Though I would not venture into Southwest Bay with the yacht, I knew that we should have to carry our stores and tools there by boat and land them on the beach opposite to the treasure ravine; for to transport them by land from the easy landing-place near the cascade would be an almost impossible undertaking.
According to the dead pirate’s statement, he and his comrades had surveyed Southwest Bay and discovered the best channel between the rocks. He gave the directions for finding this channel to Captain P⸺, and its existence had been verified by both the South Shields explorers; but as they had brought back an alarming account of its dangers, and boats had been lost in it, I considered that it would be a wise precaution for me to land at the pier in the first place, walk—or rather crawl and climb, for there is not much walking to be done on that journey—across the island and survey Southwest Bay from the hills above it, before attempting to beach a boat there.
In the evening we held a council in the saloon over our pipes, and I explained my plans for the following day.
I had explored the island pretty thoroughly while here before, and I knew that it mainly consisted of inaccessible peaks and precipices, among which there were very few passes practicable for men. In many places the cliffs fall precipitously into the sea, affording no foothold. I had landed in both Northwest Bay and the bay beyond it, and, though there were sandy beaches in both these, still, one could go no further, for sheer promontories on either side and mountains equally insurmountable at the back cut off all communication between these coves and the rest of the island. I also knew that it would be impossible for me to walk along the beach from the pier to Southwest Bay, for between these were the two capes that bound West Bay, both opposing barriers of precipices to one’s advance.
But while here with the Falcon, after a difficult and dangerous search which has been fully described in the narrative of that voyage, I at last discovered a pass, and I believe it is the only one, by which the mountains at the centre of the island can be traversed and the windward shore attained.
First, I ascended the steep ravine down which the cascade flows. Having arrived at the summit of the ravine I crossed the groves of tree-ferns, and, after making several descents into ravines which terminated in precipices and so compelled me to retrace my steps, I succeeded in discovering a gully which led me to the beach on the northeast side of the island. From here I found it possible to walk along the beach to South Point, for no insurmountable capes intervened; and from Southeast Bay there was an easy pass under the Sugarloaf Mountain by which the Treasure Bay could be reached. This was the journey which I intended to make once again on the following morning. This route, together with others taken in the course of our explorations, are I believe the only accessible ways on the island.
I knew by experience that the passage over the mountains to the windward beach was both arduous and perilous, and that to climb to Southwest Bay, survey it, and return to the pier would occupy the best part of three days.
The doctor volunteered to accompany me, and I decided to take him with me. It was indeed important that he should make himself acquainted with the pass, for it had been settled that whenever I remained with the yacht he should be in command of the party working on shore, and, as the only reliable water-supply I knew of was at the cascade, it might become necessary for him to lead the men across the mountains to it should a water-famine occur at Southwest Bay. Again, it was certain that bad weather would occasionally make the landing of boats at Southwest Bay impossible for weeks at a time, so that, if there were some urgent reason for communicating with the yacht, this could only be done by crossing to the pier landing-place, at which I am of opinion that one can land ten times with safety to once in Southwest Bay. It had been my intention to form a depot of stores at the pier, but this we found to be