1. ARPS welcomes the opportunity to comment on the draft report. 2. ARPS recognizes the expertise and knowledge reflected in the report, and the quality of writing, and commends the Task Group on producing a valuable reference document. 3. However, ARPS wishes to invite the Task Group to consider the following comments and suggestions in the interests of completeness and of anticipating future criticism. (a) It could be useful to include in the Task Group report a fuller discussion of reports in the literature concerning other possible dose-response relationships at low doses, including radiation hormesis. For example, Section 2.4.3 ii), dealing with occupational epidemiological studies, mentions in passing the work of Smith and Doll who studied the health of radiologists, but the results of the study are not analysed. Similarly, there is a well-known study of shipyard workers in the United States (Matanoski), that appears to show evidence of better health among exposed workers than unexposed workers, that the Task Group report does not mention at all. There are many other reports, such as the paper presented at IRPA10 by Mitchel and Boreham, that merit comment. In order to assist in resolving as far as possible the differences of opinion concerning the adoption of the LNT hypothesis, it would be helpful for the Task Group either to provide an analytical review of pertinent studies or to make a statement as to why they do not deserve attention in this report. Otherwise, an impression can be created that the report is selective in the work it assesses. (b) For similar reasons, the Task Group might consider commenting upon any evidence that is available concerning the health of populations living in areas of high background dose. There appears to be a lack of evidence of harm (which of course is not evidence of lack of harm) in populations receiving from five to more than fifty mSv per year. The discussion in Section 2.4.2 argues that enormous study populations would be needed to discern statistically significant effects at low doses, but our understanding is that the population numbers given (e.g. 624,000 at 10mGy) relate to a single dose. What would be the corresponding number for a population receiving 10mSv every year? (c) The Task Group report focuses principally on low LET radiation. However, the extensive literature concerning health effects of radon and its daughter products, and of inhalation and ingestion of other radionuclides, is of relevance in making decisions of judgement to adopt a LNT hypothesis for the purpose of establishing radiation protection systems. In particular, the very recently published work of Darby et al (BMJ 2005; 330:223-226) suggests statistically significant results for radon concentrations in homes corresponding to a few mSv/y in terms of effective dose. Recognizing that the draft Task Group report predates the publication of Darby et al, it would be helpful for the report to now make some comment on this and the earlier studies. (d) It may well be that the above matters are substantially dealt with in other analytical reviews, such as UNSCEAR 2000 and NCRP 2001. The comment in Chapter 1 (p.15) concerning the focus on ‘updated coverage’ is noted. Nevertheless, a reasonably informed but neutral reader of the Task Group report may anticipate at least a summary analysis of these other areas of information and be puzzled by their absence. A less neutral reader might suspect selective reporting, and that would be unfortunate.