Motor-respiratory coordination is the synchronization of movement and breathing during exercise. The relation between movement and breathing can be described using relative phase, a measure of the location in the movement cycle relative to the location in the breathing cycle. Stability in that relative phase relation has been identified as important for aerobic efficiency. However, performance can be overly attracted to stable relative phases, preventing the performance or learning of more complex patterns. Little research exists on relative phase dynamics in motor-respiratory coordination, although those observations underscore the importance of learning more. In contrast, there is an extensive literature on relative phase dynamics in interlimb coordination. The accuracy and stability of different relative phases, transitions between patterns, and asymmetries between components are well understood. Theoretically, motor-respiratory and interlimb coordination may share dynamical properties that operate in their different physiological substrates. An existing model of relative phase dynamics in interlimb coordination, the Haken, Kelso, Bunz model, was used to gain an understanding of relative phase dynamics in the less-researched motor-respiratory coordination. Experiments 1 and 2 were designed to examine the interaction of frequency asymmetries between movement and breathing with relative phase and frequency, respectively. In Experiment 3, relative phase stability and transitions in motor-respiratory coordination were explored. Perceptual constraints on differences in stability were investigated in Experiment 4. Across experiments, contributions relevant to questions of coordinative variability were made using a dynamical method called cross recurrence quantification analysis. Results showed much consistency with predictions from an asymmetric extension of the Haken, Kelso, Bunz model and theoretical interpretation in the interlimb coordination literature, including phase wandering, intermittency, and an interdependence of perception and action. There were, however, notable exceptions that indicated stability can decrease with more natural frequency asymmetries and the connection of cross recurrence measures to categories of variability needs further clarification. The complex relative phase dynamics displayed in this study suggest that movement and breathing are softly-assembled by functional constraints and indicate that motor-respiratory coordination is a self-organized system.