A summary is given of the work done on the thermal behavior of air-cooled graphite channels. The differences in results between this work and work done by Robinson and Taylor (5) appear to arise from the different models used. The results reported here and re-evaluation of the Robinson and Taylor work indicate that their model is not confirmed by their experiments and does not predict the limit in permissible temperatures imposed by graphite oxidation in air-cooled graphite moderated reactors.