Extreme enrichment and postmagmatic hydrothermal mobilization of the rare earth elements (REE), Zr, and Nb have been reported for a number of anorogenic peralkaline intrusions, including the world-class REE-Zr-Nb deposit at Strange Lake, Quebec, Canada. Establishing lithogeochemical vectors for these types of deposits is a challenging task because the effects of hydrothermal processes on element distribution are poorly known and the relationships of alteration types to mineralization stages have not been well documented. Here, we present results of a detailed mineralogical and geochemical investigation involving a dataset of over 500 mineral and bulk-rock analyses of a northeast-southwest section through the potential ore zone at Strange Lake. Based on these data, we develop a model that explains the role of hydrothermal processes in concentrating metals in peralkaline granitic systems and identify lithogeochemical vectors for their exploration. The B zone, located along the northwestern margin of the Strange Lake pluton, contains a lens-shaped, pegmatite-rich domain comprising subhorizontal sheets of pegmatites hosted by granites with a total indicated resource of 278 million tonnes (Mt) grading 0.93 wt % total rare earth oxides (TREO), of which 39% are heavy rare earth elements (HREE). Within this resource, there is an enriched zone containing 20 Mt of ore grading 1.44 wt % TREO, of which 50% are HREE. The pegmatites are characterized by a core enriched in quartz, fluorite, and light rare earth elements (LREE) fluorocarbonates, and a granitic border enriched in zirconosilicates and granitic minerals. The pegmatite sheets and surrounding granites evolved in three essential stages: a magmatic stage (I), a near-neutral hydrothermal stage involving their interaction with NaCl-bearing orthomagmatic fluids (II), and an acidic hydrothermal stage (III, comprising high-[IIIa] and low-temperature [IIIb] substages) that resulted from their interaction with pegmatite-sourced HCl-HF–bearing fluids. Stage IIIa led to pseudomorphic mineral replacement reactions (e.g., Na-Ca exchange during replacement of zirconosilicates) and formation of an aegirinization/hematization halo around the pegmatites. In contrast, stage IIIb, which was responsible for the hydrothermal mobilization of Zr and REE, is manifested by fluorite and quartz veins, zircon spherules, gadolinite-group minerals, gittinsite, ferriallanite-(Ce), and a pervasive replacement of the granite by these minerals. The distribution of REE, Zr, Nb, and Ti was controlled by the competition between hydrothermal fluids and the stability of primary REE-F-(CO 2 ) minerals (e.g., bastnasite-(Ce) host to LREE), zirconosilicates (i.e., Na zirconosilicates and zircon host to HREE and Zr), and Nb-Ti minerals (i.e., pyrochlore host to Nb and narsarsukite host to Ti), and the stability of secondary LREE silicates (i.e., ferriallanite-(Ce)), HREE silicates (i.e., gadolinite-(Y)), zirconosilicates (i.e., gittinsite and zircon), and Nb-Ti minerals (i.e., titanite and pyrochlore). Lithogeochemical vectors were identified to distinguish between the high-temperature acidic alteration (IIIa), using CaO/Na 2 O (indicator of Ca metasomatism) and Fe 2 O 3 /Na 2 O ratios (indicator of aegirinization/hematization), and the low-temperature acidic alteration (IIIb), using the CaO/Al 2 O 3 ratio (indicator of Ca-F metasomatism). Bulk-rock compositional data show that alteration was accompanied by an enrichment in heavy rare earth oxides (HREO) and ZrO 2 at the deposit scale, whereas there was no selective enrichment in the light rare earth oxides (LREO). A 2-D geochemical model of the deposit indicates that the LREO are more dispersed, whereas HREO and ZrO 2 are selectively distributed. These variations in LREE/HREE are also reflected in the mineral chemistry, especially in hydrothermal zircon crystals showing an unusual LREE enrichment and HREE depletion, contrasting with pseudomorphs, which are enriched in HREE. Hydrothermal ferriallanite-(Ce) and gadolinite-group minerals also show a clear trend of REE depletion with Ca enrichment. Controlling factors for the hydrothermal mobilization of LREE, HREE, and Zr were temperature, pH, and the availability of fluoride ions (F − ) in the fluid for the dissolution of zircon, and chloride ions (Cl − ) for the complexation of the REE. The study of rare hydrothermal minerals in conjunction with field observations and the evaluation of variations in bulk-rock composition allowed us to develop a new model for the hydrothermal evolution stage of Strange Lake.