The reproducibility of heart rate variability (HRV) measures during graded lower body negative pressure (LBNP) have not been studied in sufficient detail. Active college age men (n=14) underwent an orientation exposure and two trials of graded LBNP to presyncope or -100 mmHg, separated by 1 week. Heart rate, stroke volume (impedance cardiography), blood pressure (Finapres), and forearm blood flow were assessed, as was HRV in both time and frequency domains. The trial-to-trial responses to LBNP common to all subjects (LBNP< or =-60 mmHg and at test termination) showed parallel changes, suggesting similar responses between both trials. Good reproducibility estimates were found for the resting HRV data (lowest: R=0.62 for low frequency/high frequency ratio; highest: R=0.94 for standard deviation of normal R-R intervals). During LBNP, reproducibility estimates varied but were generally similar to that seen at rest. At test termination, they were unacceptably low (R<0.41) for the HRV data assessed in the frequency domain and expressed in absolute units. LBNP tolerance was lower in the first trial [LBNP tolerance index: 404 (21) versus 437 (15) mmHg min(-1); P<0.05] but the intraclass correlation coefficient was high (R=0.87). These data suggest that (1) HRV responses to submaximal LBNP up to -60 mmHg are consistent across trials, (2) the considerable variability seen in the HRV parameters at maximal LBNP can be reduced by expressing these data in either the time domain or using normalized units in the frequency domain, and (3) cardiovascular responses to sub- and maximal LBNP are reproducible. Data are presented as mean (SEM) unless otherwise stated.