In this study, we sought to advance understanding of trait vulnerability to bipolar disorder (BD) by testing whether instability of the 24-hr activity rhythm was a biomarker of the trait. Locomotor activity was measured over 7 days using actigraphy in participants allocated to groups of high (n = 36) and low (n = 36) trait vulnerability for BD. As predicted, the high-vulnerability group recorded a significantly lower 24-hr activity rhythm amplitude than the low-vulnerability group, indicating a less stable activity pattern. Consistent with the notion of accumulating risk, results showed the lowest activity rhythm amplitudes in participants in the high-vulnerability group with a history of depression. Secondary investigations showed that mania proneness explained more variance than depression proneness, pointing to specificity of the findings for BD. The data reinforce claims that instability of daily rhythms is a biomarker of trait vulnerability to BD.