are still somewhat in Since routine radiochemical assay procedures for curium most feasible analytithe development stage, gamma counting was considered the cal technique for this experiment. Curium concentrations in the sample material rays of were, therefore, determined by counting the 228 and 278 keV gamma er. curium-243 using a NaI(T1l) detector and pulse height analyz nties Overall measurement error was assessed by considering potential uncertai was detector The NaI(T1) in each step of the sampling and analytical scheme. were Checks . energy-calibrated with appropriate gamma-ray emitting nuclides also made for gain shifts and changes in efficiency after every 10 samples with an aliquot of the dosing solution. Backgrounds were taken before, during, and after a series of counts to confirm that contamination of the counting To establish differences in counting variability chamber had not occurred. between sample types (milk, urine, feces, etc.) and the administered dose, "spiked" samples of each type were prepared in triplicate and assayed. These "spiked" samples were recounted after a period of time to determine any time-reThroughout the lated variations due to the settling out of sample material. Tissue samples were experiment, multiple samples were collected for analysis. not homogenized but are being counted twice (top-bottom rotation). The vari- ability between these two tissue counts will be used in the overall assessment of those uncertainties not associated with analytical hardware and counting duration. RESULTS AND DISCUSSION All samples collected during this study have been analyzed at least once, but because of a few very low curium concentrations, a portion of the samples are being recounted (1,000-minute counts). Furthermore, to determine the feasibility of radiochemical assays, selected animal tissues and some metabolic (urine, milk, etc.) samples will be prepared for analysis by an outside labora- tory. As a result, data reported below they do provide significant information administered curium. Considerably more transport of intravenously administered are not totally comprehensive, but on the caprine absorption of orally information is given on the biological curium. Experimental data were initially calculated as the curium concentration per gram of sample. For milk, urine, and fecal samples, the concentration per gram of sample was multiplied by the respective total daily output (weight) of these substances. Results for the intravenously dosed goats are presented as a percentage of administered dose per total collection (Table 2) and as a percentage of dose per kg of material (Table 3). In the case of the orally dosed goats, approximately 99 percent of the dose was recovered in the feces; the majority of milk, urine, and blood samples had no detectable curium concentrations. Curium concentrations in the feces from these orally dosed goats were relatively high from 24 to 96 hours after dosing. The greatest curium output occurred in the feces at the 48-hour collection period. 171