368 SCIENCE PROGRESS 



method. The colour of the photograph is not impersonal 

 evidence of the colour of the original ; but the same qualification 

 applies to the vast majority of observation records. This much, 

 however, may be said in favour of three-colour processes — 

 if a series of photographs is made by the same process and with 

 the same materials, the discrepancies of colour would probably 

 be more easy to trace than the random errors possible when 

 the colouring is done entirely by hand. Moreover, any one who 

 is accustomed to photographic manipulations and others with 

 practice can work some of the processes, while it must be the 

 comparatively few who can ever hope to become skilled in 

 applying colours with the brush. 



