434 Talbot H. Waterman 



directionality and scattering of light under water. For instance, one might study the 

 refractive effect of abrupt density changes where these are inaccessible for direct 

 observation (Limbaugh and Rechnitzer, 1955). 



As to their biological interest, the present results would seem to be significant 

 because they greatly extend the known regions in which those animals whose eyes are 

 sensitive to them might detect natural patterns of underwater polarization. This 

 would now seem to be indicated for a good part, if not all, of the photic zone. Similarly, 

 the possibility of some sort of a polarized light sun compass, comparable to that used 

 in the sky by bees, and other animals, has been extended under water at least to 200 m, 

 since changes in the sun's position result in corresponding modifications of the 

 polarization pattern that far down. 



Of the several fields in which further research on the subject of underwater polariza- 

 tion and its significance may be profitable, the most interesting one related to the 

 present deep-sea work would be the possibility of making measurements at still 

 greater depths. Since a 20 sec exposure at f/2-8 was required to photograph the 

 interference pattern at 200 m, the present method was being pushed close to its reason- 

 able limit. This is obvious from the fact that the light energy penetrating the water 

 mass would probably be reduced to 0-1 its value at 200 m by an additional 40-50 m 

 of water. Hence to go from 200 m to 300 m would require 100 x as much time for 

 comparable negative densities! 



On the other hand, since directionality of penetrating light is maintained to the 

 lower limit of the photic zone, even though obliquity is decreased eventually to zero, 

 somewhat deeper measurements should be possible if the polarization pattern were 

 photographed in an upward direction. At least the light intensity would be greater 

 to a degree dependent on the relation between scattering and absorption in the water 

 mass concerned (Pettit, 1936; Atkins and Poole, 1952). However, the percent 

 polarization due to Rayleigh scattering would approach zero along the axis of the 

 directional light beam so that the interference figure would become weak or disappear. 



In any case it would be desirable to attempt such measurements and discover which 

 of the various factors are actually critical under these circumstances. The presence 

 of upwardly directed " telescopic " eyes in some bathypelagic fishes like Opistho- 

 proctus and Argyropelecus is suggestive of the biological importance of vertical illumin- 

 ation in deep water. 



REFERENCES 



Atkins, W. R. G. and Poole, H. H. (1952), An experimental study of the scattering of light by natural 



waters. Proc. Roy. Soc, London, B, 140, 321-338. 

 Clarke, G. L. (1941), Observations on transparency in the southwestern section of the North Atlantic 



Ocean. /. Mar. Res., 4, 221-230. 

 Jerlov, N. G. (1951), Optical studies of ocean waters. Rept. Swedish Deep-Sea Exped., 1947-1948, 



3 (1), 1-59. .... 



Limbaugh, C. and Rechnitzer, A. B. (1955), Visual detection of temperature-density discontmuities 



in water by diving. Science, 121, 395-396. 

 PExnT, E. (1936), On the color of Crater Lake water. Proc. Nat. Acad. Sci., 22, 139-146. 

 Poole, H. H. (1945), The angular distribution of submarine daylight in deep water. Sci. Proc, Roy. 



Dublin Sac, 24, 29^2. 

 Waterman, T. H. (1950), A light polarization analyzer in the compound eye of Limulus. Science, 



111, 252-254. 

 Waterman, T. H. (1954 a), Polarization patterns in submarine illumination. Science, 120, 927-932. 

 Waterman, T. H. (1954 b). Polarized light and the angle of stimulus incidence in the compound eye 



of Limulus. Proc. Nat. Acad. Sci., 40, 258-262. 

 Whttney, L. V. (1941), The angular distribution of characteristic diffuse light in natural waters 



J. Mar. Res., 4, 122-131. 



