What is the explanation for the changes to cobalt-60 tissue-air ratios in BJR Supplement 25?

Values of tissue-air ratio (TAR) in the recent British Journal of Radiology (BJR) Supplement 25 have been increased by nearly 2% over the values which have been accepted for the past 30 years. The need for this was shown by analysis of previous data using scaling laws, together with Monte Carlo calculations and careful re-measurement. However, it was not clear why previous determinations of TAR were in error: it was not, as some workers argued, because scattered radiation had been included in the absorbed dose in the miniphantom, because TAR data in BJR Supplement 17 had been derived from peak scatter factor (PSF), which is not based on the miniphantom concept. The purpose of this paper is to find the real explanation of why the PSF and, therefore, TAR were underestimated for so long. Two definitions of PSF are considered: one based on kerma and one based on dose. This paper relates PSF of either definition to measurements of air kerma by including in the derivation the scatter in the plug which replaces the chamber when it has been removed from the surface of the water phantom. The kerma-based PSF is found to be 2% higher than the simple ratio of chamber readings in phantom and in air. The value of the dose-based definition agrees with that of the kerma-based definition to within 0.2%. It is the scatter in the replacement plug in the surface of the water phantom which was effectively ignored by previous workers, and which explains the underestimates of around 2% in PSF and TAR. The value of the dose-based PSF differs slightly from that of the kerma-based PSF because of the different distributions of primary and scatter photon fluence.