Simultaneous three-frequency (30, 401, and 800 Mc/s) radar studies of the aurora were made from a site in northern Scotland. A completely steerable 142-foot-diameter antenna was used at all frequencies, providing a 15° antenna beam at 30 Mc/s and identical scaled beams of 1.2° at 401 and 800 Mc/s. Detailed results on the position in space indicate that most echoes arise from a layerlike form of auroral scatterers whose thickness is about 10–20 km. The height distribution of the auroral return was peaked at about 100 to 120 km, but heights as great as 200 km were observed. The highest amplitude and greatest percentage of auroral echoes were found to arise from the intersection of the maximum of the visual auroral zone and the 100- to 120-km orthogonality loci. Wavelength dependence and aspect sensitivity measurements at 401 and 800 Mc/s resulted in a power law dependence of λ7 and aspect sensitivity of 10 db per degree of off-perpendicular angle. Interpretation of these results in terms of the Booker scattering formula indicates that the size of the irregularities is 45 to 90 meters along the magnetic field lines and 0.7 meter across the field lines. The wavelength dependence results from the 30- and 401-Mc/s data indicate that questionable beam filling at 30 Mc/s introduces an unresolvable uncertainty of 45 db. The 30-Mc/s aspect sensitivity results indicate an amplitude drop of 10 db per 5° of off-perpendicular angle, or a length of the scatterers of 120 meters along the magnetic field line. Detailed comparison of magnetic activity auroral Doppler shifts and auroral echo amplitude indicates that the auroral Doppler spectrums and the amplitude of the echoes have a close relationship to the auroral current system. The auroral echoes seen at 401 Mc/s appear at times to be completely depolarized, possibly owing to Faraday rotation, but in general have a considerable degree of polarization retention. At 30 Mc/s, long-range auroral echoes propagated by an intervening F-layer reflection were seen as well as ground backscatter propagated by auroral sporadic E. The data indicate that the present theories for auroral scattering are insufficient to explain all the experimental characteristics observed. The plasma acoustic wave model developed for equatorial field-aligned irregularities appears to fit most of the auroral data obtained but warrants further investigation.