strike up a small valley between the western lateral slopes of the moraine of the glacier and the Mount Cook Range on our left, and for a distance of about seven or eight miles force our way through dense scrub and loose boulders from the moraine and mountain slopes, to the junction of the Ball Glacier with the Tasman. This Ball Glacier comes from the Great Southern arête of Aorangi, and is fed almost entirely by avalanches, there being no snow-fields—or névés as they are called in Alpine parlance—of any great extent at its head.
From this point upwards we strike out on to the ice on our right, and another seven miles or so brings us to a further division of the valley, Mount de la Bêche being the dividing peak. The glacier of the left-hand or northern branch is known as the Rudolf Glacier, whilst the main body of the Tasman stretches some six miles further north-eastwards to the Hochstetter Dome, where it again divides. The saddle at the head of the left-hand branch, again, has been reached by Dr. von Lendenfeld and by myself in our respective ascents of the Hochstetter Dome, and commands a superb view of the Whymper Glacier and valley, and of the Wataroa River on the west coast. The head of the branch to the right of the Hochstetter Dome has not yet been reached by man.
Taking a retrospective glance again at the peaks on either hand, and commencing at the lower end of the glacier, we have first on our right the Liebig Range till opposite the Ball Glacier, when the embouchure of the Murchison Valley occurs, followed by the Malte Brun Range, with the main peak—the Matterhorn of New
