one day, and it is well worth an effort to effect this. But by following a line which we carefully noted, a road may easily be formed by the use of the very impediments which now obstruct the track.

Being fearful that our stock of provisions would not hold out whilst we made the road passable for the mules, I turned them loose, making a cache of the packsaddles, and all surplus, stores, personal clothing, &c.; everything, indeed, that could be spared to relieve the men. As we had carefully barred the track at the head of the valley, and also at the Fish Stream, I calculated on finding the animals on our return.

Crossing a branch of the river at the entrance to the gorges, we travelled down the bed on the left side about half-a-mile. Then turning up a dry stony gully, we turned into the forest, following Dr Haast’s track (which is well defined), for about three miles. (Camp 9). Here, that track crosses the river to the right bank, immediately below the junction of a small creek; but on examining the country along the left bank, I found a very fair line of route along that side, and continued to a point opposite the junction of the Wills, where Haast’s track again crosses the river. I observed that the Wills was then very low, and quite fordable on the sand-bar at its mouth; and on our return, it was even lower. (Camp 10).

As the rain was incessant, and we were unable to find a piece of level ground of sufficient extent to pitch our tent, we rested (if the word is not misused) under the projecting ledges of a huge mass of rock which had travelled from its original bed to within three feet of the river’s edge. Here unfortunately I caught a chill in my foot from lying in a pool of water which collected during the night. The consequence was, that I suffered for three days from an acute attack of gout, under which I could only travel with difficulty and in great pain, carrying a heavy swag over the roughest possible ground. We, therefore, progressed very slowly.

Making another small cache of biscuit, we again followed Haast’s track to the confluence of the Burke river, avoiding, however, several unnecessary detours. At one point this track leads out at a saddle over a dangerous precipice, which may be avoided by keeping a little to the right. I remarked some fern trees growing just below the top of this precipice at an altitude of not less than 1200 feet above the sea-level.

Beyond and immediately opposite the Mount Bealey range (which terminates abruptly in a vertical cliff, rising perpendicularly about 3000 feet above the river, its face scarred by constant landslips and thinly clothed with patches of scrub), the ground descends to the river in a steep incline about half a mile in width. This is covered with belts of fuschia and currant bushes, between which are streams (so to speak) of rock and shingle—the debris of the blue slate rock, above, which is bare of vegetation, except on the summit, where it is fringed with timber, and rises perpendicularly with loose pendant masses, awaiting their time to crumble down and sweep all before them. Cutting through the scrub we crossed this immense landslip, and ascended the mountain again, descending rapidly to the river. This ascent maybe avoided, and a much easier track found by keeping nearer the side of the channel.

The gorges terminate about two miles above the junction of the Burke. The mountains recede, and the river flows more gently over a wider bed. We emerged from the forest, near a very remarkable water-fall, which shoots straight out from the rock, at a height of 250 feet, and falls over a vertical precipice.

The Burke is a deep and sluggish river, from which I conjecture that it passes through more level ground. Indeed, the general appearance of the country beyond this point is such as to induce the belief that a considerable extent of available land lies behind the ranges flanking the valley of the Haast, and at the rear of Mount Aspiring, the northern face of which, of all the mountains, presents comparatively little snow. And it is on that side of the chain only that a practicable pass to Jackson’s River probably exists.

Fording the Haast to the right bank above the junction of the Burke I noticed the remains of an old camp, and, carved on a tree, “Nugget Prospecting Party, Sep. 8, 1863,” with the initials of the members of the party. (Camp 12.)

Below the junction the Haast is about fifty yards wide, and the valley opens out, affording a magnificent view of the Gray and Moorhouse ranges. Here the river runs for nearly eight miles between wide flats, some of which are partially grassed, but for the greater part covered with dense scrub and timber. To avoid the sinuosities of the stream, we recrossed to the left bank, immediately below the Burke, where the river is divided by shingle spits into several channels, of which we took advantage. These spits, as we found on our return, shift their position with every flood, and are intersected at the upper end by dangerous quicksands. From this point we travelled without difficulty to the junction of the Clarke river, above which we again forded the Haast at a broad reach, and subsequently crossed the Clarke to Camp 19. Notwithstanding so much rain had recently fallen, I found the Clarke much lower than I had anticipated; running near the mouth in two broad channels, with a rapid current, between shingle beds extending fully a mile and a half from bank to bank. Some of our party crossed here safely, the second channel which is the most formidable, having about three feet of water; and the others crossed higher up over four channels in shallower water.

Before joining the Haast, the Clarke runs for about a mile between two open grass flats, containing probably 1500 acres. These, however, would seem to be liable to occasional inundations. I travelled up the valley to a stream which enters from the North, and found floodmarks on its banks nine feet above the ordinary level.

Heavy rains detained us here for a day and a half. In the afternoon of the second day it cleared up, and we again set forward, first making a cache of eight days’ provisions, with a portion of which I intended, on my return, to explore the Gray Ranges for a pass, which, I am of opinion, exists in that direction—leading out probably somewhere near Bruce Bay, from which the Clarke is distant about 25 miles in a direct line.

From the Clarke to the coast, the Haast has a bed seldom less than a mile, and sometimes one-and-a-half and two miles in width. Through this the river ordinarily flows in two or more channels—occasionally uniting and again diverging from each other. The intervening spaces are covered with loose shingle drift, interspersed with insulated patches of soil, which are of greater or less extent, and for the most part overgrown with low scrub. These are apparently the remains of the original valley, which has been torn up and destroyed by heavy floods—the vestiges of which are everywhere visible. These floods render the course of the river liable to constant changes. On our return, we found shingle where, on our down track, the water was flowing, and deep water where shingle spit had been. It is impossible, therefore, to lay down positive directions for following this part of our route. The only course is, to keep to the shingle (which affords capital travelling), wherever possible—taking to the bush when the river sets in against the banks. I may here, also, caution travellers against the “back-waters.” The shingle-beds are generally very wide, and often several miles in length; but it is not always possible to follow them to their termination. The streams, which issue from the mountains, form small water-courses inside the shingle, and close beneath the bush. These become very wide and deep at their confluence with the river, where the water backs up the entrance, so that the shingle spits often end in quicksands and impassable channels. It is necessary, therefore, to travel along the inner margins, and to cross the backwaters so soon as they begin to deepen. When proceeding up the river this precaution is unnecessary, and the rule is, to ford the backwater and cross to the spit immediately it is practicable to do so.

At the junction of the Clarke, the united rivers form a single broad channel, which when we passed down was 450 feet wide, being somewhat swollen by the rains. We therefore followed along the right bank, taking advantage of the shingle beaches when we could, but being frequently obliged to make tedious detours through the bush, and often climbing steep rocky spurs, to avoid the backwaters and the river itself. About twelve miles below the Clarke, we crossed a roaring mountain torrent, (camp 15) not without difficulty, and even danger, as it was then in a state of fresh of which we were not at the time aware. Five miles further on, the current sets close under a spur, consisting of huge rocks rising precipitously to a height of 300 feet. To avoid this we forded the river to a shingle spit in the centre, again returning to the bank below the rocks. Here, (near camp 16) about sixteen miles from the coast, the cabbage-palm first made its appearance. Two miles above the Thomas River, perpendicular cliffs again obstructed our progress; and being by this time convinced that a road along the right bank was quite impracticable, I resolved to ford the Haast to the other side. We effected the crossing at a point where the river was about a quarter of a mile in width, and about three feet six inches deep, with a slight bar in the centre, which afforded a resting point, the current running very rapidly and with great force, so that one of the men who had not taken hand with the rest of the party (a plan which we adopted for mutual support) lost his footing and was nearly swept away.

We continued on the left bank with far less difficulty than we had encountered on the right. Not only were the beaches longer and more continuous, but the rocky