Ann: Half Yearly Report and Accounts, page-11

Several reasons could cause check assays to be undertaken;
1) The REE is above the detection limit of the method. This would not usually be the case even for % level REE's using most fusion methods. If it was 20%+, maybe there's a chance they would do a repeat assay to be sure.
2) An element within the rock (could be P) is difficult to assay at high percentages, or has a lower DL than the REE's using fusion and XRF. This element may need to be assayed via a different method to provide an accurate assay. This would be more likely.
3) THe REE"s are out of normal equilibrium and show strange enrichments. Generally one way the laboratory ensures their assays a reliable would be to ensure the REE concentrations and ratios of one REE to another REE are 'normal'. However, the REE can be out of equilibrium and have strange ratios (eg; low Ce, high Dy, for example) that can make the routine QAQC process shoot up a red flag. To be sure of the result, the laboratory can do a check assay to make sure the result is accurate and precise.

I would think it's more #2 or #3 than #1. Which is not to say there may not be a high REE result, just that it's not...like, 45% REE's or something. eg; KFM didn't appear to have this issue with it's 18% TREO rock chips from Mick Well because it's within the range of the method.