In Fig. 404 we have two other diagrams yielded by shutters opening and closing at the center. The areas inclosed by the dotted lines correspond to the ideal unit of efficiency. We can see at a glance that the A shutter is much more efficient than the other, i. e., B form, which latter takes quite a comparatively long time to open as compared with the time it is fully open, but closes quicker than it opens. Examination of many shutters has led the lecturer to the conclusion that very few of them had more than about 65 per cent. efficiency.
When working the apparatus as above described, when the speed of the sectors and bars was a knowable quantity, there was no necessity to measure the speed of the revolving drum.
All that was required was a fairly steady motion, and that it be fast enough. This the lecturer easily attained for the demonstrated experiments by revolving the projected axis by means of the fingers or a piece of string in the same way that one may cause a top to spin. If, however, the measuring of the speed of the sectors be a difficulty or objection, the following device may be substituted : a vibrating (tuning) fork of known frequency has at tached to one of its arms a small reflecting mirror, which throws an independent point of light through the lens on to the face of the drum. When the fork is made to vibrate and the drum to revolve, the light from the small reflector traces a wave curve, and since the frequency of the fork's vibrations are known, this wave curve forms a time scale alongside the slit dia gram.
Numerous examples were shown and explained, one especially demanding mention being that of a shutter which, when set at a rather low speed, gave a fairly satisfactory record of its performances, but when set at higher speed clearly betrayed the fact that it rebounded to such an extent that a slight secondary exposure was made. Under such circumstances obviously it would not be likely to yield satisfactory negatives. This shutter acted satisfactorily at about second, but when its speed was about or the rebound exposure rendered it useless. A drop shutter, with elastic band, giving an exposure from start to finish of about second, yielded a diagram similar in shape to that in Fig. 403. but the time during which it was fully open was only about spa second. Generally speaking, the smaller the stop, the higher the efficiency.
The following notes on instantaneous shutters*, by Capt. W. De W. Abney, C.B., will be of interest. He says : " Perhaps on no pieces of practical apparatus have more labor and thought been expended than on instantaneous shutters. They are often constructed on very beautiful and ingenious designs; but, in many, something is wanting, and they are often far from what we may call ideal shutters, whilst others are as nearly perfection as may be.
The first question is as to the best position that the shutter should occupy. It is evident it may be on the front of the lens, at the back of the lens, or occupy a position close to the dia phragm; or, again, it may be placed close to the plate. Let us see what these several positions entail in regard to the general exposure of the plate. We will suppose for the moment that the shutter is of a guillotine type, and that it falls trom the top to the bottom, the exposure of the lens beginning first at the top, and finishing at the bottom. When it is in front of the lens, as the top part of the lens is that with which the image at the top of the picture is made, it is evident that the sky of a landscape is first exposed, then the middle distance, and finally the foreground near this base. Or let us take an example of the figure of a man moving across the view, and whose image occupies nearly the total height of the plate. The hat will be first exposed, then the face, next the trunk, and finally the legs, the last portion which is impressed being the feet. The photograph would not represent him in the position he occupied at any particular instant of time, but it would represent the position that his hat occupied at one instant, his face at another, and so on. Suppose the guillotine was a fairly narrow slit, which moved downward at a moder ate rate, it is evident that the movement of any small portion of the image would be small during exposure, though the motions gone through between the time the hat and the feet were exposed might be large. In fact, it might happen that the image on the plate would show the man as tumbling backward, and the relative position of the swinging arms to that of the legs actually reversed, and make the proportions of the figure absolutely untrue.