We investigated the genetic structure and variation of 21 populations of cattle (Bos taurus) in northern Eurasia and the neighbouring Near Eastern regions of the Balkan, the Caucasus and Ukraine employing 30 microsatellite markers. By analyses of population relationships, as well as by a Bayesian-based clustering approach, we identified a genetic distinctness between populations of modern commercial origin and those of native origin. Our data suggested that northern European Russia represents the most heavily colonized area by modern commercial cattle. Further genetic mixture analyses based on individual assignment tests found that native Red Steppe cattle were also employed in the historical breeding practices in Eastern Europe, most probably for incorporating their strong and extensive adaptability. In analysis of molecular variance, within-population differences accounted for ~90% of the genetic variation. Despite some correspondence between geographical proximity and genetic similarity, genetic differentiation was observed to be significantly associated with the difference in breeding purpose among the European populations (percentage of variance among groups and significance: 2.99%, P = 0.02). Our findings give unique genetic insight into the historical patterns of cattle breeding practices in the former Soviet Union. The results identify the neighbouring Near Eastern regions such as the Balkan, the Caucasus and Ukraine, and the isolated Far Eastern Siberia as areas of ‘genetic endemism’, where cattle populations should be given conservation priority. The results will also be of importance for cost-effective management of their future utilization.