Genetic factors are believed to be involved in the aetiology of unipolar depressive disorders. We have previously described a model built up by selective breeding of mice with different responses in the tail suspension test, a screening test for potential antidepressants. In this model, helpless H/Rouen mice are essentially immobile in this test, as well as in the Porsolt forced-swim test, whereas non-helpless NH/Rouen mice show the opposite behaviour, i.e. very low immobility. However, it is unclear whether or not the other phenotypic differences (forced swim test, locomotor activity, sucrose test, sleep patterns, effect of fluoxetine) observed between H/Rouen and the NH/Rouen mice may be attributed to a genetic drift phenomenon during the selection step, rather than being related to the trait of selection. In this study we used reciprocal crossbreeding between H/Rouen and NH/Rouen mice and obtained a segregating F2 population in order to determine whether phenotypic differences between the two lines co-segregate with the trait of selection. In the segregating F2 population, we found significant and strong genetic correlations between helplessness in the tail suspension test and some phenotypical features associated with depressive disorders such as 'alterations of sleep patterns', behavioural response to fluoxetine, immobility duration in the forced swim test, and anhedonia. Our results converge with clinical observations in depressed humans. These results strengthen the validity of the H/Rouen mouse as a model of depression, notably for preclinical studies with antidepressants. In addition, this model should open the way to identifying genes related to depression-like behaviours.