Stiffening the Bridge.



The bridge was a pretty good one, except for a slight unsteadiness between the center and either end. When Uncle Ed saw it he showed us at once where the trouble lay. Our intermediate cross beams were hung from the center of the spars, and consequently made them bend, because the strain came across their length, while at the center of the bridge there was no chance for the spars to bend, because the strain was exerted along their length, that is, it tended merely to push the ends of the spars deeper into the banks. To remedy the trouble he proposed propping up the center of each spar with a brace running from the center crosspiece. The dotted lines in Fig. 100 show how these braces were applied. They made the floor perfectly solid throughout, and gave the bridge a much better appearance. Uncle Ed told us that the structure might be called a "king rod truss," except that in place of rods we had used wires.



The King Post Bridge.

The other bridge on the Schreiner property was built in the following summer, just before we started on our second expedition to Willow Clump Island. It spanned the brook at the gorge, and was therefore a more difficult engineering feat. Mr. Schreiner himself asked us to build it, and we felt greatly honored by the request. A search was made in the Van Syckel library for a suitable type. At last we found one that seemed properly suited to the requirements. It was called a "king post truss," and was very similar to the king rod bridge. While the design of the bridge was simple, yet it required some ingenuity to put it together. In setting up the other bridge the scow had been anchored in the center of the stream and used as a working platform, from which it had been an easy matter to put the various parts together. In this case our scow was obviously of no use, so we laid a couple of long logs across the chasm, and a few slats were nailed across them to provide a temporary bridge or working platform. The platform sagged considerably at the center, because the span was fully eighteen feet; but the logs were large, and we knew they were strong enough to support our weight. However, as an extra precaution, we tied the ends to stakes driven in the ground, so that they could not possibly slip off the banks.



First we set about constructing the king posts, which were made as shown in Fig. 101. Two stout posts 7 feet long were connected at the top by a tie stick, which spaced them 4 feet apart. To make a secure fastening they were notched together and strengthened with diagonal braces. Each king post was notched on opposite sides, at about thirty inches from the top. A temporary tie piece was also nailed across the lower ends of the king posts. The frame thus formed was set up at the center of the span and temporarily held by nailing the lower tie piece to the working platform. Four stout spars were now cut, each about fifteen feet long. Taking a pair at a time, we planted their lower ends firmly in the opposite banks and sawed off their upper ends until they could just be hammered into the notches in the king post. This required careful fitting, but by making the spars a little too long to start with, and then shaving them down with a draw-knife, we managed to make fairly good joints. A couple of long wire nails in each spar made the structure perfectly secure. The king posts were now sawed off just above the temporary tie piece, and the permanent cross beam was fastened to these ends with straps of heavy wire wound tightly about them. The working platform sagged so much that we were able to lay this cross beam above it. From the ends of the cross beam diagonal braces extended to the king posts (Fig. 103). Our working platform was now removed and replaced with the permanent floor beams, which were firmly nailed to the center cross beam and to the inclined spars at the shore ends. The floor beams were quite heavy and needed no support between the king posts and shore. A rustic floor was made of small logs sawed in two at Mr. Schreiner's sawmill. Light poles were nailed to the flooring along each edge, giving a finish to the bridge. We also provided a rustic railing for the bridge of light poles nailed to the king posts and the diagonal spars.



CHAPTER X.

Canvas Canoes.

Like all inhabitants of islands, we early turned our attention to navigation. Our scow was serviceable for transporting materials back and forth across the strips of shallow water between our quarters and the Jersey shore. We never attempted to row across, because progress would have been entirely too slow, and we would have drifted down to the rapids long ere we could reach the opposite side. But on Lake Placid matters were different. Although there was no settlement near us on the Pennsylvania shore, to occasion our crossing the water for provisions and the like, yet the quiet stretch was admirably suited to boating for pleasure, and mighty little pleasure could we get out of our heavy scow.



Uncle Ed's Departure.

Owing to a sudden business call Uncle Ed left us after he had been with us nearly three weeks. But, before going, he explained carefully to Bill just how to construct a canvas canoe. Jack, the cook, who was anxious to lay in a second supply of provisions, accompanied Uncle Ed as far as Millville, the next town below Lamington. Here Uncle Ed bought five yards of canvas, 42 inches wide, several cans of paint and a quantity of brass and copper nails and tacks. These supplies, together with the food provisions that Jack had collected, were brought to us late in the afternoon by Mr. Schreiner. Mr. Schreiner also brought the necessary boards and strips of wood for the framework of our canoe.



A Visit from Mr. Schreiner.

We invited Mr. Schreiner to spend the night with us, and this he did after fording with some difficulty the swift-running river. In the morning we showed him our quarters, our filter, the roads we had built, the spar bridge across to Kite Island, our surveying instrument and the chart we had made of the vicinity. He was greatly pleased with our work, and it was then that he gave us an order for the bridge over the gorge. From that day on he became our staunchest ally, so that when my father and Mr. Van Syckel complained that we were loafing away a lot of time which could be more profitably spent in study or work, Mr. Schreiner stood up for us and declared that our experiences on the island were doing us far more good, both physically and mentally, than any other work that they could conceive of; that before condemning us they should pay us a visit and see how we were employing our time.



The Sailing Canoe.



Immediately after Mr. Schreiner's departure we started work on the canoe. A strip of spruce 1 inch thick, 3 inches wide and 12 feet long served as the keelson. At the stern a post 1-1/2 inches thick, 3 inches wide and 13 inches high was secured to the keelson with brass screws. This was braced as indicated in Fig. 104. At the bow a stem piece was attached to the keelson. This stem was cut to a somewhat semicircular form, as shown in Fig. 105. The outer edge was tapered with a draw-knife to a thickness of 1/4 inch and a brace was nailed to the inner edge. Our next work was to cut out three forms, one of the shape shown in Fig. 106 and two like that shown in Fig. 107. The first form was set up on the keelson midway between the stem and stern, and the other two were spaced about four feet each side of the center form. The center form was used only for shaping the frame of the boat, and was not intended to be permanently affixed to the canoe. Therefore, we fastened it to the keelson very lightly, so that it could be readily removed. The other two forms, however, were made permanent parts of the frame, serving as bulkheads. The gunwales were now secured in position. These were of spruce 3/4 inch thick and 2 inches wide. The ends were beveled off so as to neatly fit the stem piece and the stern post, to which they were fastened by brass screws. Then we applied the longitudinal strips, or rib bands, which were of 1/4-inch thick spruce 1 inch wide. Ten of these bands were used, equally spaced apart on the center form, to which they were lightly tacked; but they were nailed securely to the bulkheads and the stem piece and stern post. The cross ribs were made of barrel hoops which we had soaked in water for a day or so to render them pliable enough to be bent into place. These hoops were split to a width of 1/2 inch, and secured first to the keelson, then to the longitudinal strips and finally to the gunwales. Copper tacks were used for nailing the ribs in place, and these were long enough to be passed through the rib bands and clinched on the outside. Forty cross ribs were nailed on, and at the center of the canoe they were spaced about three inches apart. The center form was then removed and cut along the dotted lines shown in Fig. 106. The semicircular pieces thus obtained were now strengthened with strips on their inner edges, and wedged in between the keelson and the gunwales, to which they were nailed, as shown in Fig. 108. A pair of cleats nailed to the cross ribs served as supports for the seat of the canoe. The frame of the boat was completed by nailing in place two deck beams of 1/2-inch square pine and four corner pieces between the gunwales and the bulkheads, so as to make an elliptical well hole or deck opening. Before laying on the canvas covering the edges of the gunwales, keelson, deck beams, stem and stern posts were smoothed down with sandpaper.



Stretching on the Canvas.



The frame was laid in the center of the canvas and the latter drawn around it. Then with a large needle and strong twine we sewed both edges of the cloth together with long stitches, lacing the canvas over the frame as a shoe is laced over a foot. This done, the boat was turned deck downward and the canvas was tacked to the keelson. In each case, before driving in a tack a daub of white lead was applied, to water-proof the spot. At the stem and stern a gore (narrow triangular piece) was cut out of the canvas so as to make it lie smooth on the frame, and white lead was painted in between the overlapping edges. The canoe was then turned deck upward and the lacing tightened, while we carefully worked out all wrinkles in the cloth. After tacking the canvas along the gunwales on the outside, it was trimmed off, leaving sufficient margin to be brought over the gunwales and tacked inside. Two triangular pieces were cut out for the decks, and these were lapped over the outer canvas and tacked to the gunwales. A narrow molding along the edge of the boat served to cover the tack heads and added a certain finish to the canoe. A keel plate 2 inches wide and 1 inch thick was attached to the outside of the boat, and then, after wetting the canvas, it was given a coat of white lead and oil. When this was perfectly dry it was sandpapered and the second coat applied.



The Rudder.



The canoe was now complete except for the rudder, which was cut from a 1/2-inch board to about the shape shown in Fig. 114. Strips 1-1/2 inches wide and 1/2 inch thick were nailed to each side of the blade, forming a post, to the top of which a crosspiece or tiller was fastened. A cleat nailed to the pillar at each side of the rudder post served to greatly strengthen the joint. The rudder was hinged to the canoe by a rod, which passed through four brass screw eyes, two threaded into the rudder and a corresponding pair screwed into the stern. For convenience in steering we ran our tiller rope clear around the boat, through screw eyes in the gunwales and a pulley at the stem, so that the steersman could guide his craft from any point in the canoe.



The Deep Keel.



We planned to use our canoe as a sailboat, and had to provide a deep keel, which, for convenience, was made detachable. This keel was 6 inches wide, 1/2 inch thick and 6 feet long, and was fastened at the center of the canoe. Screw eyes about twelve inches apart were threaded alternately into opposite sides of the keel plate. Corresponding hooks were attached to the keel in position to hook into the screw eyes, and thus hold the keel firmly in place.



Canoe Sails.



Our boat was fitted with two masts, a mainmast and a mizzen or dandy mast. The former was 6 feet long and the latter 4 feet long, and each measured 1-1/2 inches in diameter at the base, tapering to about 1 inch diameter at the upper end. They were held in brass bands, or clamps, bent around them and secured to the bulkheads, as shown in Fig. 117. The sails were of the lanteen type. The mainsail measured 8-1/2 feet along the boom, 9-1/2 feet along the yard and 10 feet at the leach. The dimensions of the mizzen sail were: along the boom, 5 feet; along the yard, 5-1/2 feet; and at the leach, 6 feet. The boom was attached to a strap of leather on the mast, and was thus given freedom to swing around in any desired position. The yard was similarly attached, and was raised by a cord, which passed through pulleys at the top and at the base of the mast and extended to a cleat within easy reach of the occupant of the boat. A double paddle was fashioned from a board 1 inch thick, 6 in wide and 6 feet long. The blades were shaved down to a thickness of 3/8 of an inch at the edges.

It will be observed that we used no iron in the construction of this boat. Uncle Ed has warned us not to, because iron rusts out so easily and is apt to damage both the canvas and the wood with which it is in contact.

A canoe is rather a tipsy thing to sail in, as we soon learned, and it was lucky that we could all swim, else our vacation might have ended very tragically; for the very first time Bill and I tried the boat an unexpected gust of wind struck us and over we went. We were very poor sailors at first, but it didn't take us long to catch on.



Lee Boards.

One thing that bothered us greatly in sailing was the keel of our canoe. It was forever getting twisted, particularly when we tried to make a landing. There were only a few places along the island where the water was deep enough to permit our coming right up to shore without striking the keel. The fastening was not very strong, and every once and awhile it would be wrenched loose. The matter was made the subject of a special letter to Uncle Ed, and in due time his answer was received. As usual, he offered a first-class solution of the difficulty. "Don't use a keel," he wrote; "lee boards are much better." Then he went on to explain what was meant by lee boards: "The leeward side of a boat is the opposite of the windward side; that is, that side of the boat which is sheltered from the wind. Lee boards, then, are boards which are hung over the lee side of a boat to prevent it from drifting to leeward, and they serve to take the place of a keel or centerboard."



Following Uncle Ed's direction we fastened a strip of wood across the canoe about six feet from the bow, nailing it firmly to the gunwales. This provided a support to which the lee boards were secured. The lee boards were paddle-shaped affairs of the form and dimensions shown in Fig. 121. Each paddle near the top was hinged to the end of a board three inches wide and a foot long. The paddle was held at right angles to the board by means of a hook. Each board was fastened with door hinges to a baseboard which extended the width of the boat and was attached to the crosspiece of the canoe by means of a couple of bolts. The bolt heads were countersunk, so that the hinged boards could lie flat over them. To the top of each lee board two ropes were attached, one passing forward around a pulley and thence back to a cleat within easy reach of the occupant of the canoe, and the other passing directly back to this cleat. By pulling the former rope the lee board was lifted out of the water, while the latter rope was used to swing the board into working position. When tacking to port (left), the board on the left side of the canoe was lowered and the other was raised, as shown in Fig. 123, and when tacking to the starboard (right) the board on the right side was lowered, while the left one was raised.



The Indian Paddling Canoe.



Our sailing canoe proved such a good one that we decided to build a second. This was to be much lighter, for paddling only, and of the true Indian shape, with wide, bulging sides and raised stem and stern. The dimension of the forms used are given in Figs. 124 and 125. These forms, it will be observed, were notched to receive the keelson and gunwales. The keelson was formed of 1-inch spruce 3 inches wide and 10 feet long. The stem and stern, which were both of the same shape, were cut from a 12-inch board to the form shown in Fig. 126, and were firmly secured to the keelson. This made the boat 12 feet long. The forms were then set in place on the keelson, one at the center and the others three feet each side. The gunwales were formed of 3/4-inch by 2-1/2-inch spruce, and the twelve rib bands used were of the size used in our first boat. As none of these forms was to remain in the boat, nails were driven very lightly into them, with heads projecting so that they could easily be withdrawn when it was time to remove the forms. The cross ribs were passed under the keelson inside of the rib bands and outside of the gunwales, as shown in Fig. 128. After they were set in place and firmly secured with copper tacks, a band was nailed to the keelson to form the keel. To produce the raised stem and stern, four wedge-shaped pieces were nailed to the tops of the gunwales, as indicated in Fig. 129. The forms were then removed and were replaced with cross sticks braced between the gunwales. The center cross stick was provided with two corner pieces, as shown in Fig. 130, adapted to fit under the gunwales and against the rib bands. The canvas was then applied in the manner described before, but was tacked to the upper edge of the gunwale instead of the outer side, and the tacks were covered by a half-round molding which extended around the entire boat. After the lacing was cut the edge of the canvas was secured to the under edges of the gunwales. The canoe was then completed by fastening on a 1-inch square keel and treating the boat with two coats of paint. The paddle was a duplicate of the one described in connection with the sailing canoe.

I remember that we eventually equipped our paddling canoe with a sail and a pair of lee boards, though no record of this fact appears in the chronicles of the society.



CHAPTER XI.

HOUSE BUILDING.

One afternoon Fred, who had waded over to Lumberville after some provisions, came splashing back holding aloft a large square envelope. It was from Uncle Ed and contained a photograph of a group of Wichita Indians building a large grass lodge. In a brief explanatory letter Uncle Ed suggested that we build a similar hut on our Island.



The Grass Hut.



The grass lodge appealed to us as very picturesque, and we set to work immediately on its construction. We made our hut much smaller, however, only 12 feet in diameter, and 8 or 9 feet high. First we procured two dozen light poles between 10 and 12 feet long. These we set up about 18 inches apart in a circle like a stockade, the sticks being buried in the ground to a depth of 12 inches. At one side a space of 3 feet was allowed for a doorway. Inside the stockade we erected a working platform of planks supported on barrels, and standing on this we took two opposite poles, bent them inward and lashed their upper ends together. Then a second pair of opposite poles were similarly bent inward and tied, and so we proceeded until the entire stockade had been converted into a dome-shaped cage. Around these poles we laid lighter sticks, or bands, tying them at the points of intersection. At the doorway two posts were set firmly in the ground, projecting upward to a height of 4 feet. A lintel nailed across the top of the posts completed the door frame. Sticks were nailed to the lintel and to the side posts, extending to the main frame of the hut, to which they were tied. We were now ready to thatch our hut. Reddy and Dutchy went over to Lumberville for several bales of straw. We tied the straw in bunches and applied it to the frame, copying, as best we could, the process illustrated in the photograph.

But for its location the hut would have proved a very serviceable habitation. In order to have a good, dry dwelling without laying down a board flooring, we had selected for its site the sandy shore at Point Lookout. This part of the island was not sheltered with trees, and the hot sun beat down on our hut so strongly that we found the quarters very uncomfortable indeed. It was this fact that led to the construction of a tree hut—a building that would be perfectly dry and yet shaded and cool. Bill had read of such houses in the Philippines and felt confident that we could build one. We couldn't decide at first where to locate our hut until Dutchy moved that we build it in the gnarled oak tree overlooking the "Goblins' Dancing Platform." Immediately the motion was seconded and unanimously carried.



The Goblins' Dancing Platform.

Just above the town of Lumberville there was a cliff which rose sheer 200 feet above the level of the river. So perpendicular was the cliff that a stone dropped from the overhanging ledge at the top would fall straight down to the railroad track below without touching a twig in its course. Back of this broad ledge there was a very peculiar formation. A column of stone rose abruptly 40 feet higher and was topped with a large slab about 12 feet in diameter. This was known all over that region as the Goblins' Dancing Platform. The only possible way of gaining the summit of the column was by climbing a scraggly oak tree which grew on the high ground back of the pillar, crawling out on an overhanging limb, and then dropping down to the platform below. It was in this oak that we decided to build our house. It was a very inaccessible spot, and to reach it we had to make a wide detour around the back of the hill, and through the fields of a cranky farmer, who more than once threatened to fill us with bird shot for trespassing on his property. How were we to carry all our building materials up to this great height? One would think that the difficulties would be enough to discourage us, but not so with the S. S. I. E. E. of W. C. 1. Nothing daunted us.



Dutchy Takes a Dare.

Our first task was to try some other approach to the top of the cliff. At one side of the overhanging ledge there was a fissure in the rocks which ran from the base of the pillar to the foot of the cliff. Down this zigzag crevice Dutchy had scrambled, one afternoon, on a dare. We were rather frightened when he started, because it was a very hazardous undertaking, and we watched him anxiously, peering over the edge of the precipice. By bracing his back against one of the walls of the rock, and digging his feet into the niches and chinks of the opposite wall, he safely made his way to a shelf about half-way down, where he paused to rest. From that point on the fissure widened out, and a steep, almost vertical incline, sparsely covered with vegetation, led to the railroad track below. I think he must have become rather frightened at his position, because he hesitated long before he resumed his downward course, and when he finally did make the attempt his foot slipped upon the moss-covered rocks and down he fell, scratching and clawing at every shrub within reach. Believing him to be killed, we rushed down the hill and around to the foot of the cliff. It probably took us about fifteen or twenty minutes, though it seemed ages before we came upon our venturesome comrade coolly trying to pin together a rent of inconvenient location and dimensions in his trousers.

"Say, Dutchy, are you killed?" cried Bill, breathlessly.

"Killed, nothing," he replied, with scorn. "I suppose you fellows think I had a fall. Well, I didn't."

"You didn't, eh? We saw you slip."

"Oh, go on. I came down that way on purpose. There was no use in picking my way down like a 'fraid cat, when I could just as well take a smooth and easy toboggan slide on the bushes all the way down."

Smooth and easy toboggan slide! Well, you should have seen the hillside. The course was well defined by the torn and uprooted shrubs and the pile of branches and vines at Dutchy's feet. Whether the hare-brained Dutchy really imagined he could glide easily down on the shrubbery, his frantic movements on the way certainly belied his story, and when, the next day, we proposed that he repeat the trick, somehow he didn't seem to be very enthusiastic on the subject.



A Path Up the Fissure.



It was up this fissure that we decided to haul materials for our tree hut. Our first task was to build steps and ladders in the steepest parts. We had no tool for cutting out niches in the rock, but wherever natural depressions were formed we wedged in sticks of wood between the side walls to serve as ladder rungs. If no such niches appeared for considerable height, we would stretch a rope ladder to the next fixed rung. In most places the natural formation of the rock was such as to afford sufficient footing.



Rope Ladders



The rope ladders were made of two parallel side straps, tightly stretched between the fixed sticks, and then at intervals of fifteen inches we inserted the ends of the ladder rung between the strands of the rope. Below and above each rung the rope was bound with cord. The rungs were notched at the ends to prevent them from slipping out.

After providing a means for scaling the cliff (we called it the Jacob's Ladder), we were still confronted with the problem how to cart our building materials to the top. It was a very hard task and you couldn't have hired us to do it under any other circumstances. First, Bill planned out on paper just how the house was to be built, and we cut all the pieces to the right size so as not to carry up any superfluous matter. When all was ready the boards and sticks were loaded on the scow, and ferried over to the cliff. Then we carried them on our backs, three or four at a time, up the slanting hillside to the first ledge. From there up, owing to the steepness of the ascent, we had to employ different tactics.



The Derrick.



A derrick was constructed of two sticks 10 feet long, which were bolted together at the top, and secured about five feet apart at the bottom by a cross piece, as shown in Fig. 136. The derrick was then taken apart and with some difficulty hauled piecemeal up to the next ledge above. Here it was put together again. The fall and tackle used in our aerial railway was attached to the apex of the derrick, and the latter was then erected with the legs set into depressions in the ledge and the upper ends slanting outward but kept from falling over the edge by a rope tied to one of the fixed rungs set in the fissure. With this derrick we hoisted up the boards in a few hauls. The job was a very ticklish one, but Bill used the greatest care to prevent accident. The derrick, rope and tackle were carefully tested before used, and as soon as the load was attached to the lower pulley block the two who did the loading were instructed to crawl back into the fissure so as to be out of danger in case anything gave way. At one time a stick which had been carelessly tied did fall, and it might have badly hurt some one had we not observed this precaution. When we had raised the material to the second ledge we transferred operations to the top ledge, and when the materials had been hauled up to this point we finally rigged up our fall and tackle in the old oak tree itself.



The Tree House.



The tree had two large limbs which extended out at a wide angle from the main trunk. Across these two limbs, at about seven feet out, we laid our first girder, nailing it securely in place. Then to the main trunk we nailed the second girder on a level with the first. Diagonal braces were extended from the trunk to support the ends of this girder, and a tie piece was nailed to the braces, as shown in Fig. 138, to prevent them from spreading. The girders were rough sticks about 4 inches in diameter and 10 feet long. We cut flat faces on them at the points where they were nailed to the tree, and then, to make them doubly secured, we nailed cleats, or blocks of wood, to the tree under them. The floor beams were then laid across and nailed to the girders. They were cut to a length of 10 feet so as to project beyond the outer girder to provide for a piazza overhanging the Goblins' Platform. Six floor beams were used, spaced 20 inches apart. All branches projecting up between the beams were then cut away and a flooring of slabs was laid on. To the main trunk six feet above the flooring, a stick or (to use the technical term), "wall plate," was nailed on, and its ends were supported by upright posts resting on the platform. Thirty inches from the outer end of the platform two more posts were erected eight feet high and secured by sticks nailed across from the other posts, and also by a second wall plate connecting their upper ends. Four more posts were erected, one between each pair of the corner posts, and then we were ready to enclose the framing.

The sidewalls were first clapboarded, because we were afraid the roof would not hold us until the framing had been strengthened by nailing on the siding. Slab boards were used for this purpose. Beginning at the bottom, the boards were laid on, each lapping over the one below, as shown in Fig. 141, so as to shed water. In each side we cut a window opening and nailed on a window casing of the type shown in Fig. 142, which will be described in a moment. As soon as the clapboards were applied, we nailed on the rafters and then applied the roofing. The same principle was here used for shedding water. The lowest board was first laid on, and then the others were successively applied, each lapping over the one below.



The window casings we used each consisted of a frame about 15 inches square, but with the upper and lower pieces extending 12 inches beyond one of the side pieces. On these extended pieces a slideway was formed for the window sash by nailing on two strips of wood about 3/4 inch square and over them a pair of wider strips projecting inward, so as to overlap the edges of the sash. The window sash consisted of a frame 13-1/2 inches square, made of 3/4-inch square strips over which canvas was tightly stretched and tacked. A spool was nailed on at one side for a handle. These windows were closed only in rainy weather, to keep the water out.



Sliding Doors.



We had two doors; one at the back of the house, from which a ladder extended down to the ground, and another opening out onto the veranda, from which we dropped a ladder down to the Goblins' Dancing Platform. In order to save space we used sliding instead of swinging doors. The back door frame was 5-1/2 feet high and the front door frame 6 feet high. The doors were mounted on the outside of the building. The side posts of each frame were 2-1/2 feet apart, and the lintel and sill extended 3 feet beyond the side post at one side. The upper face of the lintel was planed down perfectly smooth, and its edges were tapered off to make a track for the rollers on the door. The rollers consisted of two spools, which turned on tenpenny nails driven into the top of the door. At the lower end two more spools were mounted, turning on nails driven in the bottom edge of the door. The rims of the spools extended slightly beyond the outer face of the door and rolled against the sill. To keep the water from leaking in at the top a slanting board was fastened above it, as shown in Fig. 144. The back door was similarly constructed. Our tree house was completed by a running balustrade around the veranda.

It strangely happened that just after our tree house had been built we received a photograph from Uncle Ed of a Filipino tree house made of bamboo.



CHAPTER XII.

TROUBLE WITH THE TRAMPS.

We were a proud lot when the house was finally completed. From the veranda we had an excellent view up and down the river. We could see our camp on the island and keep watch of our goods. Late one afternoon Dutchy and I were lolling about on the Goblins' Platform, idly watching a hawk soaring above us. The rest of the boys had returned to the island in canoes an hour before and left the heavy scow for us to row back. It was drawing near supper time and we had about decided to start for home, when I chanced to see a scow up the river. It looked exactly like ours, and in it were two men, evidently drunk, from the way they carried on. A glance showed me that our scow was not at its moorings. How were we to reach the camp? One of the men had evidently seen us and was pointing us out to his companion. We rushed down the Jacob's Ladder, but by the time we reached the river bank they were in midstream and heading rapidly northward. Our shouts merely brought forth derisive laughter. We were certainly in a predicament. First we ran back up the cliff, and tried from there to gain the attention of the rest of the fellows. They evidently saw us but couldn't make out what we wanted. Then we ran down to a point opposite the island and called to them. But the wind was against us and we couldn't make them hear, so we had to plunge in and wade across.



A Council of War.

Immediately we summoned a war council. Dutchy and Jack were chosen by lot to guard the camp, while the rest of us started in pursuit in canoes. By the time we got under way the sun had dropped back of the Pennsylvania hills and the shadows were climbing slowly up the Jacob's Ladder. Swiftly we paddled up-stream, keeping close to the western shore, where the water was very quiet. We didn't expect to go far, because there were rapids less than three miles up, and we were sure that no tramps would ever be ambitious enough to row a heavy scow against the swift current at that point. As we rounded a sharp bend in the river, we noticed a camp fire a few hundred feet further up, around which five or six men were lounging, and there, just below them, was our scow. What were four boys to do against six grown men? We were each armed with a club, and could have made a pretty good fight if necessary, but after a whispered consultation we decided it would be best to wait until dark, when we could creep up quietly and steal away unnoticed with our boat.



Vengeance.

It seemed as if darkness never would come. It was scarcely dusk when our patience gave out and we paddled up stealthily, hugging the shore. Bill gained the scow unnoticed, but just as he was about to push off he discerned the body of a man within. It was one of the tramps lying there in a drunken stupor. What was to be done? Every moment was precious. A yell from the fireside decided him. With a mighty push he launched the boat out into the current, while we threw him a line and towed the boat out to midstream. With a volley of curses the men sprang up and pelted us with stones. But they were poor shots, and we escaped without serious injury. Our prisoner, in the meantime, was snoring heavily in the scow undisturbed. We took him down-stream and then unceremoniously picked him up and dumped him overboard within a few feet of the shore. It was a rude awakening, and nearly frightened the wits out of the man. But it brought him to his senses, and in a moment we were dodging more stones, sent with such good aim that we had to lie flat in the bottoms of the boats until the current carried us out of reach.



A Double Surprise.



It was now quite dark, and we had some difficulty in groping our way back to camp. There was no moon and the stars were obscured by clouds. Our only course was to follow the shore line until we got around the bend, and then we steered for the beacon fire, which, by prearrangement, had been kindled on Point Lookout. But the spirit of mischief was in us. We thought we would have some fun with Dutchy. We could see him silhouetted against the blaze. Jim and I hung back in the canoes, while Reddy and Bill went on with the scow, splashing their oars and shouting and singing in disguised voices, like drunken men. Dutchy was evidently very much agitated. His "Hello, there! Boat ahoy!" was greeted with derisive yells.

"Say, we'll lick the life out of you, the same as we did them other kids," shouted Reddy.

This was too much for Dutchy. He ran for all he was worth, yelling for Jack to come quick.

We had a merry laugh over the situation when suddenly the tables were turned. Something whizzed past Bill's ear; I was stung on the arm with a heavy nail; a large stone hit the scow; Reddy had his hat knocked off, and Fred upset his canoe trying to duck out of reach of the invisible missiles before we could make our assailants understand that we were friends and not the tramps. The joke was on us after all. We hadn't counted on Dutchy's accurate aim or Jack's skill with the crossbow.



Tramp-proof Boat Mooring.



Around the camp fire that night we discussed our adventures and made plans to prevent their recurrence. It was evident, for one thing, that we would have to moor our boats off shore in such a way that they would be out of reach of meddlesome persons, and yet could be drawn in toward shore by anyone who knew how. This was the way we did it. A pair of galvanized iron ring bolts were procured on Jack's next trip to Lamington for provisions, also a light rope about forty feet long. The ring bolts were screwed into a pair of stout anchor stakes about two feet from their, lower ends. The rope was passed through the rings and the ends were joined by tying them to a galvanized iron link. Then it was soaked for a while to shrink it before it was set in place. After the rope had shrunk sufficiently, the two stakes were driven into the bed of the river, one close to the bank and the other far enough out to hold the rope belt clear of the bottom. Both stakes were sawed off under water, just above the ring bolts, so that they were hidden from sight. When we wanted to moor our boats we secured their anchor ropes or "painters" to the link. A large stepping stone marked the spot were the inner stake was driven, and standing on this stone we were able to reach down and haul in on the lower strap of the belt to draw the boat out a safe distance from shore, and then when we wanted to use our boat again we would haul in the upper strap to draw the boat in toward shore.



CHAPTER XIII.

WIGWAGGING AND HELIOGRAPHING.

Our tramp adventure was really quite a blessing to us, for it taught us the necessity of a good signaling system between the Goblins' Platform and the island and led to our learning how to wigwag, and later to the construction of a heliograph. Uncle Ed, when he read of our experience, sent us the U. S. Army "Manual of Signaling." Fred, the tailor of our camp, made us two white flags with red centers. Each flag was two feet square and was fastened to a light staff about five feet long. Then we got out the manual and practised sending signals, at first within shouting distance, until we got to be quite expert.



Wigwag Signals.



There were only three different movements that could be made with flags, but in the book different combinations of these movements were given to represent each letter of the alphabet and the numbers from 1 to 0. All these movements were begun and ended by holding the flagstaff upright, directly in front of the body, as shown in Fig. 147. The first movement was to swing the flag down to the right and back (Fig. 148), the second to the left and back (Fig. 149), and the third forward and back (Fig. 150). The following table gives the different combinations used for various letters:

The Wigwag Alphabet.

A 22 J 1122 S 212 B 2112 K 2121 T 2 C 121 L 221 U 112 D 222 M 1221 V 1222 E 12 N 11 W 1121 F 2221 O 21 X 2122 G 2211 P 1212 Y 111 H 122 Q 1211 Z 2222 I 1 R 211 tion 1112

Numerals.

1 1111 4 2221 8 2111 2 2222 5 1122 9 1221 3 1112 6 2211 0 2112 7 1222

The numbers 1, 2 and 3 indicate respectively the first, second and third movements. For instance, A was represented by the combination 22, which means that the flag must be swept to the left and back twice. B is represented by the combination 2112, that is, a sweep to the left, two sweeps to the right and a final sweep to the left, as shown in Fig. 151. The end of a word was represented by a sweep forward and back: the end of a sentence by two sweeps forward and back, and the end of a message by three sweeps forward and back. It will be noticed that the same combinations are used for 2 and Z, 3 and tion, 4 and F, 5 and J, 6 and G, 7 and V, 9 and M, and 0 and B. The following abbreviations were given in the Manual:

Abbreviations.

a after n not uc your b before r are w word c can t the wi with h have u you y yes

These abbreviations saved a lot of time, for when we wanted to signal the word after instead of spelling it out—22-2221-2-12-211-3—we used the signal for A—22—followed by 3 to signify that it was the end of the word. Before was represented by 2112-3, your by 111-211-3, etc. It took quite a little practice to learn the different combinations. Fred and Reddy soon became experts, and could flash the signals back and forth at a great rate.



Wigwagging at Night.

At night we used a torch in place of a flag. The torch consisted of a roll of dried birch bark tied with wire to the end of a staff. It was found necessary to place another torch on the ground directly in front of the signaler so as to fix a central point and enable one to determine whether the moving torch was swung to the left or right. A later improvement was to use three lanterns, one in each hand and one attached to the waist to fix the central position. It was quite an advantage to have a lantern in each hand, for it saved changing over from one to the other when a second movement followed a first or a first movement a second.



The Heliograph.

The book that Uncle Ed sent us had in it a description of a heliograph, that is, an instrument for sending signals with flashes of sunlight. Although our wigwagging system was good enough for our requirements, yet we thought it would be more scientific to use the sun instrument, and besides, the latter could be used for signaling many miles.



The Single Mirror Instrument.



The first thing we did was to procure a small mirror about 4 inches square, mounted in a wooden frame. Then we got a pair of small square head bolts about 1/4 of an inch in diameter and 1 inch long, also two strips of brass 1/2 inch wide and 3 inches long. In the center of each brass strip we drilled a hole just large enough to admit the shank of one of the bolts, and then the strips were fastened with screws tight against opposite edges of the mirror frame, with the heads pressed against the frame and the shanks sticking out at each side, as shown in Fig. 153. These projecting shanks served as "trunnions" (that is, pivots) for the mirror to turn on when it was mounted in place. After the trunnions had been set in place we made a peep hole in the center of the mirror by cutting out a piece of the wooden back of the frame and scratching away the silver from the back of the glass. Only a very small hole was required, about 1/8 inch in diameter. Great care was taken to have the unsilvered spot exactly on a line with the trunnions and just half-way between them. This done, we took two sticks of 3/8 inch wood, 1 inch wide and 3-1/2 inches long. In the upper end of each stick a slot was cut 1/2 inch deep and 1/4 inch wide. Into these slots the trunnions of the mirror were placed, and then the nuts were screwed tightly on, clamping the sticks against the sides of the mirror. The sticks were now connected by nailing a 1/2-inch strip at the bottom, and braced by a couple of corner pieces. This formed a swiveled frame for the mirror, which was clamped to the base of the instrument by means of a bolt 1-1/2 inches long. The bolt passed through the bottom board of the frame, squarely under the peep hole of the mirror and through the baseboard of the instrument near one end. The baseboard was 2 inches wide, 10 inches long and 3/4 inch thick.



The Sight Rod.



At the end opposite to where the mirror frame was swiveled we mounted a sight rod, which was merely a round stick of wood 1/2 inch in diameter and about 8 inches long. We cut the stick from one of the rounds of an old broken chair. The upper end of the rod was whittled to a point and one side was flattened as shown in Fig. 155. Out of a piece of heavy white cardboard we cut a round disk about 1/4 inch in diameter, with a shank 1 inch long sticking out at one side. This was fastened with a single tack to the flattened end of the rod in such a position that the point lay exactly against the center of the disk. The disk could then be turned up or down, to cover or uncover the point of the rod, as desired. The rod was fitted snugly into a hole in the baseboard, and could be raised or lowered to any extent desired, but we had to provide some sort of an arrangement for making it stay where it was put. A small hole was drilled from the edge of the baseboard through to the hole in which the rod was fitted. A square socket was chiseled out around the small hole to receive a nut. The nut was firmly wedged in and held in place by driving in nails along the edges. A bolt or machine screw was threaded through the nut, so that its inner end pressed against the sighting rod. By tightening this screw the rod could be secured at any height desired.

The instrument was mounted on a tripod similar to the one used for our surveying instrument. To this it was attached by means of a bolt, which passed through the center of the baseboard and the tripod head.



The Screen.



The screen, or shutter, of the heliograph was mounted on a separate tripod, so as to prevent shaking the mirror when it was operated. It was made something like a window shutter. We cut out two slats, each 2-1/2 inches wide and 6 inches long. They were made of hardwood 3/8 inch thick. The upper and lower edges were tapered down to a thickness of 3/16 inch. Light nails were driven into the slats at the ends, and the nail heads were then filed off so that the projecting ends formed trunnions for the slats to turn on. The slats were linked to a connecting rod with double point tacks. A small double point tack was driven into the upper edge of each slat about 1/2 inch from the right hand end. Then through each of these tacks we hooked a second double point tack and drove it into the rod. The tacks on the rod were placed just 2 inches apart. A substantial frame was then made of 3/4-inch stuff 1-1/2 inches wide. The frame was square, with an opening that measured 6 inches each way, into which the slats were fitted. Before nailing the frame together we drilled holes in the side pieces for the trunnions of the slats to turn in. These holes were just 1-3/4 inches apart. After the slats had been set in place, the frame was fastened together and then nailed to a baseboard, which was fastened by a bolt to the tripod. The shutter was operated by a key something like a telegraph key. It was made of a narrow stick of wood hinged at one end to the lower strip of the shutter frame, and a spool sawed in two was fastened to the other end to serve as a handle for the key. A string connected the key with the connecting rod. The slats were kept closed by a spring, which was fastened at one end to the connecting rod and at the other to the top of the frame. At first we used a rubber band for this purpose, but it soon wore out, so we then made a spiral spring out of stiff spring brass wire by wrapping it around a pencil. When the key was pressed down the slats would be turned open, as shown in Fig. 159; but as soon as the key was released the spring would pull them back again.



Focusing the Instrument.



We were now ready to commence operations with our instruments. The heliograph was set up on the ledge at the top of the cliff. First the disk was turned down, uncovering the point of the sighting rod. Then Bill sighted through the unsilvered spot in the mirror and shifted the rod up and down until the tip end came squarely in line with the door of our straw hut, where Jack was seated, notebook in hand, to take down our message. Reddy stood by him with his wigwag flag to answer back. When the instrument was properly sighted the shutter was set up directly in front of it and the sighting disk turned up to cover the point of the sighting rod. Then came the rather troublesome task of focusing the mirror. The mirror reflected a square panel of light, in the center of which there was a small shadow spot made by the unsilvered peep hole. The object was to get this shadow to fall on the center of the sighting disk. We knew that then the mirror would reflect the sunlight squarely on the straw hut. We found it quite easy to direct this shadow spot to the disk by holding a sheet of paper in front of the mirror six or eight inches away, and following up the spot on the paper until it reached the disk.



Heliograph Signaling.



When at last we succeeded in properly focusing the mirror Bill pressed the key down three times, sending three quick flashes to Jack as a signal that he was ready to begin. Reddy wigwagged back O. K., and then the first heliographic message was sent from the ledge to the island. It was a rather mixed-up message, and kept Jim and Reddy wigwagging back and forth very strenuously to straighten matters out. It was my duty to keep the mirror focused. As the sun moved across the sky the shadow spot would move off the disk, and I had to keep shifting the mirror to bring the spot back where it belonged. We used the International Telegraph Code, which we had been studying every evening for a week, but it was many weeks before we learned how to use it correctly, even slowly. The International Telegraph Code is as follows:

A . - B - . . . C - . - . D - . . E - F . . - . G - - . H . . . . I . . J . - - - K - . - L . - . . M - - N - . O - - - P. - - . Q - - . - R . - . S . . . T - U . . - V . . . - W . - - X - . . - Y - . - - Z - - . . 1 . - - - 2 . . - - - 3 . . . - - 4 . . . . - 5 . . . . . 6 - . . . . 7 - . . . 8 - - - . 9 - - - - . 0 - - - - -

The three short flashes Bill sent represented the letter S, which stood for the word "signal." A was formed by a short flash followed by a long flash; B by a long flash followed by three short ones, and so on. The key was held down three times as long for the long flash as for the short one. We found the best way of learning to send the signals properly was to count 1 for each short flash, and for each pause between parts of the letter, and 3 for each dash and for each pause between letters. Between words we counted 6. Thus, for the letter A the key would be down when we counted 1, up when we counted 2, down while we counted 3, 4, 5, and up while we counted 6, 7, 8, for the pause after each letter. It was rather a confusing code, I admit, but in time we mastered it, all but Reddy and Fred, who never would learn, but instead used the wigwag code, letting a short flash stand for 1, a long flash for 2 and a double long flash for 3.



The Double Mirror Instrument.



Our heliographing instrument did excellent service sending flashes from the cliff to the island, but we couldn't make it work very well sending messages from the island to the cliff, because we had to face almost due north, and then the sun was nearly always at our backs and couldn't shine squarely on the mirror. This led to our building a double mirrored heliograph the following summer. To begin with, we built an instrument which was the exact duplicate of our first heliograph; then, in addition, to fit in the socket of the sighting rod, we rigged up a second mirror, which was mounted in exactly the same way as the first. The second mirror was called the station mirror, and differed from the other, or sun mirror, in having a small patch of white paper pasted at the center instead of a peep hole. When using this instrument, we set it up so that the station mirror faced the ledge, then by sighting through the hole in the sun mirror at the reflection in the station mirror we could see just what was in focus. The station mirror had to be moved until the patch at its center hid the ledge from view. After that the sun mirror was shifted until the shadow spot fell on the white patch of the station mirror. When once the station mirror was focused, it could be clamped tightly in place by screwing up the trunnion and swivel nuts. But the sun mirror had to be constantly shifted to keep the shadow on the patch. Another way of focusing the mirrors was to stand behind the instrument with the head close to the station mirror, shift the sun mirror until the entire station mirror was reflected in it, with the white patch squarely over the unsilvered spot; then still looking at the sun mirror, the station mirror was shifted until the reflection of the distant station was brought squarely in line with the unsilvered spot on the mirror. The station mirror was now firmly bolted and the sun mirror adjusted until the shadow spot fell on the paper patch.



CHAPTER XIV.

ICE BOATS, SLEDGES AND TOBOGGANS.

As our vacation was drawing to a close, we began to make plans for the Christmas holidays. Our previous Christmas vacation had been so completely taken up with preparations for the trip to Willow Clump Island that we had had no time for the trip itself. We resolved this time to have everything ready beforehand, so that we could spend the entire two weeks in solid pleasure. Our skate sails and snow shoes were stored in the attic, ready for use. If we were to make a trip in the snow we would need a sledge, and then, too, we wanted to make an ice boat. It would hardly pay to build these on the island and then cart them home, so it was decided to break up camp a couple of weeks before school commenced.



Breaking Camp.

Consequently, on the first day of September we gathered up our belongings, corraled our chickens, packed our goods, and the next day started for home. Mr. Schreiner, in response to a letter from the secretary, came down with a large wagon in which the majority of the things were packed. The rest of our luggage was stowed in the scow and the canoes, and these were towed down the canal, as before. We reached home late in the afternoon, tired and hungry. It was a treat to sit at the table again and eat some of Mother's appetizing dishes. And say, wasn't that pie great, though! My, how ravenous we were! And then a soft, comfortable bed with spotless white sheets and pillow cases. How soundly we did sleep that night! You can just bet we were all glad enough to get back to civilization, though, of course, no one could have dragged out the confession from a single one of us.



The Ice Boat.



School commenced on the 20th of September that year, so we hadn't much time to spare. Work was begun immediately on the ice boat. Our first ice boat was rather a crude one. A 2 by 4 inch scantling 14 feet long was used for the backbone of the boat. The scantling was placed on edge, and to lighten it and improve its appearance it was tapered fore and aft from a point 4 feet from the bow end. The thickness of the ends of the backbone was but 2 inches, as shown in Fig. 163. To the under edge of the backbone, 5 feet from the forward end, a crosspiece was nailed. This crosspiece was a 1-inch board 6 inches wide and 9 feet long. Braces were then run from the ends of the crosspiece to the forward and rear ends of the backbone, and at the rear end several boards nailed across the braces served as a seat for the boat.



Our next task was to rig up the runners. For these we used skates, which were so arranged that we could remove them whenever we wanted to. Three blocks of wood were used for the runner shoes. Two of them were cut from a 2 by 4 scantling and measured a foot in length. The third block was only 1 inch thick, but was otherwise of the same dimensions. The skates were laid face downward on the blocks with the clamping levers open; then we marked the places where the clamping jaws touched the wood and drilled holes at these points. The forward end of each block was also tapered off to fit flat against the face of the skate. Then by inserting the jaws in the holes and closing the levers, the skate was clamped to the block, just as it would be to a shoe. The two 2-inch blocks were bolted to the ends of the crosspiece, but the third block needed further attention, as it was to be used for the rudder or steering runner.

The rudder post was shaped from a block of hardwood 3 inches square and 10 inches long. Two inches from the lower end saw cuts were made in the side of the block to a depth of 3/4 inch. Then with a chisel the sides were split off, forming a large pin with a square shank 8 inches long. Next the corners of the shank were cut off, rounding it to a diameter of 1-1/2 inches. The runner block was fastened securely to the head of the rudder post with screws. A 1-1/2-inch hole was now drilled into the backbone at the stern end to receive the rudder post. A tiller was next cut out of a 1-inch board to the shape shown in Fig. 167. A slot was cut in the end of the tiller, and the latter fitted snugly over the top of the post, where it was held in place by screws threaded in through the sides.



The mast of our boat was a pole 8 feet long, tapering from a diameter 2 inches at the base to 1-1/2 inches at the top. A step for the mast was cut from a 2 by 4 block 8 inches long. A 2-inch hole was drilled into the face of this block. We had no drill large enough to bore this hole, but accomplished the same result by drilling eight 1/2-inch holes inside of a 2-inch circle (Fig. 168), and then used a chisel to cut off the projecting pieces. The mast step was firmly bolted to the backbone at its thickest part, that is, just four feet from the forward end. The mast was braced with stay ropes stretched from the top to the forward end of the backbone and to the ends of the crosspiece. A 9-foot pole, tapering from 1-1/2 inches to 1 inch in diameter, was used for the boom of the mainsail, and for the gaff we used a 6-foot pole of the same diameter.



The dimensions of the mainsail are given in Fig. 169. For mast hoops we used curtain rings. Five were attached to the sail along the luff, and one was fastened with a piece of leather to the end of the gaff. We used a different scheme for holding the boom to the mast. The forward end of the boom was flattened at the sides and a couple of cheek blocks were bolted on, forming jaws of the shape indicated in Fig. 170. The jaws were whittled out to fit nicely around the mast, and were kept from slipping off by a piece of rope passed around the mast and threaded through the ends of the cheek blocks. Half a dozen small pulley blocks were now procured, of the type used on awnings. A rope called the throat halyard was strung from the throat or forward end of the gaff through a pulley block near the top of the mast, and led down to the backbone, where it was "belayed," or wrapped around a cleat. The cleat, which was whittled out of a stick of wood, was made in the form indicated in Fig. 171. A short length of rope was strung through a pulley block and tied with some slack to the upper end and to the center of the gaff. This rope is called a "bridle," and to the pulley block on this "bridle" a rope was attached called the "peak halyard." The peak halyard was passed through a pulley block at the top of the mast, and belayed on a cleat at the side of the backbone. For the main sheet (that is, the rope used for guiding the mainsail) two pulley blocks were fastened to the backbone, one just in front of the seat and the other a few feet further forward, and two more were lashed to the boom, midway between these blocks. The sheet was fastened near the aft end of the backbone and then strung through the blocks in the order illustrated, the free end of the sheet being brought back to the seat, where a cleat was provided, to which it could be secured when desired.



The jib-sail was now cut out to the dimensions given in Fig. 172. The foot of the sail was lashed to a jib-boom 3 feet 4 inches long. The jib-boom was attached to the backbone at its fore end by means of a couple of screw eyes. The eye of one of these was pried open, linked through the other and then closed again. One of the screw eyes was now screwed into the head of the jib-boom and the other was threaded into the end of the backbone. The upper corner or "head" of the jib was tied to a jib-halyard, which passed through a block at the top of the mast, and was secured on a cleat on the backbone. On the jib we used two sheets. They were attached to the end of the jib-boom and passed on opposite sides of the mast through blocks on the crosspiece to the stern of the boat, where separate cleats were provided for them.

This completed our ice boat, and a very pretty little boat she was. It was with great reluctance that we furled the sails, unstepped the mast, and stowed away the parts in our attic until old Jack Frost should wake up and furnish us with a field of smooth ice.



The Sledge.



Our sledge was patterned after a picture of one used by Peary in one of his Arctic expeditions. First we got four strips of hickory 1 inch thick, 1-1/2 inches wide and 8 feet long for the runners and side rails. Beginning 18 inches from the ends, each stick was tapered gradually to a thickness of 1/2 an inch. Then we made eight spreaders or spacing blocks, each 1-1/2 inches thick, 2-1/2 inches wide and 11 inches long. In each end a notch 1/2 inch deep was cut to receive the runners and side rails. In the edge of each block, midway of its length, a slot 1 inch deep was cut to receive the cross sticks of the sledge. First we nailed the runners and rails to the blocks, fastening them with screws, spacing the blocks 16 inches from the ends, and 20 inches apart from center to center. Then we bent the ends of the rails and runners together, fastening them with bolts, as in Fig. 175. Four crosspieces, or floor beams, were cut out of a 1-inch board, each 2 inches wide and 30 inches long. These were fitted into the slots in the space blocks and secured with screws. A cross stick was also fastened between the rails and runners at the forward end. On the floor beams we nailed a flooring of 1/2-inch slats, 2 inches wide and 6 feet long. At the rear end these slats projected 8 inches beyond the last space block and over them a cross slat was nailed. A stick of hickory 4-1/2 feet long was soaked in hot water, as described on page 39, and was bent to an U-shape. The ends were then fitted over the first cross stick, and under the first floor supports, and securely nailed in place. Another stick of hickory 6 feet long was similarly bent, and the ends slipped over the rear cross slats and fitted against the rear space blocks, in which position the stick was securely nailed. It was our intention to shoe the runners with strips of brass, but these were not procurable in our village, and we had no time to go down to Millville. However, the village blacksmith came to our rescue and shod our sledge with sleigh runner iron.

We had planned to make two more devices for our winter sports—a toboggan and a peculiar looking contrivance called a "rennwolf," a picture of which Dutchy happened to unearth in one of his father's books. Unfortunately Bill and I had to return to school before either of these was completed. However, the work was entrusted to Reddy, who was quite handy with tools, and Jack, who was made secretary pro tempore, took notes on the work.



The Toboggan.

The toboggan was made of light flexible hickory boards, 1/4 of an inch thick, 6 inches wide and 8-1/2 feet long. Three of these boards were used, and they were fastened together with cross sticks or battens, about 3 inches wide and 1/2 an inch thick. There were six of these battens spaced about 15 inches apart, and secured to the floor boards with flathead screws introduced from the under side and countersunk so that the heads would not project below the bottom of the toboggan. At the forward end we screwed on a head piece of oak, 3/4 of an inch thick, 1-1/2 inches wide and 20 inches long. The head piece was fastened to the under side of the boards, so that when they were curved up into a hood it would lie on top. The ends of the head piece, which projected 1 inch each side of the boards, were notched to hold the rope, which was tied fast after the boards had been steamed. The boards were steamed by wrapping them in burlap for a distance of 2 feet from the forward end, and pouring boiling water over them, as was done with the snow shoes (page 39). Before bending the boards we had fixed screw eyes in the ends of each batten, except the forward one; a rope had been strung through these screw eyes and the ends were now tied to the head piece and drawn tight so as to bend the boards into a graceful curve. In this way the ropes were of service not only for curving the front end into a hood, but also for side rails, to hold on by when shooting swiftly around curves.



The Rennwolf.



The runners of the rennwolf were made of hickory strips, 1 inch thick, 2 inches wide and 8 feet long. At their forward ends these strips were tapered down to a thickness of 1/2 an inch and curved upward. About 30 inches from the rear end of each runner an upright post was nailed. The post was 3 feet long and was braced by a diagonal brace 24 inches long, as shown in Fig. 179. A tie bar was nailed to the post about 6 or 8 inches from the bottom and connected with the forward curved end of the runner.

The two runners were now placed parallel to each other about 18 inches apart, and connected by four cross bars, one at the forward end, and three on the upright posts, in about the positions illustrated. The upper cross bar was extended 6 inches beyond the posts at each side, and served as a handle for guiding the queer craft. An 18-inch square board was used for the seat of the rennwolf. It rested on the second cross-bar of the post about 12 inches from the runners, and the forward end was supported on legs nailed to the tie bars. On each runner back of the posts a loop of leather was nailed, large enough to receive the toe of one shoe.

When using this odd sled one foot would rest on the runner with the toe in the strap, and by kicking out against the snow or ice with the other foot the rennwolf would be made to spin along at a rapid rate. Of course, when coasting both feet would rest on the runners and the sled was steered by an occasional side push at the right or left. Owing to the great length of the runners the rennwolf would easily ride over uneven surfaces and thin spots in the ice.



Ice Creepers.

In order to provide a better hold for the propelling foot, we fastened around the toe a strap of leather, through which a number of long tacks projected. Their sharp points would stick into the ice, and prevent the foot from slipping. The seat of the rennwolf was convenient for carrying a coat or any light luggage, and it was often used to give a friend a very exhilarating ride.



CHAPTER XV.

THE SUBTERRANEAN CLUB.

I am afraid we were not very glad to get back to school that fall. It seemed very hard to give up the sport we had been having, and our heads were brimful of new schemes which we could hardly wait to put into practice. But we soon learned that there are many things that could be done during recreation hours at school. We had intended building a cave on our island that summer, but our vacation came to an end before we got around to it. There seemed no reason why we shouldn't dig one in the woods at the back of the schoolhouse.



A Cave-in.

Bill had read somewhere that if you dig a cave under a tree the roots of the tree will support the ground on top and make a natural and substantial roof. It sounded very reasonable, we thought; in fact, we never questioned the truth of the statement, because we had somehow gotten the notion that books were never wrong, and that whatever was set up in type must surely be so. But events proved that the man who wrote that book had never attempted to build a cave in the manner he described, at least not in the loose, sandy soil of south Jersey. A large spreading cedar was selected as the tree which should support the roof of our cave. It was situated on a mound at the edge of the woods. First a passageway, or ditch, was dug at the bottom, and then we begun tunneling in the side of the mound under the roots of the tree. For a while the ground above held, and our tunnel had reached a length of about four feet, when suddenly, without the slightest warning, the sandy soil gave way and we were engulfed. Bill, who was furthest within the cave, was almost entirely covered, while I was buried to the shoulders. A crowd of boys came to our assistance and dug us out. Poor Bill was almost smothered before they scooped the sand away from around his mouth and nose. The boys made slow work of it, having to dig with their hands and a couple of shingles, because the two spades we had were buried with us at the bottom of the cave.

Of course, this little episode gave us a scare, but it was only temporary. We swore everyone to secrecy, so that Mr. Clark, the principal, wouldn't hear of the mishap and suppress any further cave building. It was obvious that the only roof we could depend on for our cave would be a wooden roof. If we had been at Willow Clump Island we would have gotten any amount of slabs from the lumber mills across the river.

One of our schoolmates, a day scholar, came to the rescue. His name was Chester Hill, a little bit of a chap, about the shortest for his age that I have ever seen. His name was so at variance to his size that we called him "Hillock," for short. Now Hillock lived on a farm about eight miles from school, and used to drive in every day on a farm wagon. He had helped us dig the cave under the cedar tree, and when he learned that we would need some lumber to build a safe cave, he told us that he had an uncle who owned a lumber mill on the Morris River, from whom he was sure we could get all the slabs we wanted. Of course, we were delighted, and laid our plans for an elaborate cave house. Hillock promised to be on hand on the following Saturday afternoon with his load of lumber.



Excavating for the Cave.

We immediately set out to make the necessary excavation. The side of a bushy knoll was chosen as a suitable site. First we carefully transplanted the bushes that grew in the square we had marked out for the cave, and cutting the sod into squares, piled it all neatly to one side. Then we shoveled away the top-soil and heaped it up for future use. After that we dug away the sandy subsoil. The cave proper we planned to make about 8 feet by 10 feet, with a passageway 2 feet wide and 6 feet long, leading in from a large bush at the base of the knoll. Our excavation was therefore somewhat T-shaped (see Fig. 182). At the deepest part we had to dig down about 10 feet.



The digging was all done by Saturday, when Hillock pulled up with a big load of slabs. Slabs are a very unsatisfactory kind of wood for most purposes. Being the outside cut, they are usually very irregular and weak in spots. In many places they are almost clear bark. Of course, had our pocketbooks permitted, we would have used stout scantlings for the corner posts of our cave house and substantial boards for the walls, roof and flooring, but we had to be content with materials at hand. Eight of the best slabs were selected for our corner posts; four of them we cut to the length of 8 feet and the others to a length of 6 feet. The long slabs were set up at the rear of the cave, two at each corner, one flat against the rear wall, with its edge buried in the corner, and the other against the side wall, with its edge tight against the rear slab, as in Fig. 183. The same was done at the forward corners with the shorter slabs. A couple of slabs were now set up on each side of the passageway, and a corresponding pair against the rear wall. The upper and lower ends of the uprights were then connected with slabs, called stringpieces.



The sides were now boarded up with upright slabs nailed to the stringpieces. An opening 3 feet 6 inches high was left in the forward wall for a passageway. Several slabs were now placed on the edge across the bottom of the cave, to serve as floor beams, upon which a flooring of slabs was laid. Next the rafters were set in place, one on each upright slab. Slots were cut in the ends of the uprights to receive the rafters, which were slabs placed on edge. As the forward uprights were 2 feet shorter than the rear ones, the rafters were given a good slant, so that the roof would properly shed any water that might soak in through the ground above.



The roof was laid on the same way that we had made the roof of our tree house; that is, a slab was first nailed at the forward end of the rafters with its edge projecting far enough to make a good eave; then the second slab was nailed on, with its edge overlapping the first, and a third with its edge overlapping the second, and so on with the rest. At the rear end of the roof a hole was cut, into which we fitted a piece of stovepipe. We didn't plan to have a fire in the house, but set the stovepipe in place to provide the necessary ventilation. As the pipe had an elbow in it, there was no danger of rain or dirt falling through it. The upper end of the stovepipe was concealed among some rocks at the top of the knoll.

A suitable flooring was now laid in the passageway, and the sides were boarded up to a height of 2 feet from the floor at the entrance to a height of 3 feet 6 inches at the inner end. A roof of slabs was nailed on, and then we were ready to cover our slab house with dirt.



Covering the Cave.

We avoided piling on the dirt very deep, because there was danger of breaking in the roof with a heavy load. A thin layer of sand covered with the top-soil brought up the level to about that of the rest of the knoll. Then the sod was laid back in place and well watered, and the few bushes planted back in their original positions. Our sodding should have been done in the spring for best results. The frost soon killed the grass, and the bushes withered away. But a few cents' worth of grass seed was sowed in, and in time gave the knoll a very natural appearance. A bush at the bottom concealed the entrance of the cave, so that no one who was not in the secret would have suspected that beneath that innocent looking knoll were gathered the members of the "Big Bug Club."



The Big Bug Club.



Of course, we had to organize a secret society, to occupy our subterranean dwelling. In that I fear we overstepped the rules of the school. Of course, Mr. Clark knew of our cave, in fact he visited us there once, lowering his dignity sufficiently to squeeze into the narrow passageway, and playing Bill a game of chess at our club table. He seemed quite pleased with our work, and complimented us very highly on the masterful way in which we had built the underground house. We told him that we had organized a club of the older fellows to play indoor games and have occasional spreads, but we did not tell him that most of our spreads were held at the dead of night, when there was no moon and the stars were hidden by clouds. At 10 o'clock each night the bell rang for us to turn out our lights, and after that the six members would each, in turn, keep a half-hour watch, that is, first one would sit up and try to keep awake for half an hour, after which he would waken the next fellow, who at the end of a half hour would rouse the third, and so on, until 1 o'clock, when the sixth watcher would wake up the entire club. Then we would all creep out the back window in the hall, onto the roof of the rear annex of the schoolhouse, and thence climb down a rope ladder to the ground.



Midnight Banquets.

I suppose we could have just as easily have tiptoed downstairs and out the back door, but it would have spoiled the romance of it all. The absolute stillness and the pitch black darkness of the night were awe-inspiring. The roll of a pebble or the crack of a twig under foot would set us all a-tingle as we stole out to our cave house. Sometimes the night was so black that we could hardly find the entrance of the cave. Once inside, in the light of a few candles, the nervous tension was relieved, and we reveled in a banquet of cold victuals and dainties, purchased out of the monthly club dues. Our meetings in the cave lasted scarcely half an hour. In fact, the meeting, and even the banquet, were mere incidentals. The main enjoyment consisted in stealing out to the cave and back again, always at the risk of getting caught. Usually when we got to bed again we would be too excited to fall asleep right away, and when we did finally drop off our sleep was so sound that several times the breakfast bell caught one or more of us still napping.



The Club Pin.

The only other charm our secret club afforded was the wearing of a mysterious club pin. It was a silver beetle, with the letter G engraved on the head and the letter B on the body, while down the center of the back was the letter I (see Fig. 187). In public we called ourselves the G. I. B.'s, but it was only the initiated members who knew that these letters were to be read backward, and, with the beetle on which they were engraved, signified the "Big Bugs." Of course, we had some secret signs and signals, a secret hand grasp, a peculiar whistle as a warning to run, another meaning "lie still," and a third signifying "all is well."



The Combination Lock.

We found it necessary to close the entrance of our cave with a door fastened with a padlock, so as to keep meddlers out. The entire school had watched us build the cave house, and, of course, knew just where our entrance lay. Then, in addition to the outer door, we put in another one, half-way down the dark passageway. On this Bill rigged up a simple combination lock which would baffle anyone who managed to pick the padlock. This inner door opened outward. It was hinged to the floor of the passageway, and swung up against a frame set in the passageway. At the top was a board whose lower edge lay flush with the edge of the door when it was closed. For the combination lock we used a couple of spools, each with one head cut off and the central hole plugged up with a stick of wood. In the door and the top board of the frame, holes were drilled just large enough for the shanks of the spools to fit snugly in them. Next we made a trip to a hardware store for a file and a couple of large copper washers, about 1-1/4 inches in diameter. The washers were fastened to the inner ends of the spools after they had been pushed through the hole. The washer on the door came just to the edge of the door, while the other extended below the door frame and lapped under the door washer. Then in the edge of the washer on the frame a notch was filed, while in the other washer two notches were filed, so as to leave a tooth which fitted snugly into the notch of the first washer (see Figs. 188, 189). The door was locked by turning both the washers until the notch and tooth came in line with each other, then pushing the tooth through the notch, and turning the washers so that the frame washer hooked over the door washer. Then the door could be opened only when the tooth and notch were brought in line.

On the head of each spool we pasted a disk of white cardboard, the edge of which was graduated, as in Fig. 190. Then we had a secret combination, say 11-19, which meant that when the spools were turned so that the number 11 on the door spool came in line with the number 19 on the frame spool the tooth and notch would be in line, and the door could then be opened. Of course, this combination was known to the members of the club only, and anyone outside who tried to open the door might have tried for some time without bringing the tooth and notch into line with each other. Occasionally we changed the combination by loosening the screws which held the washers, and turning them so that the notch and tooth came opposite different numbers on the dials. This was done so that if anyone should chance to learn our combination he could not make use of it very long.



CHAPTER XVI.

SCOOTERS.

"Hello, Dutchy! What in thunder have you got there?" It was Bill who spoke. We were on our way home for the winter holidays, and had been held up at Millville by Reddy Schreiner, who had informed us that Dutchy was down by the river with the boat to give us a sail up to Lamington.

A vision of a fleet ice boat skimming up the river at express train speed swam before our eyes. But the next moment, as we turned the corner into River Street, we were surprised by the sight of our old scow just off the pier at anchor, and in open water. It was rigged up with a jib and mainsail, which were flapping idly in the wind. It had also been altered by decking over the top, with the exception of a small cockpit, evidently for the purpose of keeping out the water when she heeled over under the wind. We were disappointed and quite annoyed at not finding the ice boat on hand; furthermore, our annoyance was considerably heightened by Dutchy's broad grin of evident delight at our discomfiture. "The river wasn't all frozen over," he explained, "and we couldn't bring the ice boat down, so we rigged up the scow and she came down splendidly."



A Sail in the Scow.

There was nothing to do but to jump in, though I, for one, would have taken the train in preference had there been one inside of two hours. Dutchy, however, seemed to be in a surprisingly good humor, and kept up a lively chatter about things that the club had made in our absence. The skis, which have already been described on page 42, had been built under Reddy's guidance, and they had already used them on Willard's Hill, coasting down like a streak and shooting way up into the air off a hump at the bottom. Then there was the toboggan slide down Randall's Hill, and way across the river on the ice.



Our Craft Strikes the Ice.



Dutchy talked so incessantly that we hadn't noticed the field of ice which we were nearing. Just at this point Bill turned around with an exclamation.

"Here, Dutchy, you crazy fellow, where are you going to? Hard to port, man—hard aport—or you will crash into the ice!"

But Dutchy only grinned nervously.

"I tell you, you will smash the boat!" Bill cried again, making a dive for the steering oar; but just then the boat struck the ice, and both Bill and I were thrown backward into the bottom of the boat. But the boat didn't smash. There was a momentary grinding and crunching noise, and, much to my surprise, I found that the old scow had lifted itself clean out of the water, and was skating right along on the ice. Then Dutchy could control himself no longer. He laughed, and laughed, as if he never would stop. He laughed until the steering oar dropped from his hands, and the old scow, with the head free, swung around and plunged off the ice ledge with a heavy splash into the open water again. Then Reddy, who was almost equally convulsed, came to his senses. "Now you've done it, Dutchy; you're a fine skipper, you are! How do you expect to get us back to shore again?" The steering oar was left behind us on the ice, and there we were drifting on the open water, with no rudder and no oar to bring us back.



The Scooter Scow.



The only thing we could do was to wait until the wind or current carried us to the ice or land. In the meantime Dutchy, who had suddenly sobered down when we took our water plunge, explained how he had rigged up the scow to travel both on ice and on water. He called the rig a sled boat, but the name by which such a rig is now known is a "scooter." It was Dutchy's idea primarily, but Reddy had engineered the work. Along the bottom of the scow two strips of hickory had been nailed to serve as runners. The hickory strips had been bent up at the forward end, as shown in Fig. 191. Each runner was shod with a strip of brass, fastened on with flathead screws, which were countersunk, so that the heads should not project below the brass. This virtually made a sledge out of the old scow, and didn't spoil it for use on the water.



A Sprit Sail.



A sprit sail and jib were rigged up. The dimensions of these sails, which were taken from a book in Mr. Van Syckel's library, are given in the illustrations. A sheet of heavy muslin was made to measure 7 feet square, as indicated by dotted lines in the drawing; then the corners were cut off along the full lines shown in the illustration. The edges were now hemmed all around, and the lower edge of the sail was lashed to a boom, 7 feet 6 inches long. To the luff were attached a number of mast rings, which were slipped over a stout mast projecting about 5 feet 6 inches above the deck of the boat. The peak of the sail was held up by a spar called a sprit. The sprit was sharpened at each end, and the point at the upper end was inserted in a loop of heavy cord fastened to the peak of the sail, while the lower point of the sprit rested in the loop of a rope on the mast, called a "snotter." The snotter was a short piece of rope with a loop at each end. It was wrapped around the mast, as shown in the drawing, with one loop holding it in place, like a slip knot, and the other supporting the end of the sprit. A single halyard was used to raise this sail. It was attached to the boat and passed over a block in the mast. When raising the sail it was first partly hoisted, then the sprit was hooked in the loop and the snotter, after which the throat halyard was drawn taut. Then the snotter was pulled up the mast as far as it would go, flattening out the sail. The jib-sail was made out of the large corner piece left when cutting the mainsail. The dimensions of the jib-sail are given in Fig. 194. It was such a small sail that no boom was used with it. In place of a rudder the steering oar had to be used. This was made of a rake handle with a large trowel blade fastened to the end of it. The sharp blade cut into the ice, and so steered the scow when it was running as an ice boat, and in the water the blade offered sufficient resistance to act as a rudder.



Scooter Sailing.

But to return to our sail home to Lamington, we were not out on the open water long before the current carried us back to the ice ledge. Reddy jumped off and soon returned with the steering oar; then we proceeded on our way homeward, now in the water and now on ice. Once or twice the scow was unable to climb out of the water, because she had not sufficient headway, and was clumsy and heavy with four boys aboard. Then we had to push off until we could get a sufficient start. It struck me that while Dutchy was quite clever to think of such a rig, yet it was very clumsy and capable of much improvement. Bill wasn't saying very much all this time, and I could see he was doing a lot of thinking. Evidently he was planning some improvement, but Bill was a very considerate fellow, and did not want to spoil Dutchy's pleasure just then by telling him how much better a scooter he might have built. It wasn't until after supper, when a meeting of t he S. S. I. E. E. of W. C. I. was called, that Bill carne out with his scheme.



A Meeting of the Society.

"Why not mount the sailing canoe on runners, instead of the scow? You would have a very light rig then, and it would sail like a streak."

"Mr. President," said Reddy, "your plan sounds first-rate, but how are you going to fasten runners onto the canoe?"

"I've thought all that out," replied Bill. "If we can only get hold of a pair of sleigh runners it won't take long to rig up the sled boat."

Dutchy, who had looked rather crestfallen at a suggestion of an improvement on his pet invention, now suddenly brightened up.

"I know where we can get the sleigh runners!" he exclaimed. "Dad has an old ramshackle sleigh in the barn that is just falling to pieces with dry rot. I'll ask him for it to-night."

"Do you think you can get it?" inquired Bill.

"I guess so," Dutchy answered, rather doubtfully. "But say, suppose we send a delegation to see him about it?"



An Interview with Mr. Van Syckel.

This was agreed upon, and in the morning, as soon as breakfast had been downed, the entire society marched in a body into Mr. Van Syckel's library. I was appointed spokesman, with Bill to back me, while the rest of the party were strung out behind, with Dutchy bringing up the rear. Mr. Van Syckel was not the man to take much interest in boys' work, but we happened to strike him at the right moment, and before our interview was over we had told him all our experiences of the summer before and all our plans for the future. Then we did a good turn for Dutchy, too. Mr. Van Syckel had always considered his boy a "know-nothing," and was very much surprised to find that he had invented the scooter scow. Why, he actually seemed proud of his son, much to Dutchy's embarrassment. After that there was no trouble about getting the sleigh runners, and Mr. Van Syckel forgot the objections he had offered at first.



The Scooter Canoe.



Naturally we were very much elated at our success, and straightway made for the barn, where we began operations on the scooter canoe. The sleigh was an old-fashioned affair, with rather broad wooden runners. First we removed the body of the sleigh, and then the runners were cut down to a height of about 15 inches. We spaced them apart about 28 inches, and connected them with four crosspieces at the top. The runners were now placed over our larger canoe, with forward ends about on a line with the mast, and the crosspieces were fastened with screws to the gunwales. As an additional security, a pair of crosspieces were now run under the canoe at each end and fastened with screws to the keel. At the bow the keel was shod with a strip of brass. The rudder was taken off the boat, and an oar lock was fastened to the stern to hold the steering oar. In place of lee boards we nailed a couple of thin boards over each runner, as shown in the drawing. We were in a hurry to finish this, as our vacation was short, so we used on the scooter canoe the sails that we had made for our ice boat. This required a bowsprit, but as we had little time to spare we used the jib-boom of the ice boat, nailing it to the deck beam of the canoe. We decided that the jib-sail could be used without a boom, as we had done with the scow. The mast was braced by stays attached to the ends of the runners and bowsprit. This spread of canvas was far greater than that originally provided for sailing the canoe, but the heavy runners on each side helped to keep the boat on even keel, and then to further balance the sail a board was nailed across the aft end of the boat. This overhung the runners about 18 inches each side, and in a strong wind we could sit out on the windward end of this board, thus preventing the scooter from heeling over too far.



CHAPTER XVII.

AN ARCTIC EXPEDITION.

As soon as our scooter canoe was completed we prepared for the long-planned winter expedition to Willow Clump Island. The weather conditions were ideal. We had had ten days of steady cold weather, which had followed a heavy fall of snow, so that we could tramp up the island on snow shoes, or we could use our scooter canoe and scooter scow on the river. It was out of the question to use our skate sails or the ice boat on the river, and the canal would be serviceable only in case the wind should blow from a southerly quarter. But we stowed them on the sledge for use on Lake Placid.

On the Tuesday morning following Christmas we made the start. Bill in the scooter canoe and Dutchy in the scooter scow sailed up the river, and the rest of us, on snow shoes, took the tow path of the canal, hauling the sledge along. We carried provisions for a week and a good supply of blankets. The island was reached without mishap, except that Dutchy had to be helped several times in dragging the heavy scow around the rapids. Bill reached the island long before we did, and after unloading the canoe came racing back under a stiff breeze for a second load. Then he took his turn at hauling the sledge, while Reddy sailed the reloaded scooter canoe up to the island.



Willow Clump Island in Winter.

We brought no tent with us, as we expected to take up our quarters in the straw hut. When we reached the hut we hardly recognized it. It was almost completely covered with snow and looked like an Eskimo house. The snow had drifted well up over the north side, completely closing the entrance. We had to set to work at once with a shovel and open up a passageway, and then we had to shovel out a large pile of snow that had drifted into the hut from the open doorway.



Kindling a Camp Fire.

In the meantime Jack scoured the island for some dry wood. In this he was not very successful, because everything was covered with snow, and when he tried to kindle a fire in the open space in front of our hut he found the task an exceedingly difficult one. Unfortunately we forgot to bring the oil stove with us, and the prospect of something warm to eat was exceedingly remote. We hadn't yet learned the trick of building a camp fire in wet weather. After exhausting our stock of paper Fred and I started over to Lumberville for several newspapers and a can of kerosene. We went to old Jim Halliday's, who had befriended us on one or two occasions the previous summer, and made known to him our troubles.