A brief review is presented on historical aspects of the dynamic nuclear polarization (DNP) research in 1960-1970s of the last century. Much attention is given to development of the many-particle approach, from pioneering studies of the 3-spin electron-nuclear cross effect to development of spin-temperature concept and thermal mixing mechanism of DNP in solids. Applicability of different models to modern DNP data is briefly discussed.This brief review is focused primarily on historical aspects: we will try to recollect the atmosphere of research in the fields of dynamic nuclear polarization (DNP) and solid-state spin dynamics, typical for the 1960s and 1970s of the last century. The paper does not claim to be exhaustive; in particular, the original references are often replaced by books and reviews. Much attention is given to studies performed in the Soviet Union; some of them are little-known because of the former Iron Curtain and the language barrier (the publications were in Russian). We hope to improve this situation and apologize in advance for a possible bias.At that time, the DNP technique was already well known and successfully applied, mostly in polarized proton targets. The physical interpretation of the DNP in non-conducting solids was restricted to the simplest two-particle picture including an electron spin S i of paramagnetic impurity coupled with a nuclear spin I j of the host lattice by magnetic dipole-dipole interaction. Following the ''solid effect'' (SE) mechanism suggested by Abragam and Proctor [1] (see also the