A common laboratory experiment in radiative heat transfer involves measuring the cooling rate of two water‐filled metal cans, one can left unpainted and the other painted black. The shiny unpainted can cools more slowly, and the student can be led to the incorrect conclusion that bodies that are poor emitters (good reflectors) of light are also poor emitters of heat. A variation of this experiment was done with the shiny unpainted can replaced by one painted white. The experiment still compares a good and poor reflector of light, but in this case there is essentially no difference in cooling rates. These results are explained and are discussed in terms of other phenomena involving radiative heat transfer.