Another very interesting piece of work which has been carried out at the Laboratory is that of measuring the speeds of different flashlight compounds. There are several methods of doing this. One of them is to touch off a charge of powder behind a sheet of opal glass covering a circular opening in an opaque screen, in front of which a metal disk, with a radial slit about one-fifth of an inch wide is revolved at a given speed. The disk is photographed at the moment the flash powder is fired and when the plate is developed the angle of the sector formed can be measured and the speed of the flash easily calculated.

The shutter testing apparatus at the Laboratory will make very accurate measurements, but is not so convenient as the motion-picture camera method which is extremely simple. The camera is set up in a comparatively dark room. It is operated by a motor and made to run at a constant speed, taking thirty-two pictures per second. The powder to be measured is ignited behind an opal glass screen and its intensity and duration recorded as faithfully as the smile of a movie favorite.

Figures 5 and 6 show the negatives obtained by such exposures. Figure 5 being the record of a fast burning powder, and Figure 6 that of a slow burning powder. If we measure the density - the amount of the silver deposit on each square and plot it off against time, we get the time-intensity curve shown in Figure 4, the curve with the short base representing the quick flash and the longer based curve the slower flash.

Fig. 4

Fig. 5

Fig. 6

Each square or picture represents a time interval of one thirty-second part of a second, so that the total duration of the flash may be taken as four thirty-second parts of a second, and nine thirty-seconds of a second respectively. In Figure 6 we have the slow flash showing nine exposures, indicating that the flash lasted nine thirty-seconds of a second. From Figure 5 we see that the flash had a duration of four pictures or four thirty-seconds of a second. A reference to the curves in Figure 4 shows that both powders reached their highest intensity almost immediately and during the second exposure. Owing to the limitations of the half-tone reproductions we cannot distinguish the depth of intensity in Figures 5 and 6, but they are clearly shown in the plotted curves in Figure 4. The last four or five pictures in Figure 6 show a rapid and marked falling off in light intensity. The greatest density - highest illumination - is reached in both cases in from two to three thirty-second parts of a second.

The effective speed of most flash powders is usually faster than one-twelfth of a second, though the speed may drop with age, especially if the powder becomes damp. In this connection, the meaning of the word "effective" will be clear on reference to Figure 6. It is seen that the last three or four images are very weak, compared with the second or third, that is to say, very little impression is made on the film during the later stages of the flash.

Working with a strong auxiliary light while focusing or combining flashlight with daylight and using a fast flash powder, both the stare before and the wink after the flash can be eliminated, and results can be produced that are quite as good as those made by daylight alone or any other illumination.