parts by passing it through two equal lamp resistances, each of which, at the Pittsburg testing station, consists of a series of five 16-candlepower lamps. These leads are then, after independently traversing the cartridge file at the initial and final points, jointly connected to one of the poles of the primary coil of the induction coil through which the current passes to the return conductor. The secondary coil of the induction coil is then connected by one pole to the two platinum terminals and by the other pole to the base supporting the drum as described. As is well known, in the induction coil any change of tension in the primary coil sets up an induced current in the secondary coil, and this mutual induction between the coils results in the production of a higher potential difference at the terminals of the secondary one so that sparks of considerable length and intensity may be obtained.

The vibration tachometer, by which the speed of rotation of the drum is measured, is connected to an auxiliary shaft which engages the main shaft of the drum by gears, thus preventing any irregularity in recording the speed due to slipping. This tachometer measures the number of rotations of the drum, and as the circumference of the drum is accurately known, the distance which any point on the periphery travels may easily be calculated. Hence, at the highest speed of 105 revolutions per second, the distance of travel is 52.5 meters. At 50 revolutions it is 25 meters. At 86 revolutions it is 43 meters per second. With this number of revolutions it is possible with this instrument to measure the one-four-million-three-hundred-thousandth part of a second of time.

A more recent and simpler method of measuring the rate of detonation is that devised by M. d'Autriche,1 which was described at the congress in London, in 1909, by Dr. A. M. Comey, as follows:

The method of M. d'Autriche depends upon the use of a special detonating fuse having a uniform velocity of 6,000 meters per second. A suitable length of fuse, according to the length of the column of explosive to be tested is taken for the test and the exact middle of the fuse is determined by measurement and marked. A fulminate cap containing a charge of 15 grains (1 gram) is slipped over each end of the fuse and crimped securely. The fuse is then laid upon a piece of 32-pound sheet lead (1 inches by 15 inches by inch) (38 by 380 by 13 millimeters), so that the center of the fuse is about in the center of the sheet of lead, and the point coinciding with the middle point of the fuse is marked plainly on the sheet lead (M). The fuse passes along the entire length of the sheet of lead, and its ends are bent around so that they nearly meet. The two ends of the fuse covered with the detonating caps are inserted a short distance, at two points, into the column of explosive, the velocity of which is to be tested, and the distance between these points accurately measured. This may be called (A). A fulminate cap with fuse or electric connections is placed in one end of the stick of explosive. When this cap is detonated, the explosive wave proceeds through the detonating fuse in both directions and meets at a point

1 Comptes rend. 143, 641 and 144, 1030.

F-246

RPM. 37.5. DURATION OF FLAME 1539. MILLISEC. HEIGHT OF FLAME 50.21 IN. A BLACK BLASTING POWDER.

1

F-3/2

R.P.M. 2400. DURATION OF FLAME 342 MILLISEC. HEIGHT OF FLAME 1979 IN. A PERMISSIBLE EXPLOSIVE

(T) where a sharp line is formed by the effects of the detonation itself, and the lead is often broken through at this point. The distance from (M) to (T) Is accurately measured and designated as (b). If the two ends of the fuse are detonated simultaneously (M) and (T) fall together; that is, the detonation proceeds at the same rate through the two halves of the fuse and meets at the middle, but when a certain length of an explosive is placed in the circuit we have on one side one-half the length of the fuse and on the other side one-half the length of the fuse plus a certain measured length of explosive. We have thus, letting

X=Velocity of detonation of the explosive tested.

V=Known velocity of the fuse (6,000 meters per second).

A Distance between two ends of fuse, or length of explosive tested.
b Distance between M and T.

As to the accuracy of the test, it was found that by using the fuse alone (M) and (T) always coincided to within one-eighth of an inch (3 millimeters). It is easily seen that errors in measurement will be diminished by increase in the length of explosive tested, and it can be calculated, with velocities of 4,000 to 6,000 meters per second, using 15 inches (38 centimeters) of powder, that an error of one-fourth of an inch (6 millimeters) in measurement of the distance (M) to (T), which is a very large one under the conditions, introduces an error in the determination of the velocity of about 5 per cent.

Comey and his associates have tested this method quite fully at the eastern laboratory of the Du Pont Powder Co., and have found that it gives not only a ready and accurate means of determining the velocity of detonation through a column of any desired length of explosive, but that it is also possible by this method to determine the velocity with which a detonation wave travels through the air.

It is obvious that the flame-giving qualities of an explosive plays a most important part in its liability to ignite fire damp and other combustible mixtures, and that, all other conditions being equal, that explosive which gives the shortest flame for the briefest time is most suitable for use. Hence latterly much attention has been given to the study of the flames from explosives, and many devices have been constructed by which to photograph them.

Among these is the one employed at the Pittsburg testing station, where the flame is photographed on a moving film. The charge of explosive is fired from a "cannon" of the type used in the gallery tests by means of an electric detonator or igniter, but in this test the "cannon" is mounted vertically in a concrete foundation at a distance of about 18 feet from the lens of the camera. To cut off extraneous light rays, so that the tests may be made at any time, the cannon" is inclosed in an iron cylinder 20 feet in height and 43 inches in diameter, which is connected with the dark room by a light-tight iron conduit, as shown in plate 11. The cylinder, or

97578°-SM 1910-20

stack, is provided with a door in the side, through which the "cannon" can be loaded, and with a vertical slit 8 feet in length by 2 inches in width, which is so placed that its vertical center coincides with that of the conduit and also with that of the lens by which it is viewed. At the time of firing the top of the stack is covered with black paper. The conduit is closed at the point where it ends in the wall of the dark room by a shutter.

The camera consists of a drum on which the sensitized film is mounted, an electric motor by which the drum is revolved at a known rate, a quartz lens, a semicircular shield in which a stenopaic slit has been cut, and a shutter by which to control the slit in the shield. All of these except the motor are inclosed in a light-tight box. The semicircular shield is placed close to and concentric with the drum to prevent any light reaching the film except that passing through the stenopaic slit. A lens of quartz is used because it focuses not only the visible light rays, but also those invisible violet rays which occur to a large extent in the flames from explosives.

By means of a tachometer both the number of revolutions per minute of the motor and the peripheral speed of the drum are directly read off. The maximum peripheral speed of the drum is 20 meters per second, and this rate is employed when detonating explosives are tested, but with slow-burning explosives the drum is run at a slower rate. At the 20-meter rate 1 millimeter width of flame equals 0.05 millisecond of time, and as the measurements of the flame photographs are read to the nearest quarter of a millimeter the smallest time interval measured is the 0.0125 millisecond. The charge of explosive used in the test is 100 grams, and these charges are fired both with and without stemming.

The result of this test on black blasting powder and on a permissible explosive are shown in plate 12, figure 1. By the term “ permissible explosive" is meant an explosive which has satisfactorily passed all the prescribed tests at the Pittsburg testing station and is regarded as suitable for use in coal mines.

One of the most novel of modern tests is that devised by J. Thomas, who has employed the X rays for ascertaining the condition of the powder core in Bickford or running fuse. The cause of misfires and delayed ignitions has been the subject of much speculation, and among other theories proposed was that of a break in the continuity of the powder cores. In plate 12, figure 2, which is a copy of Thomas's X-ray picture, the interruption of continuity in two pieces of the fuse shown is very apparent.

1 J. Chem. Met. Soc. S. Africa 9, 183; 1908.