Comparisons between the INAA results for the USGS standard rocks and the recommended values (Govindaraju, 1989) were all good and show that the expected precision and accuracy for the INAA element suite is of the order of 5% to 10% for most elements. Six elements—Cr, Zn, Rb, Sr, Ba, and Ce—that were determined by INAA were also measured by the shipboard XRF (Barker, Camerlenghi, Acton, et al., 1999). Differences between these two sets of analyses are small and can largely be attributed to analytical scatter and small interlaboratory biases. It is also important to note that the INAA sample is not from exactly the same powder as the XRF sample, and it was not washed to remove salts, so some variability might also reflect real differences between samples or differences in preparation.
LOI ranged from 3% to 5% in most samples to nearly 15% in two samples from Site 1101 (Table T8). This mass loss at high temperatures is most likely related to loss of water in hydrated phases, such as smectites, and CO2 loss caused by decomposition of calcite. The highest LOIs in the samples from Site 1101 were known to contain about 22% calcite (Shipboard Scientific Party, 1999), which would produce about 10% CO2. However, the CaO reported in the microprobe data is probably derived from both feldspars and calcite. The iron oxide reported in the microprobe analyses is given as FeO, but its oxidation state is unknown. The LOI was part of the normal XRF procedure and was designed to oxidize all Fe to Fe2O3. However, the glass beads were formed on a strip of metallic Mo under an Ar atmosphere, and some reduction back to FeO is possible.