light which falls upon the subject, whilst the other end of the cylinder is unfastened and set swinging, each swing being half a second. One of the pointers is set to the number of seconds, and the other pointers are set to the various factors of plate number, subject number, and diaphragm, when the last pointer will indicate the exposure required. An improvement was intro-duced some years ago into this instrument in the shape of a small circle of pale blue glass, covering the bromide paper, which renders the matching of the standard tint, previously a difficulty sometimes met with, very easy. Similar instruments are now made in watch form, and so simplified as to be self explanatory. The degree on Watkins' exposure meter is often assumed to be equal to one and a half times the Hurter and Driffield figure.
Wynne's Exposure Meter. This is now made in watch form, and, like that previously described, depends for its datum of light intensity upon a standard tint obtained by exposing sensitive paper.
Decouduris Photometer. This instrument consists practically of a small brass circular box containing sheets of tissue paper in increasing thickness, and it was introduced commercially in 1888. The method of using this instrument is as follows : - The instrument is held with the right hand on the ground-glass of the camera, whilst the focussing cloth must be carefully wrapped round the operator's head, so as to exclude any light but that passing through the ground-glass. The operator now observes from the distance of normal vision the small aperture to the left of the photometer, where three small holes and one large one are seen. The milled head in the centre of the instrument is now turned till the three small holes are no longer visible, and the instrument is then removed, turned over, and a letter will be observed, which gives an index of the exposure required, as against the letter in the table of the instrument will be found the exposure. Weber (" Phot. Mittheil.," vol. xxv., p. 37) has pointed out that this instrument possesses a scale of brightness of 9-48 to 0.35, to which the exposure is theoretically inversely pro-portional - that is to say, the time of exposure should be from 1 to 27 - whilst Decoudun gives his scale of exposures from 1 to 750, which is quite sufficient to condemn it. Goerz, of Berlin has introduced an improved and simplified form, which, however, is also open to the same theoretical objections.