Carotid intima-media thickness is a well established non-invasive surrogate marker of cardiovascular disease, however there is evidence that structural modification of the arterial adventitia also accompanies cardiovascular risk factors and might be involved causally in atherosclerosis. We sought to determine the relative contributions of the intima-media and adventitia to variation in ultrasound and magnetic resonance derived measures of carotid wall thickness.\ Carotid ultrasound and magnetic resonance imaging were undertaken in 20 participants. Carotid intima-media thickness, carotid extra-media thickness (which incorporates the arterial adventitia) and total wall thickness (a combined near-wall intima-media thickness and carotid extra-media thickness) using high-resolution ultrasound, and wall thickness using magnetic resonance imaging, were obtained.\ All ultrasound-derived measures of the arterial wall thickness were highly correlated with wall thickness by magnetic resonance imaging (all P<0.001); as expected the total wall thickness by ultrasound measure was most tightly correlated (correlation coefficient=0.814, P<0.0001). In multivariable models, there was evidence that both carotid intima-media thickness and carotid extra-media thickness contributed independently to the variance in wall thickness by magnetic resonance imaging, especially for the most severe focal thickening. Measures of carotid wall thickness that incorporated all three layers of the arterial wall were more closely correlated with the number of cardiovascular risk factors than carotid intima-media thickness alone.\ These results indicate that the arterial adventitia is an important contributor to the wall thickness measure derived by magnetic resonance imaging, and that carotid extra-media thickness likely provides additional information concerning arterial structure than that obtained from carotid intima-media thickness alone.