Data from the EUV Imaging Spectrometer on Hinode taken on 2007 May 1 (May Day) are used to investigate the thermal properties of a coronal loop in AR 10953. For background subtraction, we have taken cuts across the loop near the apex and the footpoint where the background is relatively simple. Three density-sensitive line ratios give statistically different answers, and emission measure loci plots indicate that the loop plasma in not isothermal. Therefore, we have done differential emission measure analysis on these data and found a two-component model that can reproduce the background-subtracted intensities. Since both of these components are broadened, they cannot simply represent two isothermal strands of the EIS loop or two isothermal loops along the line of sight. They could, however, represent either two dominant ensembles of strands for the observed EIS loop or the dominant ensemble of strands for two individual loops along the line of sight. The TRACE image of the active region can help us determine which of these models best describes the data. It shows what appears to be two distinct loops that cross, one behind the other, at the approximate position of our cut near the EIS loop apex. It seemed natural to conclude, therefore, that the two-component DEM distribution represents two ensembles of strands, one for each of the loops seen in the TRACE image.