The spatial connections of endogenous deposits with anomalies of the vertical density gradient in lithosphere, calculated on the formulas for the spherical gravity sources are analyzed for Amur Province. The spatial parameters of tin-bearing, tungsten-bearing and gold-bearing regional ore-magmatic systems (ore regions, belts, zones and areas) are characterized by correlation of deep gradient density anomalies in the crust and upper mantle with the extensivity of superficial ore mineralization. Tin-ore deposits are mainly connected with density heterogeneities of the upper crust layer (up to a depth of 20 km), and tungsten, molibdenium and polymetallic deposits-with anomalies of the density gradient in the low crust layer (at a depth of 30 km) and in the upper mantle (at a depth of 60～70 km). Gold-ore regions are characterized by the greatest vertical range of fluid-magmatic systems (about 100 km from the surface) and arethrough ones in relation to mantle and crust tectonic structures. Opportunity is shown for the 3D density gradient model of the Amur Province lithosphere to be applied for constructing the forecast metallogenic models. The spatial connections of endogenous deposits with anomalies of the vertical density gradient in lithosphere, calculated on the formulas for the spherical gravity sources are analyzed for Amur Province. The spatial parameters of tin-bearing, tungsten-bearing and gold-bearing regional ore-magmatic systems (ore regions, belts, zones and areas) are characterized by correlation of deep gradient density anomalies in the crust and upper mantle with the extensivity of superficial ore mineralization. Tin-ore deposits are mainly connected with density heterogeneities of the upper crust layer ( up to a depth of 20 km), and tungsten, molibdenium and polymetallic deposits-with anomalies of the density gradient in the low crust layer (at a depth of 30 km) and in the upper mantle (at a depth of 60-70 km ). Gold-ore regions are characterized by the greatest vertical range of fluid-magmatic systems (about 100 km from the surface) and are through ones in relation to mantle and crust tectonic structures Opportunity is shown for the 3D density gradient model of the Amur Province lithosphere to be applied for constructing the forecast metallogenic models.