Summary Objective: Measurement of rectal temperature is the most common method and considered gold standard for obtaining body temperature in dogs. So far, no study has been performed comparing agreement between rectal and auricular measurements in a large case series. The purpose of the study was to assess agreement between rectal and auricular temperature measurement in normothermic, hypothermic, and hyperthermic dogs with consideration of different environmental conditions and ear conformations. Materials and methods: Reference values for both methods were established using 62 healthy dogs. Three hundred dogs with various diseases (220 normothermic, 32 hypothermic, 48 hyperthermic) were enrolled in this prospective study. Rectal temperature was compared to auricular temperature and differences in agreement with regard to environmental temperature, relative humidity, and different ear conformations (pendulous versus prick ears) were evaluated using Pearson’s correlation coefficient and Bland-Altman analysis. Results: Correlation between rectal and auricular temperature was significant (r: 0.892; p < 0.01). However, Bland-Altman plots showed an inacceptable variation of values (bias: 0.300 °C; limits of agreement: –0.606 to 1.206 °C). This variation was above a maximal clinical tolerance of 0.3 °C, which was established by experts’ opinion (n = 16). Relative humidity had a significant influence (p = 0.001), whereas environmental temperature did not. Conclusion: Variation between the two methods of measuring body temperature was clinically unacceptable. Clinical relevance: Although measurement of auricular temperature is fast, simple, and well tolerated, this method provides a clinically unacceptable difference to the rectal measurement.