THE PROPERTIES OF SEA-WATER 77 



The fact that light pénétrâtes to an appréciable extent down to 

 550 fathoms was not previously suspected. 



Regnard has devised an apparatus for determining- the 

 length of the day at différent depths, a photographie film on 

 a cylinder being passed by a clock-work arrangement before 

 an apertiire in the " caméra." Experiments made at Madeira 

 showed that at 11 fathoms the day lasted 11 hoiirs, at 16J 

 fathoms 5 hours, but at 22 fathoms, although the sun was 

 shining brightly, the film only exhibited a slight influence of 

 light for a quarter of an hour about 2 p. m. 



As a gênerai ruie pure water is less transparent the higher 

 the température. In spite of this, the tropical seas are more 

 transparent than the temperate. Salinity does not appear to 

 be of much importance in this connection. The Baltic and 

 North Seas are of like transparency but very différent salinity, 

 and the same is true of the Red and Sargasso Seas. 



The colour of the océan has been the subject of fréquent 

 misunderstandings and misdescription. Reflection from the 

 surface causes the various différences described by différent 

 writers and painted by différent artists. 



Omitting reflection from considération, the colour of local 

 seas is green, that of the tropical seas, on the other hand, blue. 

 Occasionally the colour of the sea takes on a whitish, yellowish, 

 reddish, or olive-coloured tint, but thèse appearances are 

 invariably due to the présence of suspended matter or of 

 organisms. The normal colour of the sea is consequently 

 either green or blue. 



For the exact détermination of the colour of the sea various 

 instruments and methods hâve been applied, of which one 

 only need be mentioned hère. Forel invented a scale or 

 xanthometer which indicates ail colour changes from pure 

 yellow through green to deep blue. Two solutions were made, 

 the first of i gram of copper sulphate with 9 grams of ammonia 

 and 190 grams of water; the second of i gram of neutral 

 potassium chromate in 199 grams of water. The first solution 

 was taken as the index and was numbered o. Then mixtures 



