range as any known, showing ranges greater than any obtained with paper envelopes in experiments at Hallet's Point and Willets Point. (2) That between the range of certain explosion and that of certain failure there is a very wide interval where the result cannot be predicted. (3) That a 10-pound primary charge will explode secondary charges with certainty within a range of 27 feet. (4) That there is little or no increase in the range of certain explosion by increasing the primary charge from 10 pounds up to 25 pounds. (5) That there is reason to believe that at long ranges secondary explosions occur, which are of a low order or only partial. (6) That the range of certain explosion is substantially the same for secondaries of Atlas A or of Atlas B when the primary is 10 pounds of dynamite No. 1. Application to the blast at Flood Rock.-Applying these facts to the case of the final blast at Flood Rock, it is my opinion that if primary charges of 10 pounds of dynamite are placed 25 feet apart along the galleries from which the holes are drilled, and the bottom cartridge in each hole is thin copper packed with dynamite and projecting from the hole not less than 4 inches, that the explosion of the primaries will cause the explosion of every charge in the mine with as much certainty as if these charges were themselves connected with the batteries. In this connection it is to be remembered that the 4 inches of dynamite projecting from each hole will itself weigh 1 pound, and will itself have a sympathetic range of 18 to 22 feet, neglecting entirely the effect of the rest of the powder in the drill hole, and that these holes are only about 5 feet apart. I should place the primaries 25 feet apart, and put adjacent ones on different circuits, to provide against a possible break in either, and put a fulminate cap in each secondary, to make certain of detonations of the first order. Summary of experiments.—The accompanying table gives a summary of the 170 results obtained with dynamite No. 1, inert, absorbent, and 75 per cent. glycerine. Charges that exploded are marked by a small x. Those that failed are marked by a small o. Where the result is not absolutely certain, but there are special reasons for believing that the charge exploded or failed, it is recorded by a capital X or O. Results that are uncertain are marked?. All the primary charges are dynamite No. 1 (inert absorbent, and 75 per cent. nitroglycerine). DETAILS OF EXPERIMENTS WITH DYNAMITES. Cartridge packing. The cartridges used were of the thinnest sheet copper obta nå ble (0.005 inches), 24 inches in diameter and 8 inches long. They were loaded with 14 pounds of dynamite No. 1 (inert absorbent, 75 per cent. nitro-glycerine), packed with the greatest cars and as compactly as possible by expe rienced workmen, who understood the importance of uniform packing for the work in hand. The charge occupied about 54 inches of the length of the cartridge, and no amount of care (I tried it myself) would make it any more compact. In every case a round stick inc:es by 5 inches was then driven into the center of the cartridge and left in it. The driving of this stick caused the top of the charge to rise slightly in the can, showing that it was packed as tightly as possible. Sealing. The cans were then sealed with one-half inch of wax (5 tallow, 10 beeswax, and 15 resin). Attachment.-When about to fire, the cartridges for the shot were fastened with eight or ten copper tacks to a green yellow pine sapling 4 feet long, the end of which had been trimmed to enter the month of the cartridge aud fit down tight on the wax. The sapling was then weighted with two bricks and suspended from a raft anchored in deep water by a line of proper length to give the desired submergence, which was 20 feet, except in shot X, where it was 10 feet. The raft. The one-pound primaries were fired from a raft consisting simply of two 45-foot sticks crossed at right angles and lashed together, the primary being sn8pended at the center and the four secondaries at measured distances on the radial arms. For the larger charges the raft consisted of a regular hexagon, with sides of 80 feet, braced for stiffness, having a hexagonal opening in the center about 60 feet across. Six radial ropes ran to the center of the opening and were there lashed together. The primary charge was suspended at the center, and sometimes as many as twelve secondaries at different points, so distributed as not to be within exploding range of each other. Location. The work was done about 2 miles below Jacksonville, on the Saint John's River, in water from 7 to 11 fathoms deep. Current. The current varying with the tide was sometimes quite swift (about 4 miles up to 4 miles), but its effect on the location of the charge was eliminated by casting the raft loose from its anchor when ready to fire, and allowing time for the raft to take the velocity of the water. To make certain that this expedient answered its purpose, in all shots after number XI the charges were slid down by means of a wire loop along a one-half inch iron rod 24 feet long, which was so fastened to the raft that it could swing freely in any direction. These rods, several feet of which stood above water, invariably stood plumb when properly swing as soon as the raft had properly got under way, showing that the method was sufficiently accurate. Evidence of explosion.-It was expected that the condition of the end of the sapling would show conclusively whether the secondary charge had exploded or not—that it would necessarily be broomed in case of explosion. In the general case this proved to be so, but many saplings were recovered that were not injured at all. Sometimes the can was still on, showing almost no sign of any action; at other times the can was very much corrugated; at others, almost destroyed, but still holding considerable quantities of unexploded dynamite. These cases had clearly failed to explode. But in many other cases the can had disappeared entirely, all the tacks having been pulled out or the copper torn loose from them, or else a ring of copper showing signs of violent action still remained around the end of the sapling, the latter being sometimes slightly damaged, at others perfectly intact, and even in one case having all the wax still sticking to it. It was thought possible the wax might be the cause of the difficulty, and a number of shots were therefore fired without wax, the dynamite being in contact with the end of the stick. But the same results continued. A variety of expedients were tried to get more evidence on this class of cases, such as making a safety line fast to the can to recover it, if it was simply knocked off the end of the sapling; but nothing conclusive was reached until shots XXVIII, XXIX, and XXX, and shots Atlas VIII and IX, which were fired with the charge of dynamite entering 2 inches into a 24 inch iron pipe. In several of these cases (five out of thirty-five), when the condition of the iron pipe proved that the charge had exploded, the end of the sapling was not damaged at all. It follows from this that failure to broom the sticks cannot be considered proof of failure to explode. In the thirty-five cases above referred to the wax between the powder and the sapling was 24 inches thick in some, and in others there was 24 inches of sand, and then half-inch of wax to fill the can; the five failures to broom the stick were of the first description. In all cases except those fired with iron pipe the wax was only one-half inch thick. Failures at short range. -The most remarkable results obtained are the failures with large charges at short range. Of these the one that failed in shot XV was recovered with the can intact. In shot XVII the can was all gone, but some wax and a half. teaspoonful of unexploded dynamite remained in a crack in the end of the sapling. In XXV all the wax and a half-teaspoonful of dynamite remained on end of stick. Finally, in XXVII, both cans were recovered; one had a large hole torn in it, the ean much corrugated, but containing about 1 pound of dynamite, and the other almost destroyed, but still holding about one-quarter pound of the charge. Respectfully submitted. GEO. MCC. DERBY, ENGINEER OFFICE, UNITED STATES ARMY, Respectfully forwarded to General John Newton, Chief of Engineers, Washington, D. C. The conclusions reached by Lieutenant Derby, which have been indicated in my reports from Jacksonville, are concurred in, viz: 1. That rackarock, when each cartridge is prepared with a 30-grain fulminate primer reinforced with an ounce of dynamite, may be regarded as available for the final explosion at Flood Rock. 2. That whatever explosive is adopted for filling the blast-holes, dynamite No. 1 prepared with a fulminate primer should be contained in the cartridges which project from the blast-holes into the galleries. 3. That with this arrangement the blasting charges need not be electrically connected; but they will be effectively exploded sympathetically by the electric explosion of a number of ten-pound charges of dynamite No. 1, placed along the axes of the galleries at intervals of 25 feet. WALTER MCFARLAND, REMOVAL OF PILGRIM ROCK FROM EAST RIVER, NEW YORK. ENGINEER OFFICE, UNITED STATES ARMY, New York, N. Y., March 21, 1885. GENERAL: I have the honor to submit the following report upon the removal of the rock off East Nineteenth street, in the East River, New York, known as "Pilgrim Rock," this name having been given to it in |