The rifle-light which resembles an ordinary rocket, is fired from a rifle and is designed for short-range use.  It consists of a steel cylindrical shell a few inches long fastened to a steel rod.  A parachute is attached to the cardboard container in which the illuminating mixture is packed and the whole is stowed away in the steel shell.  Shore delay-fuses are used for starting the usual cycle of events after the rifle-light has been fired from the gun.  The steel rod is injected into the barrel of a rifle and a blank cartridge is used for ejecting this rocket-like apparatus.  Owing to inertia the firing-pin in the shell operates and the short delay-fuse is thus fired automatically an instant after the trigger of the rifle is pulled.

Illuminating “bombs” of the same general principles are used by airmen in search of a landing for himself or for a destructive bomb; in signaling to a gunner, and in many other ways.  They are simple in construction because they need not withstand the stresses of being fired from a gun; they are merely dropped from the aircraft.  The mechanism of ignition and the cycle of events which follow are similar to those of other illuminating shells.

The value of such artificial-lighting devices depends both upon luminous intensity and time of burning.  Although long-burning is not generally required in warfare, it is obvious that more than a momentary light is usually needed.  In general, high candle-power and long-burning are opposed to each other, so that the most intense lights of this character usually are of short duration.  Typical performances of two flares of the same composition are as follows: 

Flare No. 1  Flare No. 2
Average candle-power            270,000       95,000
Seconds of burning                10            35
Candle-seconds                 2,700,000     3,325,000
Cubic inches of compound           6             7
Candle-seconds per cubic inch   450,000       475,000
Candle-hours per cubic inch       125           132

The illuminating compound was the same in these two flares, which differed only in the time allowed for burning.  Of course, the measurements of the luminous intensity of such flares is difficult because of the fluctuations, but within the errors of the measurements it is seen that the illuminating power of the compound is about the same regardless of the time of burning.  The light-source in the case of burning powders is really a flame, and inasmuch as the burning end hangs downward, more light is emitted in the lower hemisphere than in the upper.  The candle-power of the largest flares equals the combined luminous intensities of 200 street arc-lamps or of 10,000 ordinary 40-watt tungsten lamps such as are used in residence lighting.

It is interesting to note the candle-hours obtained per cubic inch of compound and to find that the cost of this light is less than that of candles at the present time and only five or ten times greater than that of modern electric lighting.