“…In the latter deposits, the carbonatites were strongly overprinted by hydrothermal fluids resulting in remobilization and precipitation of the REE in vug-and vein-hosted assemblages. Although it is widely believed that these fluids originate from the carbonatite (Currie and Ferguson, 1971;Kresten 1988;Wyllie, 1989;Gittins et al, 1990) and fluid inclusion studies (e.g., Bühn and Rankin, 1999;Bühn et al, 2002; have documented the nature of the fluids exsolving from carbonatites, the controls on the REE mineralization are poorly understood. In many carbonatite-hosted deposits, there is a close association of the REE mineralization with fluorite (e.g., Amba Dongar, Doroshkevich et al, 2009;Bear Lodge, Moore et al, 2015), and this has led some authors (Möller et al, 1980;Salvi and Williams-Jones, 1990;Williams-Jones et al, 2000) to suggest that the REE are transported as fluoride complexes, particularly as such complexes are known to be extremely stable (Migdisov et al, 2009 i.e., REE-fluorocarbonates that are associated with barite, fluorite, and quartz (Doroshkevich et al, 2009).…”