being, an Icelandic farmer is; and yet Einar is an exceptionally enterprising one.

As the Sufird flowed from the south-east, the direction in which we wished to proceed, we followed its course for about four miles, passing through an extensive sandy waste-not lava-desert as shown on Gunnlaugsson's map-bordering the Oddtiahraun on the north-west, named after the river, the Sufirasandr. The river, I discovered, was fed with the surface water from the Oddftahraun in a somewhat remarkable manner, beautifully clear springs bubbling briskly forth in innumerable shallow pools in the sand. The phenomenon is easily accounted for. The surface lava in the desert lying at a higher level than the Siiftrdsandr, the melted snows and rains percolating through for centuries have worn channels in the porous lava and through these the surface water now flows until it reaches the sand, where it rises as we have seen.

We quitted the river at a still more remarkable pool, one in which the order of things was reversed, the water being in rotary motion around a small hole at the bottom, a few inches in diameter, through which it was descending to some underground channel. Here the ponies were allowed to drink to their hearts' content, their riders likewise imbibing as much as possible, as in a few minutes we should find ourselves amidst the lava where no water would be met with for at least six hours.

A brief description of the Oddftahraun, which we were now about to cross, will doubtless prove interesting. Its area, as before stated, exceeds 1,100 square miles; and it forms part of a desert region lying between the Jdkulsd, draining the Vatna, on the east and the Sfcjdlfandafljot on the west, stretching south from the range of mountains encircling Mjvatn on the south and east right to the glaciers of the Vatna Jdkull, which extend to the coast; a black, fire-blasted, scarcely passable wilderness of lava and volcanic sand, at least two thousand square miles in extent. The surface lava of the Oddftahraun has an altitude of about 1,500 feet, and it consists chiefly of countless lava-floods, varying greatly in age, some being thousands of years old and clothed with lichen, while others are as black and new-looking as those which flowed from the mountains east of Myvatn a century and a-half ago. The newer lava-floods in the vicinity of the Dyngjufjoll on the north have flowed from rifts that have been time after time torn in Ashja's encircling mountain-wall, and at its base; while the south-western portion of the desert, according to Mr. Watts, is covered with those which have flowed from the Trolladyngja (Trolls-bower), a volcano lying about fifteen miles distant from Askja in that direction. Mr. Watts is the only man, I believe, save his guides, that has visited this mountain of late years, or crossed the desert between it and the Dyngjujjoll, and he says, p. 98 of 'Across the Vatna Jokull.' at the summit he 1 found a small but perfectly formed crater, about 500 yards in circumference, but of no great depth, while in the centre rose a ridge of burnt lava, which gave the mountain the black tufted appearance I had noticed in the distance.' This crater-'bower' therefore is an utterly insignificant one compared with the ' bower3 within the monarch of the Dyngjujjoll. Mr. Watts furthermore says the Trolladyngja is ' nothing but a huge mound of basaltic lava, partially covered with snow, rising by a very gradual slope to about 4,000 feet above sea level.' I may as well here observe that notwithstanding this mountain is named on Gunnlaugsson's map Skjaldbreift efia (or) Trolladyngja, and Mr. Watts and other writers invariably designate it by the first name, the Icelanders, or at least the people living around Myvatn, do not know it by that appellation, but always call it Trolladyngja; therefore I have done likewise, it being less likely to cause confusion, there being another Skjaldbreift, near pingvellir.