The standard deviation of the final kinetic energy distribution (σe) as a function of mass of final fragments (m) from low energy fission of 234U, measured with the Lohengrin spectrometer by Belhafaf et al., presents a peak around m = 109 and another around m = 122. The authors attribute the first peak to the evaporation of a large number of neutrons around the corresponding mass number, i.e. there is no peak on the standard deviation of the primary kinetic energy distribution (σE) as a function of primary fragment mass (A). The second peak is attributed to a real peak on σE(A). However, theoretical calculations related to primary distributions made by H.R. Faust and Z. Bao do not suggest any peak on σE(A). In order to clarify this apparent controversy, we have made a numerical experiment in which the masses and the kinetic energy of final fragments are calculated, assuming an initial distribution of the kinetic energy without structures on the standard deviation as function of fragment mass. As a result we obtain a pronounced peak on σe(m) curve around m = 109, a depletion from m = 121 to m = 129, and an small peak around m = 122, which is not as great as that measured by Belhafaf et al.Our simulation also reproduces the experimental results on the yield of the final mass Y (m), the average number of emitted neutrons as a function of the provisional mass (calculated from the values of the final kinetic energy of the complementary fragments) and the average value of fragment kinetic energy as a function of the final mass (e). From our results we conclude that there are no peaks on the σE(A) curve, and the observed peaks on σe(m) are due to the emitted neutron multiplicity and the variation of the average fragment kinetic energy as a function of primary fragment mass.