The development of the present land forms presents a complex problem which requires detailed analysis and consideration of alternative explanations. To clarify the statement of the various lines of evidence, the following general conclusions are given in advance:—
1. The whole Fiordland región was reduced to a peneplain surface either in Cretaceous times, as usually held, or, as now seems more probable, during the later part of the Tertiary period. In its present elevated position the surface then formed is hereinafter termed the Fiordland Peneplain.
2. Before the peneplanation, if it occurred in Tertiary times, but after it if in the Cretaceous, the region was raised early in the Tertiary period with a marginal seaward downwarp. Dissection and marine erosion took place, followed by subsidence, resulting in the deposition of coarse arkosic sediments succeeded by finer and more or less calcareous beds.
3. Renewed elevation was accompanied by marginal corrugation along north-west-south-east axes and development of north-east-south-west undulations. The synclinal axes of the two series cross one another at the sites of the present Preservation and Chalky Inlets.
4. Consequent drainage systems developed in the north-east-south-west synclines, and eroding headward into the rising Fiord-land region tended to become adjusted to the structures therein. The controlling structures were the faults, shatter-zones, or joints in the slates, schists, and plutonic rocks. On the hypothesis of the late Tertiary origin of the Fiordland Peneplain, the major planation occurred at this stage and was followed by further uplift. On the hypothesis of the Cretaceous age of the Fiordland Peneplain, the uplift and dissection of the region after the deposition and folding of the Tertiary sediments was a continuous process.
5. While dissection of the uplifted peneplain progressed in the headward portions of the main valleys, the intervening coastward slopes of the peneplain were considerably reduced, with the production of a relatively low and undulating strip of land extending ten miles or more back from the shore, a foothill region in front of the residual promontories of the peneplain. Slow and moderate subsidence then followed, permitting the planing of this lowland region by wave-action during the transgression of the sea over it. Accompanying this there was necessarily much deposition of detritus in the submerged valleys. Such subsidence continued until the original surface of the peneplain in the region discussed in this paper had been lowered to within about 2000 feet of the sea-level, and remained in that position long enough to permit the destruction and disappearance of nearly all of the sea-cliffs cut by the advancing sea, while the main valleys became matured for a long distance inland.
6. Then a movement of elevation commenced which raised the inner margin of this wave-planed strip to a height of about 1500 feet above the sea. It may indeed have been a gentle arching, for the surface of the plateau is convex east of Puysegur Point, though this is possibly better explained by decrease in the rate of submergence towards the end of the previous phase of movement, and, consequently, the more nearly horizontal extension of the wave-planed surface, which is visible in some though not in all parts of it. The smoothed raised surface thus laid bare will hereinafter be termed the “Coastal Plateau.” It is perhaps possible that its uplift was accompanied or followed by a revival of old lines of fracture and the subsidence of certain crust-blocks, which may have taken a part in the formation of the present broad depressions in the coastal plateau, namely Preservation Inlet, with Otago's Retreat, and the Western and Eastern parts of Chalky Inlet, with Southport. If that were the case, the two larger islands and Gulches Peninsula remained as horsts between these depressions. It does not appear clear, however, that such differential crust-movements occurred.
7. As the uplift began, streams leaving the margin of the Fiord-land Peneplain extended their courses over the emerging coastal plateau. The major streams re-occupied their earlier recently-infilled valleys, from which they speedily removed the detritus, thus bringing about a rejuvenation that finally led to a perfecting of earlier adjustment to structure, which may perhaps have been favoured by the accompanying revival of movement along the early lines of fracture. The intervening minor streams entrenched their consequent extended courses in deep canyons across the plateau, discharging into the open sea.
8. At the commencement of the Pleistocene period glaciers moved down the main valleys, overwhelmed the islands in the inlets, and overflowed on to the coastal plateau on which there were deposited sheets of morainic and outwashed material. It is difficult to estimate how much of the lowering of the land surface should be assigned to preglacial erosion and how much to glaciation or to crustal movement. At all events, at the retreat of the ice the removal of detritus from the valleys was completed and benches along their sides were much modified. Further, the Sounds had a depth comparable with their present depth, whilst masses of moraine lay plastered on the sides of the inlets and probably also formed much of the thresholds at their mouths, and together with some outwash material covered portions of the coastal plateau.
9. Post-glacial activities have been:—
(a) A very small amount of stream erosion within the Sounds and much dissection of the morainic sheet on the coastal plateau. Delta-accumulation is well displayed in some of the Sounds.
(b) Cliff-recession under wave attack with the transport of detritus and morainic material and further growth of the thresholds, building of spits, tying of islands, and levelling of the sea-floor.
(c) Depression of the coast to some extent.
(d) A rather smaller elevation with brief still-stand intervals about fifteen and forty feet above the present sea-level. It is not possible at present to determine the sequence of events in (c) and (d).