Solution volumes and concentrations were proportional (as closely as possible from available Los Alamos records) to permit an arbitrary scale of 2.2 ft/cm. The synthetic waste solutions were then delivered to the test specimen in the same sequence as at the site at a controlled rate of 2 ml/hour using a metering pump. Disposal at the site, however, was mot continuous but rather sporadic. Since no record of the cycle is available, no attempt was made to reproduce it. We fully realize the complications such simplification of the model would produce but felt the chance to compare even a crude model to an actual site justified this approach. The effluent of the column was collected and prepared for analysis. At the conclusion of the experiment, the specimen was removed from the apparatus, dried, cut into 0.5 cm sections, and also analyzed by gamma spectrometry to determine the distribution of the actinides in the rock. RESULTS Examination of the actinide content in the core (fig. 3) showed that detectable amounts of these nuclides did indeed migrate downward under the prevailing conditions to 20 feet below the bottom of the trench. Elevated concentrations of both nuclides are found at less than 1 foot and at 13 feet. The bulk of the americium is centered around the 9 foot mark. Noteworthy is the abrupt diminution of plutonium and americium below 13 feet; in fact, only 0.2% of the detected plutonium wss below that level. On the cther hand 45% of the plutonium was 1 foot or less from the bottom of the trench. The distribution of actinides in the simulated model (fig. 4) is'in excellent agreement with our earlier modeling work using tracers and simulated rain (Fried et al., 1976). Characteristic of these data are strong retention of actinides at the near surface with exponential tailing downward. In addition, plutonium exhibits a second concentration band beyond this tail. The percent of plutonium in this band is a function of the prior chemical treatment of the tracer material. The chemistry of plutonium is complex, existing in a IV, V, and a VI oxidation atate. Polymeric states are also known at high pH. The strongly retained large peak at the surface is most likely Pu(IV). The faster moving species is not so readily characterized. The amount of fast moving Plutonium in the model here depicted is much smailer than earlier models, Yeflecting, perhaps, the lower pH of the feed solution. It appears in the histogram as a slight rise at 6 cm with a rapid drop. An examination of the eluent (fig. 5) from the model shows no readily discernable pattern. It contains no americium and only 5% of the plutonium. It does indicate the mobilizing influence of the various waste solutions upon plutonium, Most interesting is the rapid drop in the elution of plutonium once water has tinsed the waste solutions from the rock and is the only eluting agent. This agrees with previous work which shows that distribution of plutonium is dependent upon the conditions of loading on the rock such as chemical composition and rate. Once plutonium fs adsorbed by the rock, the actual water induced migration ig extremely slow. Metering pump FIGURE 2. Experimental modeling apparatus. 120 121