The currently most widely accepted theory of polar mesosphere summer echoes (PMSE) assumes that the echoes originate from turbulence-induced scatter in combination with a large Schmidt number caused by the presence of charged ice particles. We test this theory with calibrated observations with the European Incoherent Scatter (EISCAT) Svalbard Radar (ESR) at 500 MHz (Bragg wavelength 30 cm) and the Sounding System (SOUSY) Svalbard Radar (SSR) at 53.5 MHz (Bragg wavelength 2.8 m), which are collocated near Longyearbyen on Svalbard (78°N, 16°E). Our observations in June 2006 yield volume reflectivities ranging from values of 2.5 × 10⁻¹⁹ m⁻¹ to 1 × 10⁻¹⁷ m⁻¹ for the case of the ESR echoes and from 5 × 10⁻¹⁶ m⁻¹ to 6.3 × 10⁻¹² m⁻¹ for the SSR echoes. In the frame of the above-mentioned theory the expected reflectivity ratio should be equal to or larger than the ratio of the frequencies to the third power (i.e., larger than (500 MHz/53.5 MHz)³ = 816). Our experimental results show that 94% of the observations satisfy this expectation. The remaining 6%, which show too small ratios, can be tentatively attributed to calibration uncertainties and an incomplete filling of the scattering volume of the SSR, which is significantly larger than that of the ESR. Hence our observations are largely consistent with the predictions of the above-mentioned theory even though we note that it cannot prove it, which would require additional observations at different frequencies. However, this consistency is used as sufficient motivation to apply the assumed theory to the observations in order to derive Schmidt numbers and radii of the charged aerosol particles. Corresponding results are in excellent agreement with expectations from microphysical models and independent satellite and lidar observations, thereby corroborating our initial assumptions.