Between the undoubtedly excellent natural harbours, which exist both at the Pacific and Atlantic ends of this route, the distance across the isthmus is but thirty-nine miles, and only about thirty miles of actual cutting would be necessary. According to Gisborne’s report, no engineering difficulties stood in the way of making a cut of sufficient capacity to form an uninterrupted navigation free from locks from sea to sea. The course of the projected canal was a perfectly straight one, and the greatest depth of cutting required was estimated to be about 150 feet for a distance of two miles. It was claimed that no dredging or deepening of the River Savana would be required, or any other work, such as the construction of dams or locks, be necessary. A concession from the Government of New Granada was obtained, and a company formed and provisionally registered. There was nothing to be done but to make a simple cut some twenty-five or thirty miles long, thirty feet deep and one hundred and forty feet wide at bottom, and all at an estimated cost of only £12,000,000; and yet the scheme fell through. [Illustration: INTERIOR OF A SHACK ON THE ISTHMUS.] The glowing accounts of both Cullen and Gisborne as to the suitableness of the locality, and the absence of difficulty in the carrying out of the work, cause considerable wonder as to the reason for the abandonment of the scheme; for not till twenty years later did Commander Selfridge prove the statements of Cullen and Gisborne to be erroneous, when in the course of an able survey of this region, he showed that a canal through it would necessitate a tunnel of ten miles in length. At least there was no lack of public interest in the question of piercing the isthmus, for farther south in the Darien three particular routes were being investigated. The first of these, by the way of the rivers Atrato and San Juan, had aroused hope on account of a report common amongst the natives that there was in the divide, between these two rivers, a low depression which the Indians used as a portage for their canoes when travelling from sea to sea. Indeed there was a tradition of a waterway having been cut through the short distance separating the higher reaches of these two rivers, but this was never verified. A second Atrato route was by using that river in conjunction with the River Bando, whilst still a third proposed to cross from the Bay of Cupica to the River Atrato.