Two experiments were carried out to evaluate the sensory conflict and the postural instability theories of motion sickness. The central hypothesis of sensory conflict theory is that motion sickness is caused by conflict between the current pattern of sensory inputs about self-movement and the pattern that is expected on the basis of previous experience. A subsidiary hypothesis is that the degree of motion sickness is proportional to the magnitude of sensory conflict. The central hypothesis of postural instability theory is that motion sickness is caused by loss of postural control. A subsidiary hypothesis is that the degree of motion sickness is proportional to amount of postural instability, which can be manipulated by physical restraint. In both experiments there were two levels of sensory conflict and two levels of postural restraint. Dependent variables were latency of onset and severity of motion sickness. The widespread occurrence of motion sickness in both experiments clearly confirmed the main hypothesis of sensory conflict theory. The results from Experiment 1, that there was significantly more motion sickness in the restrained condition, and from Experiment 2, that there was no significant difference in symptoms between the two restraint conditions, provide no support for the subsidiary hypothesis of postural instability theory. Evidence relating to the subsidiary proposition of sensory conflict theory was inconsistent.