Perceived stress can lead to dysregulated cortisol patterns, including blunted peaks and flatter slopes, which are associated with increased morbidity and mortality risks. Couples' interdependence provides a prime opportunity for partners' stress to disrupt a healthy cortisol pattern. This study examined how individuals' own perceived stress and their partners' perceived stress shape cortisol levels and slopes across the day, as well as how positive and negative behaviors during conflict discussions impact associations between stress and cortisol. Both partners of a married couple (n = 43 couples, 86 individuals) completed a full day in-person visit. Each partner completed the Perceived Stress Scale, and all couples engaged in a 20-min marital problem discussion which was recorded and later coded for positive and negative behaviors using the Rapid Marital Interaction Coding System (RMICS). Partners also provided five salivary cortisol samples across the day, two samples before the conflict and three after the conflict. The dyadic design and analyses provided a way to account for the interdependent nature of married couples' data, as well as to use the Actor-Partner Interdependence Model (APIM) to assess the mutual influence of spouses' stress on cortisol. Individuals with more stressed partners had flatter cortisol slopes than individuals with less stressed partners, who showed steeper and thus healthier declines across the day. Individuals' cortisol levels at the beginning of the day were similar regardless of their partners' perceived stress, but individuals with more stressed partners had higher cortisol levels 30-min, 1 h, and 4 h after the conflict discussion than those with less stressed partners. Couples' behavior during the conflict moderated the relationship between partner perceived stress and average cortisol; when couples used more negative and less positive behaviors, individuals with more stressed partners had higher average cortisol levels than those with less stressed partners. On a day couples experienced conflict, having a partner with higher perceived stress is associated with dysregulated cortisol patterns, including higher levels and flatter slopes, but having a partner with lower perceived stress is linked to steeper and thus healthier cortisol declines. A partner's stress was particularly consequential for one's own cortisol when couples used more negative and fewer positive behaviors during a conflict discussion. This research adds to the growing literature on pathways connecting marital interactions to important biorhythms and health. Copyright © 2020. Published by Elsevier Ltd.