Previous Magnetic Resonance (MR) studies of carotid endarterectomy (CEA) specimens have been directed at interpreting image contrast to determine plaque composition and stability. Such studies may give misleading results since it is believed that the acquired MR data is affected by the conditions used to store the excised tissue. This has been investigated in a high-resolution imaging study of the changes in contrast occurring with a change in specimen temperature. MR images were acquired from 20 CEA specimens. The initial MR examination was made within four hours of excision in tissue maintained and imaged at body temperature. Specimens were subsequently cooled and then re-examined at different times. The MR data was interpreted by comparison with histology obtained from equivalent sections. With the exception of signals arising from blood, changes in MR image contrast occurring in the 24-hour period after the CEA were relatively slight. Plaque lipid T2-weighted image intensity was initially relatively hyperintense in some samples and decreased with time on cooling reflecting a decrease in the lipid transverse (T2) relaxation time. High-resolution imaging allowed a detailed description of plaque calcification and showed that clot retraction was probably responsible for the marked change in the distribution of signals assigned to intraluminal blood. The use of high-resolution MR imaging to examine freshly excised specimens at body temperature showed previously unreported image features. This is important in the design of in vivo MR studies directed at assessing plaque stability by identifying the composition of the atheromatous tissue.