Polarization measurements, in addition to their interest for deciphering the internal physics of radio sources, are relevant to cosmology because estimates of ne, the thermal electron density, may be obtained from measurements of the Faraday effect. We have:-

The very low upper limits on distributed intergalactic (i.g.) HI (Gunn and Peterson 1965, Wampler 1967), and H2 (Field et al. 1966) have made it clear for some time now that if a smooth distributed i.g. gas exists in significant amounts (Ωig ≳ 10^−3), it must be ionized. X-Ray emission from rich clusters of galaxies such as Coma, indeed show that an intracluster gas can be seen (cf. Field 1974). Also persuasive in this connection are the head-tail radio galaxies in clusters, whose prototype is NGC 1265 (Miley et al. 1972, Jaffe and Perola 1973).

So far, we have mainly discussed what might be called classical methods for investigating cosmological models using radio observations. These mostly involve the number counts as a function of measured flux, angular diameters, and spatial distribution of radio sources at cosmological distances, as well as the measurements of the microwave background. In this review, I have been asked to give a brief description of some new, or non-classical, measurements that might be made. It seems that such measurements fall into one of three categories: Observations which are being made now and which may bear fruit in the next several years, observations which may be possible in the intermediate future several decades from now, and observations which may be possible in the far future - if ever. Naturally I'll try to concentrate on the first two categories since they will be of more interest to most living astronomers during their working lifetimes. Also, since time is short and it is very difficult to predict the future, I'm afraid that this review will be somewhat impressionistic rather than exhaustive. I hope people will describe additional methods during the discussion.

Half a century after Hubble expansion was recognized and a quarter century after the identification of Cygnus A we must acknowledge that progress in observational cosmology is slow, the efforts are laborious, corrections are subtle and correlations elusive. Part of progress is the realization that there are no short-cuts, no easy methods or quick successes. The result of this awareness has been increasingly careful attention to the completeness of samples, to selection criteria and the hazards of intercomparisons from survey to survey. This week in Cambridge has shown both modest progress and progress in modesty.