within a particular site. This results in parts of an area being unminable and places certain boundary design conditions on mining equipment which are different for different sites.

With presently envisioned mining systems, the lowest wet concentration of any size that can be economically mined

is about 1-1/2 lbs/ft.2 (7.32 kg/m2). In a typical mine site, therefore, a small but significant percent of the deposit contains concentrations too low to be mined economically.

Surface and Subsurface Ore Potential: For purposes of reserve determination, only the nodules which occur on the sea floor surface can be considered in the tonnage determinations. It has been determined by careful sectioning of a large number of box cores (8" x 12" x 2 feet deep) taken at numerous deposits that minable deposits do not occur below the first inch or so of the upper thin layer of the sediment water interface. Figures 10 and 11 are photographs of a core of heavy concentration as seen from the surface and as sectioned. These figures vividly demonstrate the surficial nature of the deposit.

Assay: Many published assay values for manganese nodules are too high because they do not represent the total nodule. Every nodule contains varying amounts of noneconomic material such as clay silt, rock fragments, or biologically formed debris. As assay of a nodule as an ore must include these materials as part of the original weight. Assays are normally reported as a percentage of the dry weight of the nodule. It should be noted that the nodules are complex materials and difficult to sample, and the proper analysis can be subject to substantial procedural errors. Many of the literature values are also suspect because of potential analytical error.

Figure 9B shows the variability of analysis for copper and nickel in nodules from a deposit of interest. In this area, copper varies from 0.6 to 2.2 percent, while nickel varies between 0.8 and 1.5 percent. Each station represents four separate samples (sample weight for each station can be obtained by multiplying the station concentration by eight), and the assay is the average for the station. A study of data from many stations shows local fluctuations in analyses to be substantial although the average analyses

in typical potential mine sites can be constant over a wide area. In one such potential mine site copper contents varied from 0.71 percent to 1.52 percent and nickel varied from 0.96 percent to 1.63 percent. Thus, it is apparent that significant parts of a mine site may be below economic grade.

It is also noted that one company surveyed a dozen or so locations where literature data indicated a high nodule grade to have been sampled. In only four of these stations were nodules found in an abundance of greater than 1 lb/ft.2 with analysis comparable to those reported. The remainder were either of low abundance and limited areal extent or had obviously not been properly analyzed. One such station was Msnk shown on page 202 of "The Mineral Resources of the Sea" by John Mero. Analyses were respectably high but the abundance was very low.

The overall conclusion is that not only is abundance frequently in error in literature sources but so are analyses. Furthermore, even attractive analyses in the literature are, most often, representative of stations with poor abundance or adverse topographic features.

The question of analyses of candidate mining sites is a sensitive one for the Industry Representatives preparing this report. Accordingly, the above data seek to avoid disclosure of such sensitive information. Therefore, additional data on this question may be submitted on a confidential basis to support these conclusions in greater depth.

Figures 12, 13 and 14 contain assays on a percent dry-weight basis after being dried at 150°c. They all include the total material contained in the nodules and represent a bulk sample weighing from 1 to 10 pounds. Assays based on a single nodule are often not a representative sample.

The assays in Figure 12 are from a broad area and are remarkably uniform. The variation between subsamples from a single dredge haul is nearly as large as the total variation if sample 9 is omitted. The false sense of uniformity, however, is destroyed when one looks at Figure 13. This latter table contains closer spaced samples from within a portion of the broad area covered in Figure 12. Here, 30% of

the samples (1, 2, 7, 8) fall below a desirable level, but would be marginally economic.