VOYAGE TO ALGERIA. 223 



nary remark or two will clear our way towards an ex- 

 planation. Colour resides in white light, appearing 

 when any constituent of the white light is with- 

 drawn. The hue of a purple liquid, for example, is 

 immediately accounted for by its action on a spectrum. 

 It cuts out the yellow and green, and allows the red 

 and blue to pass through. The blending of these two 

 colours produces the purple. But while such a liquid 

 attacks with special energy the yellow and green, it 

 enfeebles the whole spectrum. By increasing the thick- 

 ness of the stratum we may absorb the whole of the 

 light. The colour of a blue liquid is similarly accounted 

 for. It first extinguishes the red ; then, as the thick- 

 ness augments, it attacks the orange, yellow, and green 

 in succession ; the blue alone finally remaining. But 

 even it might be extinguished by a sufficient depth of 

 the liquid. 



And now we are prepared for a brief, but tolerably 

 complete, statement of that action of sea-water upon 

 light, to which it owes its darkness. The spectrum 

 embraces three classes of rays the thermal, the visual, 

 and the chemical. These divisions overlap each other ; 

 the thermal rays are in part visual, the visual rays in 

 part chemical, and vice versa. The vast body of thermal 

 rays lie beyond the red, being invisible. These rays are 

 attacked with exceeding energy by water. They are 

 absorbed close to the surface of the sea, and are the 

 great agents in evaporation. At the same time the 

 whole spectrum suffers enfeeblement ; water attacks all 

 its rays, but with different degrees of energy. Of the 



never seen the Lake of Geneva, but I thought of the brilliant 

 dazzling- dark blue of the mid- Atlantic under the sunlight, and its 

 black-blue under cloud, both so solid that one might leap off the 

 sponson on to it without fear ; this was to me the most wonderful 

 thing which I saw on my voyages to and from the West Indies.' 



