This opportunity to add to my larder could not be neglected, and I set to work to devise a plan for capturing them. I thought, at first, of making a hook from thorns; but this idea was abandoned as not apt to be practical, and I hit upon a plan for making a net. The first inspiration gradually developed into a trap, and took definite shape as I revolved the matter in my mind. It was a simple device, but I spent much time and patience in perfecting it.

First selecting one of the supple vines, about half an inch in thickness, I bent it into the form of a hoop, two feet in diameter, uniting the two ends by lashing them with smaller vines. Then, with the aid of a sharp thorn and thread from the fibre of the cocoanut, I sewed together pieces of the cocoanut cloth so as to make a bag three feet long, with an opening of the same diameter as the hoop. Then I sewed the edge of the opening of the bag firmly to the hoop, which kept the bag rigidly open.

Next, from more vines, I wove a funnel-shaped basket, the larger end fitting inside the hoop, while the smaller end, which was inserted into the bag, had an opening about six inches in diameter. The larger end of this basket, which was like an inverted cone, was lashed to the hoop, all around.

This was my fish-trap, and as soon as it was ready I took it to the brook. The water was normally low and, finding the narrowest place in the current, I built across it a wall of stones, having an opening in the centre of the wall, in width just a little less than the diameter of the hoop. The trap was then set into this opening, with the mouth pointing up-stream, the gentle current keeping the bag distended, while the hoop projecting across the edges of the opening in the wall held the bag in position.

I expected that the fish, swimming downstream, finding no other passage, would enter the opening of the bag and pass through the small opening in the lower end of the cone, thereby becoming imprisoned. From similar devices that my brother and myself had made and used in the brooks at home, I knew that, once inside, the fish would huddle in the lower end of the bag and make no effort to repass through the opening in the end of the cone.

CHAPTER XIV.

Another Exploring Trip; Tropical Fruits.

The morning following the setting of the trap I removed it and, allowing the water to drain out through the meshes in the bag, I found three beautiful silvery fish, not unlike the white perch of the northern waters. Cleaning them, I broiled them over the coals, and found them to be delicious, delicate and fine flavored. After that I had no lack of fresh fish.

Puff, meantime, had made rapid progress in the art of talking and could carry on quite a conversation--and many were the hours I spent assisting him in adding to his vocabulary. He insisted upon following me everywhere I went, always walking and never making any attempt to fly, his instructions in the use of his wings having been neglected since his capture. When walking through the bush in search of pigeons and other provisions, I always carried Puff perched upon my shoulder, and he never made any attempt to escape. He seemed as perfectly satisfied with my society as I was with his, and we kept up a continuous conversation.

My supply of salt becoming nearly exhausted, I procured a large supply, enough to last me many months, by the same process as had been employed on the first experiment while marching along the coast. This I sealed securely in joints of bamboo.