An experimental study of the scattering of light by natural waters

A blue-sensitive multiplier phototube was used to measure light scattered from a parallel beam in distilled, tap and sea water, the first named serving as a check upon errors from extraneous sources of light. Forward and back scatter are closely the same for distilled water, but with natural waters by far the greater part of the effect occurs through angles less than 25 degrees. A minimum is found for a deviation of about 110 degrees, back scattering increasing somewhat for greater angles. The relative importance of forward scatter increases with turbidity, and in sea water about three-quarters of the effect is due to matter removable by filtration through a collodion filter of average pore diameter 1 $\mu $ or by sedimentation; further passage through $\text{0.6 and 0.2}$ $\mu $ filters produces little change. Scattering is greater in blue light. Plymouth tap water scatters more than surface coastal water and the latter more than surface water 20 miles out, station E1. Surface water scatters more than deeper--the water column being remarkably homogeneous even when a well-marked thermocline had existed for weeks, but a small increase was detectable at the top of the cold water. E1 surface water increased in scattering between August and January, and decreased till May. Deep water showed little change. Extinction due to scattering between 20 and 155 degrees amounted to less than one-sixth of that found for a similar sample with a Pulfrich photometer, so probably much scattering occurs below 20 degrees. This explains why Pulfrich extinctions are so much greater than vertical extinction coefficients found in the sea. The preponderance of forward scattering within the range 20 to 155 degrees and the effects of filtration suggest that such scattering is due chiefly to refraction through transparent mineral particles, large compared with the wave-length of light. The refractive index of organic matter is too near that of water to produce refraction through angles as large as 20 degrees. Such matter may, however, be responsible for some of the scattering through smaller angles which apparently accounts for most of the turbidity found with the Pulfrich photometer.