Experimental evidence supporting Kolmogorov's refined similarity hypothesis

In a recent Letter, Hosokawa and Yamamoto [Phys. Fluids A 3, 457 (1992)] have provided evidence against Kolmogorov's refined similarity hypothesis based on the results of direct numerical simulations of isotropic turbulence [J. Phys. Soc. Jpn. 60, 1852 (1991)] with Re_λ close to 100. Here experimental results are presented from a wind-tunnel study of a cylinder wake, at Re_λ of 550, which in spirit confirm the hypothesis, especially for the smaller scales in the inertial range. The present experimental velocity spectra contain one and one-half decades of inertial range. The apparent basic disagreement between the numerical simulations and the experiments may be due to the difference in Re_λ, or may be symptomatic of a more subtle problem.