430 Talbot H. Waterman 



The whole deep-sea camera and case were mounted on a weighted frame suspended 

 from the end of the ship's hydrographic cable. When the equipment was lowered into 

 the water, the camera lens and interference analyzer were quite accurately directed 

 horizontally. This was so even while the ship was steaming ahead at slow speeds, a 

 fact which was checked by diving with the ship under way. Consequently the frame 

 of the photographic negatives has been taken as the vertical and horizontal reference 

 for analysis of the data. 



Previous work (Waterman, 1954 a) had shown that the intensity and plane of the 

 polarization of underwater illumination were dependent on the relation between the 

 line of sight and the bearing of the sun. Consequently, it was important for the present 

 deep-water measurements to control the azimuth in which the pictures of the polariza- 

 tion pattern were taken. To do so a large fin of sheet aluminum was attached to the 

 frame holding the camera in such a way that the horizontal direction in which the 

 camera would point coincided with the ship's heading. 



The procedure used in obtaining the photographic data was as follows. The camera, 

 loaded with Eastman Kodak Tri-X film, which has an American Standards Exposure 

 Index of 200, was lowered to the depth desired. Then the ship steamed slowly ahead 

 in the sun's azimuth, a manoeuvre which would shortly bring the camera's direction 

 of view into alignment with the sun's bearing. After a period which experience with 

 the resulting negatives showed was adequate for the slowly towed camera to settle 

 into this orientation, a series of three exposures was made separated by 5 second 

 intervals. 



Exposure times and diaphragm stops were first set on the basis of the estimated 

 extinction coefficients of the water mass concerned, then, if necessary, corrected from 

 experience. No photometric transparency measurements were made, but an approxi- 

 mate value was obtained for the surface layers by the use of the Secchi disc at each 

 station (Clarke, 1941) and by comparison with pubhshed data from stations made 

 in nearby areas (Jerlov, 1951). Actual exposures used varied with the sun's altitude 

 as well as depth of the camera from 1/25 sec at f/4 for 30 m to 20 sec at f/2-8 for 200 m. 



After pictures had been taken in one direction, the ship's course was altered 90° 

 clockwise, and after a suitable wait of about 5-10 min for the camera and cable to 

 swing around, another three photographs were made, this time looking 90° to the 

 right of the sun's azimuth. Similarly two more sets of exposures were made in direc- 

 tions 180° and 270° to the right of the sun's bearing. 



It should be mentioned that because of the wind (force 3-4 Beaufort) and sea 

 running at the various stations, it was not always possible to obtain records under 

 ideal conditions for each of the four azimuths tested. In some cases the wire angle 

 instead of being negligible was appreciable, particularly for downwind or crosswind 

 legs of the station. Similarly the behvaiour of the cable after a turn sometimes showed 

 that the apparatus was slow in aligning its heading with that of the ship. Reasonable 

 compromises between ideal recording conditions and practical handling circumstances 

 of the ship and gear had therefore to be made. 



In the absence of a more elaborate apparatus where the camera's direction could be 

 immediately and accurately controlled from the surface and monitored with repeating 

 compass and precise vertical reference, the reliability of the present results could be 

 insured only by the care and experience exercised in carrying out the manoeuvres 

 described. It could be checked, however, by the consistency of the data obtained, 



