408 Transactions.— Geology. 
indicate. When the head of Wakatipu Lake was raised that height and 
the lower end a corresponding height, the lake must have contained a 
glacier. At any rate, it did once contain one, as the Kingston moraine 
testifies. Keeping in mind that the lake country subsided not from the 
central chain as an axis, but from the sea coast as from a hinge ;—that in 
other words the lake did not rise at the lower end and sink at the upper, but 
that the whole of the lake sank, more of course at the head and less at the 
foot ;—we can see clearly how the deep part in the middle came to be. 
This would simply be as follows:—The valley in which the lake lies, and 
which is about 80 miles long and narrow for its length, was the bed of a 
glacier. A glacier moves like a river and would doubtless also work down 
the valley in a parabolic curve (as our president has demonstrated in a 
‘paper printed in vol. vi. of the “‘ Transactions” that the principal rivers here 
do), the shoulder of the curve being towards the upper end of the valley. 
Now the gradual lowering in height of the glacier (which I am assuming 
stopped short at Kingston) would have the effect of causing it to melt away 
at the lower end, and as the foot receded from the moraine, it would con- 
tinue to deposit debris and make the valley at that part (already stopped up 
by the moraine) to decline gradually backwards. The future lake would be 
shallower at that end—more or less as the wearing away of the glacier was 
quicker or slower. At the same time the sinking of the curve higher up 
would tend to make the slope up to the moraine steeper. Whilst this was 
going on, the glacier would slowly melt and in course of time would dis- 
appear, water would take its place and the lake (at a much higher level, 
however, than at present) would come into existence, formed partly by the 
melted ice of the glacier, and partly by summer rains and melted snow off 
the ranges. The two rivers which run in at the upper end would gradually 
deposit a delta (the present Dart Hundred) and the ultimate effect of all 
these proceedings would be to make the lake deepest at the middle. 
From the foregoing considerations, it seems doubtful that the Kingston 
Valley has ever been the outlet of Lake Wakatipu. It was of the glacier but 
not of the lake, which, if I am right in theory, must, as I have said, have 
originated by the sinking of the upper part of the valley below the level of 
the moraine. Most probably the lake always found an outlet, as at pr esent. 
In conclusion, I would remark that I am well aware I have sugge 
nothing new. But my object in writing this short paper is to excite an 
interest in a subject (the mode of formation of the large southern lakes) 
which, as these lakes lie it may be said at our very doors, it naturally falls 
to the Sonthiend Institute to a 
