The modern theory of the evolving Earth is based on integrated isotopic data obtained for the accessible part of the planet and cosmic bodies, in which the U–Pb isotope system plays a key role. The theory is tested by the isotope systematics of oceanic basalt sources. The origin of continental volcanic rocks is often interpreted in terms of the isotopic systematics of oceanic basalts. However, such interpretations, as a rule, reveal contradictions arising from differences in the history and current mantle dynamics of oceans and continents. Under the oceans, a mantle material has long lost connection with the accessible Earth tectonic units; under the continents such a connection is often established. The nature of the evolution of deep‐seated processes under the continents remains uncertain and, by analogy with the oceans, requires deciphering in terms of the components of the mantle sources for volcanic rocks. In modern lithospheric plates of the Earth, there are regions ranging in width from hundreds to thousands of kilometers, which are characterized by high strain rates and, consequently, at least one to two orders of magnitude lower viscosity relative to that of the internal stable parts of the plates. This gives them a special structural status of “dispersed plate boundaries”. The isotope‐geochemical studies of volcanic rocks from regions of the unstable Asia revealed the different nature of components in sources, for which particular interpretations have been proposed. In this paper, a general systematics of sources is defined for volcanic rocks of the latest geodynamic stage in Asia through estimating the incubation time on the 207Pb/204Pb versus 206Pb/204Pb diagram. Two domains are designated: (1) low 238U/204Pb (LOMU) derived from the viscous protomantle (VIPMA), and (2) elevated 238U/204Pb (ELMU). The mantle domains evolved from the Earth's primary material between 4.51 and 4.36 Gyr ago, 4.0 and 3.7 Gyr ago, 2.9 and 2.6 Gyr ago, 2.0 and 1.8 Gyr ago, about 0.66 Gyr ago and <0.09 Gyr ago. Melting anomalies of ELMU sources characterize the unstable mantle of Southern Asia, and those of LOMU sources belong to the Japan‐Baikal geodynamic corridor of the transitional region between the unstable mantle of Asia and its stable core. The Late Cenozoic evolution of the Japan‐Baikal geodynamic corridor resulted in cutting the LOMU domain by the Jeju‐Vitim ELMU source line.

A comprehensive model for deep dynamics in Asia has been developed from the data on the evolution of melting anomalies in the context of lithospheric plate motions, interactions, orogeny, and rifting. The key components of our model are the primary (transition layer) and secondary (upper mantle) melting anomalies (Gobi, Baikal, and North Transbaikalia; and Hangay, Sayan, and Vitim, respectively). It is inferred that the primary melting anomalies originated at the beginning of the latest geodynamic stage (ca. 90 Ma) as a result of the transition layer distortion by lower mantle flows. Such primary anomalies were caused by avalanche collapses of the slab material that had been stagnated under the closed fragments of the Solonker, Ural-Mongolian paleooceans and the Mongol-Okhotsk Bay of Paleopacific. The secondary melting anomalies occurred due to the Early-Middle Miocene structural reorganization in the Pacific-Asian and Indo-Asian interaction zones. The primary melting anomalies governed the spatial distribution of forces and processes of the latest geodynamic stage. The secondary melting anomalies resulted from the lithospheric motions relative to the primary anomalies and provided for the development of orogeny and rifting. The BaikalMongolian corridor of asthenospheric flows was limited by the lateral zones of convergent interactions between India and Asia in the southwest, and North America and Asia in the northeast. In these lateral zones, Late Phanerozoic paleoslabs and ascending mantle fluxes were revealed in the transition layer, as well as in the upper mantle, without any destruction by the asthenospheric flows.