Background: Left atrial appendage (LAA) contraction velocities are used frequently as surrogates of global left atrial (LA) function, but the validity of this approach has not been confirmed. Objective: The objective of our study was to assess the relationship between LAA flow velocities and multiple global LA variables. Methods: The correlations between LAA contraction velocities and global LA variables (difference between maximal and minimal LA volumes, mitral inflow A velocity, atrial ejection force, mitral annulus late diastolic [a′] velocity, pulmonary vein atrial reversal velocity) were assessed in 349 subjects participating in the Stroke Prevention: Assessment of Risk in a Community study (age, 67 ± 11 years; 53% men; all subjects in sinus rhythm), who were examined with transthoracic and transesophageal echocardiography. Results: The correlations between LAA contraction velocities and multiple global LA variables were poor or nonsignificant. The best correlations, albeit weak, were between LAA velocities and mitral annulus a′ velocities (r = 0.32 and r = 0.35 for correlations with septal and lateral annulus a′ velocities, respectively, P < .001). Low LAA contraction velocities (lowest 10%, velocities <56 cm/s) were not sensitive in detecting low values (lowest 10%) of LA variables (sensitivity <0.2 for most LA variables). In a subgroup of subjects with lower left ventricular ejection fraction (≤50%), the correlations between LAA velocities and LA variables were not apparently better than those in the total population. LAA velocities correlated negatively with LAA orifice diameter (r = −0.29, P = .002). The correlations between LAA velocities and global LA variables remained poor after correcting for LAA orifice size. Conclusion: LAA contraction velocities correlate poorly with multiple LA variables. Therefore, LAA flow velocities should not be used as surrogates of global LA function. (J Am Soc Echocardiogr 2002;15:433-40.)