Some comments on scaling exponents of turbulence

Abstract
Several authors have reported that in turbulence the scaling exponent of the first order velocity structure function increases when the Reynolds number Re decreases. From this result some important consequences on the transition to turbulence could be obtained. However we report experiemntal evidence that this result is coming only from an improper definition of the inertial range. Our data clearly show that the scaling exponents remain constant as a function of Re which is consistent with the selfsimilarity of spectra