Patients with restless leg syndrome (RLS) complain of motor restlessness, usually occurring while they rest in the evening. Two immobilization tests have been described to assess leg restlessness in these patients. In the first test, the patient sits in bed with his or her legs outstretched while electromyograms are recorded from right and left anterior tibialis muscles for an hour (Suggested Immobilization Test [SIT]); in the second test, the legs are immobilized in a stretcher (Forced Immobilization Test [FIT]). In the current study, the SIT and the FIT were compared in patients with RLS and normal control subjects matched for age and sex. More leg movements were seen in patients than in controls during immobilization tests, especially the SIT. These movements were periodic, occurring at a frequency of approximately one every 12 seconds. The SIT (index > 40) was found to discriminate between RLS and control subjects better than the FIT (index > 25). Patients were also recorded during two consecutive nights to measure periodic leg movements in sleep (PLMS). A SIT index greater than 40 and a PLMS index greater than 11 (highest PLMS index of 2 consecutive nights) were found to discriminate patients with RLS from control subjects with similar power. With each of these two measures, the clinical diagnosis was correctly predicted in 81% of patients and 81% of the control subjects. The SIT has several advantages over the measure of the PLMS index; it does not require an all-night polygraphic recording and can be administered several times a day to measure circadian fluctuation of motor restlessness.