Radio-quiet and normal radio-loud quasars have very similar spectral properties in the ultraviolet, optical and near infrared regions, but their radio powers differ by several orders of magnitude. Somewhere between the near infrared and the radio their spectra must diverge dramatically.

The IRAS survey detected 17 radio quiet quasars and luminous Seyfert 1's with −29. 5 ≤ Mv ≤ −21.6 (for Ho = 75). By coadding the survey data and using pointed observations, we have detections of most of these objects in all four IRAS passbands. The spectra are all rising with wavelength all the way to 100μ. We are measuring fluxes in the centimeter, millimeter, and, together with R Cutri, the near infrared and optical regions for each of these objects. Our goal is to constrain the location, shape, and spectral context of the low frequency cutoffs. Here we present the IRAS, millimeter and centimeter data. Measurements at the other wavelengths are still in progress.

Although the spectra are rising steeply between 60μ and 100μ, we find that all of our objects are undetectable at 1.3 mm with the NRAO 12-m telescope. Our limits are typically an order of magnitude below the 100μ fluxes. (Ennis et al (1982) and Robson et al (1985) have already shown that the 1 mm fluxes of some radio quiet quasars must be below the extrapolation of the near infrared continuum.) Our objects are all extremely weak or undetected with the VLA at 2 cm and 1.3 cm, at levels typically three orders of magnitude below the 100μ fluxes. The sharpness of the required cutoffs allows us to rule out the hypothesis that the infrared is synchrotron radiation with the cutoff due to an absence of low energy electrons. The high frequency of the cutoffs makes free-free absorption implausible, but not impossible. It is possible that synchrotron self-absorption is suppressing the radio. There is circumstantial evidence that the far infrared is thermal dust emission. This would require a lower cutoff in the distribution of dust temperatures, which we think we can explain.