On a Paradox in the Theory of Heat Conduction