CONCLUSIONS

Sodium polytungstate is an expensive, albeit extremely effective, method of microvertebrate fossil separation for certain lithologies. Apart from the SPT, only the hydrometer and deionized water are potentially difficult to obtain. It has been shown to increase, at three sites, the fossil concentration from ~1 in 100 (1.3%) to ~1 in 5 (19%). After operating the laboratory (using 5kg of dry SPT powder) for 24 working hours (including setup and experimental time) we managed to process all 11kg of concentrate without running so low on SPT as to ever need to stop processing (though there was a significant amount of dilute SPT from tool cleanings at the end). The lab remains in operation and, although we are recovering some SPT residue, functionally we have as much SPT as when we started, and have processed more than outlined here. The SPT process undoubtedly saved time in processing the Moncure locality, even when accounting for time taken to process the materials and set up the lab. This and the comparable fossil concentration improvement seen in the Moenave Formation site (~25% fossil in SPT-concentrated fraction) provides evidence that SPT treatment could be useful to other workers. We recommend a starting density of 2.70-2.75 g/mL, as this is low enough that the density increase during prolonged work will not result in significant fossil loss, but is also high enough to float most of the non-fossil clasts. Though every site varies, and for some sites it may be necessary to float bone and sink the rock (see McCarty and Congleton, 1994:189 for Table 8.1, showing different mineral and biological densities).

The fossils found in the float were few (see Table 5 and Figure 4) and were typically small scales embedded in larger clasts. Some fossils, such as teeth embedded in larger clasts, still sank, and after processing (picking) over 6000 grains of the float the only fossils found were scales, all of which we embedded in larger clay clasts, even though controlled experiments showed significant fossil loss (Table 4) at densities over 2.80 g/mL. Testing individual sites, and especially individual fossil types (as there is variation between them) is vital before beginning to use SPT separation.

Unlike others (Murray and Johnston, 1987:317; Cifelli et al., 1996:17), we found no noticeable viscosity increase in the SPT solution over time or with increased density. Jeppsson and Anehus (1999:57) argue that calcium carbonate and dolomite, when present in a sample, can cause an increase in viscosity, yet our main sample bore pedogenic carbonate and the viscosity was not noticeably different during the processing of it, than during the processing of the carbonate-poor Moenave Formation site. It is worth noting that calcium ions (as present in calcite, aragonite, dolomite and tap water) may cause an insoluble precipitate to form (Cifelli et al., 1996:18), and we did encounter the infrequent formation of a precipitate, though have failed to determine whether it is an SPT reaction or associated with the clays (McCarty and Congleton, 1994:198 note that clays can absorb SPT, which may cause this phenomenon). Soaking samples in dilute acetic, formic or a similar, weak acid and then rinsing, as well as screen-washing thoroughly (or even using kerosene) to remove clays, before running through SPT may be advisable, if only to avoid both the reported viscosity and precipitate issues (McCarty and Congleton, 1994:198).

Sodium polytungstate provides an efficient and reusable, albeit at a high initial investment, means to greatly improve fossil concentrations in microvertebrate samples. These fossil-dense samples are more quickly processed on the whole, and thus greatly facilitate research goals. There is a risk associated with the use of sodium polytungstate, in that many fossils (up to 16% in our trials) may be lost, and the resulting float is extremely nonfossiliferous (0.25% fossil) as to make reprocessing time-consuming and extremely difficult. As such, workers will need to evaluate the utility of sodium polytungstate on a site-by-site basis, with considerations as to the relative frequency of different fossil type, relative importance of total sampling, and total sample size (if enough sediment is collected, then the fossil loss may be outweighed by the research time gained). We hope that this contribution will help future workers make informed decisions about whether or not to use heavy liquid separation, and guide those that do through laboratory setup and processing.

Back                                                                                        Next