ESTIMATING EXPOSURES.
In view of the large number of factors, and the many sub-divisions of each, it would be extremely difficult to devise a scheme for estimating exact exposures, and but for the fact that all plates allow a cer tain latitude of error it would be im possible. Latitude, although usually greater in slow plates than rapid ones, is not by any means entirely dependent upon speed. Some rapid plates of the best quality possess greater latitude than 81,- in. by q. in. in the frame, and lay the strips on it, having rounded off the corners) so that they then have the appearance shown in Fig. 171. Now fasten strips ofr card A A across so as to hold in the strips,i in order that they may be pulled out one at A a time. This cross strip may be fastened byl driving tiny pins through it and betweenl the strips. Set at 4 ft. from a candle flame, having previously placed the frame with an ordinary half-plate dry plate inside in the position of the dotted lines and re placed the back. No. 2 strip is then pulled
out, and two seconds' exposure given ; No. 3, four seconds ; No. 4 eight seconds ; No. 5 sixteen seconds ; No. 6, thirty-two seconds ; and so on, doubling the exposure each time. On development in a normal devel oper, an image like Fig. 172 is produced, which will show that Nos. 1 to 3 are alike, after which the exposures show a gradual increase of density up to No. 8, when the densities will remain the same. If such a plate were cut in parts, and each developed a little more, the set of grada tions would move up as in Figs. 173 and 174. Now, as only about 6 to 1 can he registered on a print, this shows that modifying the length of development per not always quite the same, as experience will fully confirm.