nvestigation it was found that some of the water that leaked into the mine through the drill-holes was strongly charged with sulphur, which rapidly attacked the copper cartridges. From the locations in which this effect was found to be the most pronounced it is believed that the sulphur came from the ashes used for filling in behind the sea wall in constructing the small island on which the plant was located, and possibly also from the wrecks of coal barges, many of which have been lost on the reef.

The trouble was discovered before any great damage was done, and a remedy was readily found by coating the cartridge with a wax made of resin, beeswax, and tallow, put on hot; but, although only 536 cartridges were seriously damaged, the question of the integrity of the cartridge was of such vital importance that it was deemed prudent to take out in all 7,759, both rackarock and dynamite.

The labor of withdrawing such a number of charges from the drillholes, dipping them in melted wax, and replacing them, besides being an exceedingly delicate operation, made an amount of extra work, at a time when all hands were fully occupied, that added immensely to the difficulty of the final preparations.

When the loading was completed, the charge consisted of 240,399 pounds of rackarock and 42,331 pounds of dynamite, distributed over the whole nine acres of the reef in proportion to the volume of rock plus one-fourth the volume of water overhead.

Where the rock was very hard it was allowed 25 per cent. extra, and where exceptionally soft a reduction of 25 per cent. was made. The number of cubic yards of rock to be broken by the blast was 270,716.7, so that the charge was at the rate of 1.04 pounds per cubic yard.

The charge at the final blast at Hallet's Point in 1876 was 0.79 pounds per cubic yard; but an increase was deemed necessary at Flood Rock, because the thickness of the roof was much greater, and because the amount of rock excavated from the galleries was 2.7 times less in proportion to the amount of rock to be broken by the blast than at Hallet's Point, and consequently the whole of the reef had to be lifted to allow it to break, instead of getting the breaking effect of a fall, as at Hallet's Point.

When work of this kind was first undertaken at Hell Gate it was believed to be possible to make an excavation below the reef capable of receiving all the débris from the roof and pillars, and so gain at once the required depth of water. But on account of the seamy and treacherous nature of the rock of which the Hell Gate reefs are composed it is found that when the excavation becomes very large an inordinate amount of expensive timbering has to be done, as the falling of a single large block of stone from the roof (a very common occurrence in ordinary mining and tunneling) might result in flooding the whole mine in a few minutes. Besides, as broken stone takes up much more space than the same amount of rock before it is broken (at Hallet's Point the ratio was 1.45), to make certain of having the required depth immediately after the blast it would be necessary to excavate at least twice as much rock from the mine as it was intended to break up by the final blast. The average thickness of the roof at Flood Rock was 18.86 feet.

As this kind of excavation is necessarily costly, the ultimate cost of the result produced would be exceedingly high. It was, therefore, early decided in the course of the work at Hallet's Point to excavate only a relatively small amount (0.78 yard per yard of roof), but to thoroughly break up the roof by the final blast and remove so much of the débris as might be necessary by dredging.

The expectations of the engineer were so well fulfilled in this case, and such good results were obtained with the dredging, that it was decided to make a still further saving of expense at Flood Rock by an extension of the same policy, and during the later years of the work the excavation was limited as much as possible, only such galleries being run as were necessary to admit of the drill holes for the explosives being properly distributed.

The excavation at Flood Rock, therefore, amounted to 0.296 cubic yard only per cubic yard broken by the final blast, and instead of any greater depth on the reef being looked for after the blast, it had been computed that there should be a shoaling averaging several feet over the whole reef.

In all the work done in Hell Gate the near proximity of tall buildings in New York City and on Ward's and Blackwell's islands had to be constantly borne in mind. It had already been well settled by the experience acquired at Hallet's Point, that a large factor controlling the shock transmitted through the ground is the size of the individual charges, and that no fear need be felt in firing even as large a blast as 288,936 pounds at Flood Rock, provided it be sufficiently distributed. It was, therefore, tried as far as possible to limit the charges to 25 pounds per drill-hole, though in a few cases charges as great as 37 pounds could not be avoided. The average charge was 22.5 pounds.

The drilling of holes of sufficient size to hold such charges, slanting upwards into the roof, was a work of exceptional difficulty that had to be met with the invention of special appliances; 13,023 such holes were drilled, aggregating 113,102.15 feet in length. Only 12,561 of these holes were actually charged, a considerable number having to be rejected on account of the pressure of the water flowing in them being so great as to prevent their being used.

The arrangements for firing simultaneously this large number of charges were reduced to great simplicity. Elaborate experiments (reported last year) were made early in the season to determine with certainty to what extent sympathetic explosion could be relied on in firing the mine, and it was decided to dispense with connecting wires to the drill-holes entirely, it having been ascertained that the explosion of a charge of 10 pounds of dynamite under water will fire with certainty any number of other charges of dynamite packed solidly in thin elastic envelopes within a radius of 27 feet. Primary charges of 10 pounds of dynamite were therefore placed at intervals of 25 feet along the galleries from which the holes were drilled, and these primary charges only were connected with the batteries. There were 591 of them, arranged on 24 different circuits.

They consisted of two 24 inch by 24 inch thin copper cartridges solidly packed with dynamite, lashed on horizontal timbers from 3 to 12 feet above the floor, according to the height of the gallery; each cartridge also containing a 30-grain fulminate cap. On the top of the two cartridges a small primer was lashed, made of stiff brass, containing the platinum fuse fastened to the wires connected with the battery and half a pound of dynamite put in loose.

In locating the primary charges care was 'taken to have alternate charges on different circuits, so that in the event of a failure of any circuit its charges would nevertheless explode by sympathy, being within less than 27 feet of four other primaries on three other circuits, and part of the dynamite being solidly packed in thin elastic envelopes. The dynamite in the brass primers was put in loose, because in this shape it is but little affected by water, and it was difficult to make cer

tain of the tightness of the packing of the wires in the neck of the primer.

The circuits were made without joints, the wires being cut to the proper length and the platinum bridge soldered directly to them, and the gutta-percha insulation of the wire was carried over the whole fuze. Twenty-one of these circuits contained 25 fuzes, each connected with wires 52 feet long, of No. 18 copper insulated to No. 9, with lead wires at each end of the circuit, 600 feet long, of No. 12 copper, insulated to No. 9 A wire gauge. The remaining three circuits contained 22 fuzes, each connected with wires 106 feet long, and having lead wires of 950 feet; these circuits, intended for the extreme northern portion of the work, were nearly a mile in length. The bridges of the fuzes were of pure platinum wire 0.0005 of an inch thick. They had a resistance of 1.73 ohms cold and 2.76 ohms when heated to the exploding point. It required a current of 0.615 amperes to explode them when connected in series.

The number of fuzes and the amount of wire in each circuit were so adjusted that at the instant of explosion all should have the same reresistance, 77.43 ohms.

These values were determined in advance by the engineers of the work.

The circuits were made to order by the Laflin & Rand Powder Company, under the personal supervision of Mr. H. Julius Smith, who also made the fuzes for the Hallet's Point blast. Mr. Smith kindly volunteered to assist in laying the wires in the mine, and rendered valuable assistance in this delicate and tedious work. The wires were laid in about 26 hours, under the personal direction of Lieutenant Derby, by four gangs made up of the most intelligent of the miners.

The battery used in firing the mine was furnished by E. S. Greeley & Co., and consisted of 60 bichromate celis coupled in series. The plates were 6 inches by 9 inches, four carbon and three zinc in each cell, separated by only one-fourth of an inch. The electromotive force of the cells was between 1.95 and 2.07 volts; the internal resistance was less than 0.01 ohms. The battery was of sufficient power to send through every fuze a current just double that computed to be necessary to fire it. On account of the grave objection to having any joints in the wires under water, and the great advantage of having no resistance between the battery and the point of separation of the different circuits, the lead wires from each circuit led directly to the battery, which was located at the head of the shaft on top of the rock.

The arrangement of the circuit closer is shown in Fig. 5. Two large mercury cups formed the poles of the battery. One of these cups received the ends of the twenty-four lead wires, and the other was connected by a heavy band of copper with the mercury in a stout iron cup, in which sat a thin glass tumbler, also containing mercury. A stout cop. per strip connected the mercury in this glass with that in another mercury cup which received the twenty-four return wires.

The circuit was closed by breaking the thin glass, and allowing the mercury in it to mix with that in the iron cup. A long quarter-inch iron rod, terminating at its upper end in a small disc, was passed through a hole in the roof of the battery house, and stood up in the thin glass tumbler with its point resting on the bottom of it. A thirty-grain fulminate fuze was laid on the disc, and fastened to it with wax a few inches above the roof of the battery house, and was fired by means of a small battery at Astoria. The blow struck by the fuze on the disc and trans