86 Lord Rayleigh on the Manufacture and 



should recommend 5x4 glasses, as a margin is convenient. 

 Even when, as in the collodio-chloride process to be presently 

 described, the plate for each grating is prepared separately, it is 

 convenient to perform the preliminary operations of cleaning 

 and albumenizing on larger pieces. The cutting of prepared 

 plates requires a little care. I place them face downwards on a 

 sheet of clean paper, make the diamond cuts on the back, and 

 then, before breaking, remove as much as possible of the glass 

 powder. As it is important to prevent any grit from getting 

 between the film and the engraved face of the original, I usually 

 brush the surface with a large camePs-hair bimsh kept scrupu- 

 lously clean. 



In the preparation of the plates I have used a considerable 

 variety of methods. The process with gelatine and bichromate 

 of potash described in my previous papers has decided advan- 

 tages ; but all my efforts to obtain a mastery of it have been 

 unavailing. Plates prepared to all appearance in precisely the 

 same manner, and even at the same time, turned out differ- 

 ently, while modifications purposely introduced seemed to be for 

 the most part without effect. It required a strong scientific 

 prejudice to hold the uniformity of nature in the face of so 

 much adverse evidence. The uncertainty of this method is pro- 

 voking, as some of the results are exceedingly good ; but I gave 

 up my attempts sooner than I might otherwise have done in 

 consequence of the discovery of another method by which most 

 of the advantages of the gelatine process, namely simplicity of 

 manipulation and brilliancy of results ; might be attained with 

 much less risk of failure. 



It is very possible that a photographer skilled in the employ- 

 ment of geiatine might succeed where I failed. In case any 

 such should wish to make the attempt, I will mention a few 

 points that seemed important. The solution of gelatine should 

 be carefully filtered. For thick liquids containing gelatine, 

 albumen, &c, the best filtering material that I know of is tow. 

 The tow should be cleaned from grease by boiling with soda and 

 subsequent washing, and a small plugget of it pushed with mo- 

 derate force into the neck of the funnel. Some arrangement must 

 be adopted for keeping the gelatine hot, or the operation will 

 hardly succeed. It is important that the coat of gelatine should 

 be even, for which object the glass must be free from grease, 

 and the plate on which the prepared glasses are put away to set 

 perfectly level. Even then a good deal depends on the manipu- 

 lation ; but this is soon learned. The uniformity of the coat may 

 be tested by the colour when the plate is placed on a sheet of 

 white paper and examined in a weak white light. By candle- 

 light the colour of weak bichromate of potash is scarcely visible. 



