I purchased some sago from the chief, and when we got bearers together I started for Ekeikei. One day’s journey brought us to our destination, which was situated 1500 feet above sea level, on the foot-hills of the Owen Stanley range. This point I had already selected in my mind as the scene of my future labours, and I at once set about building a permanent camp. I chose the site in a part of the forest overlooking a fine valley, and we set to work speedily, felling the forest trees to make the necessary clearing. It was a big business, much bigger than founding our establishment at Dinawa, but I intended to erect much more permanent structures, which were to be built large enough not only to serve for scientific work, but as a depôt for expeditions to other districts. The house and two collecting verandahs were all in one building, one verandah facing the forest and the other the valley, so as to permit of work being carried on whatever the direction of the wind. The whole structure was built on poles 6 feet 6 inches off the ground, so that my natives could shelter, sling their hammocks, and take their meals below. This work occupied us three weeks, and in it we were assisted by Mavai’s people, who were helped by the villagers of the neighbouring chief, Kafulu. These came in to lend a hand for the sake of tobacco and other trade articles they needed. The best thatch to be obtained in Papua is the sago leaf, and of this the natives make roofs that are water-tight and very durable. At Ekeikei we adopted this method. Along the rafters of our house we ran horizontal bamboos, and instead of a ridge-pole roof we had two of these bamboos running from end to end a few inches apart. The frond of the sago leaf which we used for this purpose is at least 4 feet long; it measures 6 inches at the base, and tapers to a point. To begin the thatch, one takes the leaf and bends it two-thirds away from the apex. One starts from the bamboo horizontal that lies nearest the eaves, and hooks the leaf over, laying the pointed end out. On the next higher bamboo one hooks over another leaf, similarly folded, so that its long pointed end far overlaps the other, and so on until the ridge of the roof is reached. The operation is thus repeated until the whole roof is thatched. The space between the two parallels which form the ridge-pole is finally covered with grass laid thickly across and across. The sago leaf is grooved laterally, and forms, as it were, a natural water-spout for carrying off the rain. So durable is this roof that after an absence of five months we found that our Ekeikei house was still water-tight. This thatch is, however, a great harbourage for cockroaches, and there must have been millions of them in our house. At night we could hear them rustling among the dry leaves. I could not ascertain that they had done any actual damage, and they had the grace not to fall down upon us.