Recently, two Wistar rat lines, bred and selected for either high (HAB) or low (LAB) anxiety-related behavior on the elevated plus-maze, were described as a novel psychopathologic animal model. The behavioral and neuroendocrine responses to exposure to an emotional stressor were markedly enhanced in HAB rats compared with LAB rats, thus resembling patients suffering from psychiatric diseases. The present study focused on the developmental and genetic basis of the line-specific differences by using cross-fostering and cross-breeding approaches. For the cross-fostering paradigm, neonate HAB offspring were nursed by a LAB foster mother, and vice versa, until weaning. In the cross-breeding approach, HAB females were mated with LAB males, and vice versa, to create an intermediate F1 generation. Thereafter, the F1 animals were strictly sibling-mated to produce a segregating F2 generation. At 10 weeks of age, anxiety-related behavior of all animals was tested on the elevated plus-maze. The robustness of emotionality was assessed in rats of both lines throughout their entire lifetime. Serving this purpose, the frequency of ultrasound isolation calls, indicative of the emotionality of newborn rats, was monitored in regularly-fostered HAB and LAB pups on postnatal day 11. In addition, the timecourse of anxiety-related behavior was studied by repeated testing on the elevated plus-maze at the ages of 10 weeks, 6 months, 16 months, and 19 months. The cross-fostering approach failed to reveal behavioral differences between regularly and cross-fostered HAB and LAB rats, indicating no line-specific differences in maternal care or maternally-influenced development, at least after postnatal day 1. In contrast, cross-breeding resulted in F1 and F2 offspring displaying intermediate behavioral patterns on the elevated plus-maze which were exactly in between those shown in HAB and LAB control rats, thus confirming a genetic basis of the differences in anxiety. Cross-breeding revealed no differences related to the gender of the offspring or to the line-derivation of sire or dam, indicating an autosomal, rather than heterosomal, heredity of the divergent emotionality in HABs and LABs. Further, we were able to show stable and robust emotional differences in rats of both lines during their entire lifetime. HAB rats showed an enhanced frequency of ultrasound isolation calls on postnatal day 11 (p < 0.05) and a lower open arm exploration of the elevated plus-maze throughout adulthood (p < 0.01) compared with the same-aged LABs. In conclusion, the extremely divergent anxiety levels of HAB and LAB rats are maintained during their whole lives and are determined genetically, rather than being learned. These findings may be important for further studies on the genetic basis of emotionality.