Sampling was completed in the summer and fall of 1992, by staff from the Alberta Research Council, the Saskatchewan Research Council (SRC), Manitoba Energy and Mines, and the GSC. At 734 sites, till samples were obtained at depths of 1 to 2 m below surface, in order to minimize the effects of surface weathering, from the area within farmland. An additional 82 till samples were collected along road traverses which extended to the Phanerozoic limit in Saskatchewan and Manitoba. At 1273 sites in Canada and the United States, A (averaging 3-18 cm below surface) and C (averaging 40-66 cm below surface) horizon soil samples were obtained. Details of the sampling procedures are described by Thorleifson and Garrett (1993). The field data provide information on the general sampling environment and observations on the colour, texture, moisture content, and composition of the soils and till.

Geochemical sample preparation and initial recovery of <2 mm heavy mineral concentrates from the 25 litre bulk till samples was completed at the SRC. To conform with agricultural and environmental protocols the <2 mm fraction of the soils were recovered, and the <63 µm fraction of the tills were recovered for geochemical studies. Heavy mineral separations were completed by Overburden Drilling Management Ltd. of Nepean, Ontario, using methylene iodide diluted with acetone to a specific gravity of 3.2. The concentrates were then returned to the SRC and were examined for potential indicator minerals under a stereoscopic microscope. The 0.5 to 2.0 mm non-ferromagnetic concentrates were picked without further processing. The 0.25 to 0.5 mm non-ferromagnetic fraction was sorted by magnetic susceptibility using a Frantz isodynamic separator into strongly, moderately, and weakly paramagnetic fractions. Selected grains were subsequently examined by the mineralogical staff of Consorminex Inc. of Gatineau, Quebec, and minerals such as staurolite and hornblende were removed prior to grain mounting and microprobe analysis. Full details of the procedures are presented in Thorleifson and Garrett (1993).

Electron microprobe data were used to classify the minerals. Garnets were classified using a simplified version of the Dawson and Stephens (1975, 1976) and Gurney (1984) classifications. Diopsides with >0.50% Cr2O3 (Deer et al., 1982; Fipke, 1989) were regarded as chrome-diopsides. Magnesian ilmenites in every case contained well in excess of 6% MgO. Cr-spinels exceeding 60% Cr2O3 and 12% MgO were regarded as compositions comparable to those reported for diamond inclusions (Gurney and Moore, 1993). Details of the classification were discussed by Thorleifson et al. (1994).

Subsequent to the initial microprobe analyses, a number of additional studies were undertaken. Eclogitic and near-eclogitic garnets were re-analyzed at reduced detection limits. These analyses were carried out to obtain acceptable sodium data, for diamond grade prediction (Gurney and Moore, 1993), and to enhance titanium data which were being used for definition of eclogitic garnets. The G7, G9, G10, and G11 pyropes and all Cr-spinels were analyzed by proton microprobe at the University of Guelph for Ni and several other elements in order to utilize classification methods developed by Griffin and Ryan (1993). Nickel temperatures were calculated for peridotitic garnets using the equation presented by Griffin et al. (1989).

The 8 to 16 mm fraction of till samples was classified with respect to lithology using a classification based on that of Shetsen (1984). Bulk mineralogy of the 63 to 250 µm-non-ferromagnetic heavy mineral fraction was determined on the basis of identification under a stereoscopic microscope of 300 grains in an araldite mount.

Following grinding of the <2 mm fraction of the soils to approximately <150 µm, duplicates and both internal GSC and international standards were inserted in the A and C horizon soil and till sample sequences. The three sequences were randomized and renumbered prior to analysis to destroy any relationship between analysis order and spatial context. All samples were submitted for a suite of trace element analyses using Instrumental Neutron Activation Analysis (INAA) and Atomic Absorption Spectrophotometry (AAS) procedures after a total acid decomposition. In addition, the till and C horizon soil samples were analyzed for major and minor elements by X-Ray Fluorescence (XRF), and carbonate determinations were performed on the tills by the Chittick procedure. Full details are provided in Thorleifson and Garrett (1993).