of Edinburgh, Session 1881 - 82 . 
473 
in the water. These particles being so small they can reflect 
only the short waves of light, or those which belong to the 
blue end of the spectrum. The other theory explains the colour 
by supposing that water has a selective absorption for the rays of 
the red end of the spectrum — that water is in fact a blue trans- 
parent medium. 
Three different methods were adopted of testing the correctness 
of these rival theories, and all three proved the water of the 
Mediterranean to be blue by selective absorption, and show that 
light in passing through the water has the rays of the red end of 
the spectrum absorbed, and only those of the blue end transmitted. 
The first method tried was to find out what is the colour of the 
illumination of submerged objects. This was done by taking a long 
metal tube, closed at the end with a glass plate, and sinking it 
vertically in the water, and looking through it at white and different 
coloured objects fixed near the end of the tube. When this was 
done, it was found that a white object appeared of a most beautiful 
deep and delicate blue at a depth of 6 m. If the selective reflec- 
tion theory was true, submerged objects would be illuminated with 
a colour complimentary to that reflected by the fine particles, and 
would therefore appear orange or yellow, the exact colour depending 
on the amount of green in the reflected blue. 
If the blue colour of the sky, as generally supposed, is due to 
the reflection of the blue rays by small particles of matter suspended 
in the air, it obviously follows that light in passing through our 
atmosphere must become of a colour complimentary to the blue of 
the sky ; and it is asked, may not this be one of the reasons why 
the sun when near the horizon, and all artificial lights when seen 
at a great distance, appear more or less yellow ? 
The second method of experimenting was by looking at a white 
surface through a considerable length of water contained in a 
blackened tube. The light transmitted was found to be blue, 
thus showing the water to have a selective absorption for the rays 
of the red end of the spectrum. 
The third method was, by sinking white and different coloured 
surfaces under the water, and noting the change which took place 
in the colours. The colours selected were— -red, yellow, and purple. 
It was found that these colours when seen through the water 
