The 1998 April 20 spectral line data from the Coronal Diagnostic Spectrometer on the Solar and Heliospheric Observatory show a coronal loop on the solar limb. Our original analysis of these data showed that the plasma was multithermal, both along the length of the loop and along the line of sight. However, more recent results by other authors indicate that background subtraction might change these conclusions, so we consider the effect of background subtraction on our analysis. We show emission measure (EM) loci plots of three representative pixels: loop apex, upper leg, and lower leg. Comparisons of the original and background-subtracted intensities show that the EM loci are more tightly clustered after background subtraction, but that the plasma is still not well represented by an isothermal model. Our results taken together with those of other authors indicate that a variety of temperature structures may be present within loops.