After the surface temperature was raised above 70 C, different degrees of bacterial killing were observed. It was difficult to maintain frankfurter samples absolutely aligning with the central axis between the heaters during the experiments. Therefore the samples might be ‘‘wobbling’’ along the central axis. This may have caused some variation in the surface temperature development among the samples, leading to some difference in the bacterial kill. On average, L. monocytogenes on turkey frankfurters was reduced by 3.5F0.4, 4.3 F0.4 and 4.5F0.2 logs (meanFstandard error), respectively, after the surface temperature was raised to 70, 75 and 80 jC. However, the difference observed in the mean log reduction at all these three temperatures was not statistically significant
(n=5, p>0.108).