As soon as the STU structure was recovered from the sea and placed 

 on the deck of the ship, the two test channbers containing control steel and 

 aluminum alloy specimens were removed from the STU and water samples 

 carefully collected from inside each chamber. The water samples were treated 

 in different ways as follows: 



1 . "Pickled" immediately with manganous sulfate solution and 

 alkaline potassium iodide solution so that dissolved oxygen concentration 

 could be determined later in the laboratory. 



2. About 10 ml of seawater was collected aseptically using sterile 

 syringe for bacteriological tests and stored in the ship's refrigerator for later 

 evaluation. 



3. Water temperature and pH values were determined immediately 

 aboard ship. 



4. Water samples were stored in glass containers for determination of 

 salinity at the laboratory. 



The results of these tests are presented in Table 7. Seawater samples 

 had been collected from a depth of 6,000 feet near the seafloor in the vicinity 

 of the STU exposure site during previous oceanographic data collection cruises. 

 The seawater analyses of these deep-ocean water samples were conducted 

 aboard ship and the results obtained are presented in Table 7. 



As shown in Table 7, the dissolved oxygen concentration inside the 

 cylindrical test chamber in which control panels were exposed was slightly 

 higher than that in seawater samples collected at the seafloor in 6,000 feet of 

 water. A small number of microorganisms (200/ml) were again found in 

 seawater containing control panels. The microorganisms could be species of 

 cellulose-decomposing marine bacteria. If so, these bacteria could have pene- 

 trated the membrane filter which is made of cellulose ester. The effects of 

 these bacteria on corrosion of control panels is not known. 



When the exposed steel and aluminum alloy specimens were examined 

 immediately after recovery aboard ship, the test panels were found to be free 

 of any fouling organisms, except for bacterial slime growth. The steel panels 

 were covered with a moderately thick layer of red rust (Figure 38). When a 

 small section of this red rust was removed, a bright metal surface was found 

 underneath. The aluminum alloy panels, especially along the edges and ends, 

 were covered with large masses of white aluminum oxide corrosion products 

 (Figure 39). The flat surface area was relatively free of such corrosion products. 



The steel and aluminum alloy control panels which were protected 

 inside the cylindrical test chambers were relatively free of corrosion products 

 on their surfaces (Figure 40). 



43 



