For the purpose of deepening the interpretation of regional gravity datait is necessary to utilize data on the crust and the deeper part provided bydeep seismic sounding. Several geoscientific global transects completed in Chinain the past few years have furnished such possibility. We hold that only onthe basis of the gravity density model which is designed according to thesetransects and agrees with the reality can effective interpretation result beobtained. The conversion of data obtained at a few deep seismic soundingstations into gravity Moho is extremely unreliable, and the metallogenic pro-gnosis made on such a basis will not be correct. The new gravity model shouldinclude multi-layered structure of the crust, faulted block, varying densitycharacter, density ununiformity and not great undulation of Moho as well asthe possibility that fractures are gentle shear zones at depth. The two-layerstructural model neglects these basic characteristics and therefore its interpre-tation has no practical significance.