after 6 months of exposure at the surface except for slight etching 

 of the silicon cast iron; their corrosion rates were less than 0.1 

 MPY. These two alloys also were uncorroded at depths of 2,500 and 

 6,000 feet after comparable periods of exposure. 



The other cast irons and austenitic cast irons corroded at higher 

 rates at the surface than at either depth. Also, their corrosion rates 

 were lower at the 2,500 foot depth than at the 6,000 foot depth. The 

 corrosion rates of the high nickel alloy austenitic cast irons were 

 lower than those of the other cast irons except the silicon and silicon- 

 molybdenum cast irons which were uncorroded. 



The corrosion rates of the ordinary and low alloy cast irons were 

 treated as a group while the high nickel austenitic cast irons were 

 treated as another group. The corrosion rates of each group were 

 averaged to obtain representative curves. These average curves and 

 the maximum and minimum values are plotted in Figures 11 and 12 to 

 show the effects of depth and concentration of oxygen in sea water on 

 the corrosion rates. 



It is shown in Figure 11 that the corrosion rates of the cast 

 irons are higher at the surface than at both depths and that the cor- 

 rosion rates at the 6,000 foot depth are higher than those at the 2,500 

 foot depth. Also the corrosion rates of the nickel, nickel- chromium 

 and ductile cast irons are higher than those of the high nickel 

 austenitic cast irons. 



The corrosion rates of the cast irons also decreased with decreas- 

 ing concentration of oxygen in sea water as shown in Figure 12. The 

 corrosion rates of the nickel, nickel-chromium and ductile cast irons 

 were higher than those of the austenitic cast irons and they decreased 

 at a greater rate with decreasing concentration of oxygen than did 

 those of the austenitic cast irons. At the lowest oxygen concentration, 

 0.4 ml/1, the corrosion rates of the two groups were comparable. 



In Figure 13 are shown the average corrosion rate curves for the 

 steels; nickel, nickel- chromium and ductile cast irons and; the aus- 

 tenitic cast irons. The corrosion rates for the steels and the group 

 containing the nickel, nickel- chromium and ductile cast irons are com- 

 parable and for all practical purposes it can be concluded that their 

 corrosion behavior is essentially the same. The corrosion rate curve 

 for the austenitic cast irons is much lower than the other two and re- 

 flects the effect of the high nickel contents in reducing the corrosion 

 rates of cast irons. 



The effects of exposure in surface sea water on the mechanical 

 properties of two of the austenitic cast irons are given in Table 16. 

 The mechanical properties of Type 4 austenitic cast iron were not 

 affected while those of Type D-2C were adversely affected. Metallo- 

 graphic examinations of polished cross sections of Type D-2C showed 

 that it had been attacked by interdendritic corrosion which would 

 explain the decrease in mechanical properties. 



