Moonlight and Colour. 133 



and thus objects seen in moonlight and in sunlight will differ 

 on this account, as well as by simple reason of diminished in- 

 tensity of illumination. 



Let us now turn to another set of experiments, and, when 

 the moon is at or near her full, in clear weather, let one or 

 more members of a family dress themselves in strongly- coloured 

 clothes, red feathers, green shawls, blue silk dresses, etc., and 

 then walk in a moon-lit garden, leaving others to guess 

 what they have put on. Generally speaking very considerable 

 mistakes will be made, although, when close to bright red, 

 blue, green, or yellow objects, their colours will be approxi- 

 mately guessed. In one set of trials which we made, scarlet 

 was pronounced crimson, but at the same time the supposed 

 black cloth binding of a book was called purple, which was 

 found correct; thus showing that the blue and violet rays 

 could not be very deficient in force. Tertiary colours will be 

 more bewildering than primaries or secondaries, and different 

 eyes will pronounce different judgments, the discrepancy be- 

 tween a set of observers in moonlight not necessarily being the 

 same as might exist between them in daylight. 



The colour of shadows in moonlight will require long 

 study, and so will the tints or hues of trees in full leaf, near 

 and distant, both in the direct rays and out of them; and it 

 must be remembered that all moonlight effects will vary 

 according to the state of the air and the brightness of the 

 luminary. 



The coloured cards furnished with the " Colour-Top" will 

 help these inquiries, if they happen to be at hand ; otherwise, 

 books in various cloth bindings, and articles of clothing and 

 ornament will readily supply what is required. Besides these 

 a few strips of coloured glass, easily obtained from a glazier, 

 will be wanted ; but, Avhen once attention has been called to 

 the subject, an agreeable inquiry may be conducted with very 

 little trouble and no expense. 



VOL. IV. — -NO. II. 



