88 Lord Rayleigh on the Manufacture and 



tion may take an hour; but much depends upon the exposure, 

 and still more upon the temperature. The proper point to 

 which to carry the development can only be learned by experi- 

 ence ; but the beginner is most likely to err on the side of excess, 

 particularly if he uses pyrogallic acid. If, as is desirable, the 

 film be creamy and thick, the spectra of a candle do not appear 

 to advantage at this stage, in consequence of the unaltered iodide 

 of silver. For fixing, " hypo " is the safest, though cyanide may 

 be used if the film will bear it. 



Tannin plates when finished are hardly secure without var- 

 nish ; but there is considerable risk of spoiling gratings in the 

 operation if an ordinary negative varnish be used. The crystal 

 (benzole) varnish, which is applied cold, is much easier to use 

 and gives adequate protection. 



But the process which I am now most inclined to recommend 

 is that introduced by Mr. Wharton Simpson, and known as the 

 collodio-chloride process. The collodion, which may be procured 

 ready for use from Messrs. Mawson and Swan, of Newcastle, 

 consists of an emulsion of finely divided chloride of silver held 

 in suspension by the dissolved gun-cotton, together with a care- 

 fully adjusted excess of free nitrate. After a time the chloride 

 of silver is precipitated and the preparation becomes useless ; but 

 if properly mixed in the first instance, it will remain fit for use 

 for weeks or even months. In the production of gratings the 

 consumption is very small ; so that, if required for this purpose 

 alone, it is well to order it in small quantities. 



In order to secure a proper adhesion, I have found a prelimi- 

 nary coating of albumen absolutely necessary. The white of an 

 egg beaten up with a pint of distilled water gives a solution of 

 sufficient strength. The plates, previously cleaned, are coated 

 in any way that may be found convenient, and then set up on 

 blotting-paper to drain and dry. The principal precautions 

 necessary are to filter the albumen very carefully and to work in 

 a room free from dust. It will generally be convenient to pre- 

 pare a considerable number of plates at a time. Though of 

 almost infinitesimal thickness, the film of albumen produces a 

 very marked effect. Without the albumen the skin of collodion 

 will usually come right away from the glass when washed under 

 the tap ; with it the adhesion is remarkably good, and the film 

 so tough as even to bear rubbing with the finger while wet. 



The plate is coated with collodion in the ordinary way, and, 

 after resting a few seconds, is dried by heat over a spirit-lamp 

 or otherwise. After the plate has been made quite warm, it is 

 put aside in the dark to cool and to absorb a certain amount of 

 moisture from the atmosphere. This may take five or ten mi- 

 nutes. If the plate is used too soon the result is unsatisfactory ; 



