Major concerns remain about the reliability of tympanic thermometers. The aim of this study was to investigate the agreement between rectal and tympanic thermometry and compare this with the repeatability of each method in patients with acute stroke. Tympanic temperature readings were from 0.7 degrees C below to 0.8 degrees C above rectal readings. As to repeatability, we found second rectal readings to be between 0.3 degrees C below and 0.4 degrees C above first readings, and second tympanic readings to be between 0.4 degrees C below and 0.5 degrees C above first readings. The agreement between the two methods is acceptable in clinical practice. Tympanic thermometry is useful in serial measurements, whereas, for a single accurate measurement, rectal mercury thermometry remains the method of choice.