First of all, we would like to express our appreciation for the possibility which has been offered to the community to comment on the recommendation draft. Our concern is mainly with Chapter 3 (Quantities used in radiological protection). Many choices made by the Commission since 1990, and largely maintained in the present draft, appear to be dictated by an effort to define a quantity - effective dose - directly representative of biological radiation effects as measured in vivo, regardless of the physical radiation field present inside the body. However, the Commission has followed a different approach in the following cases: 1) whenever in vivo biological data are lacking 2) to define operational quantities 3) for internal irradiation The dual system so defined presents several drawbacks: 1) Lack of consistency. Biological uncertainty does not justify ambiguous or contradictory definitions, and stability should not be invoked to avoid correcting past errors. On the contrary, the fact, often mentioned in the draft report, that biological effects are only poorly known, should allow a definition of protection quantities based on a fully coherent physical model. 2) Lack of universality: a given type of radiation should not be treated in a way dependent on how or where it is generated (externally or internally), but on the radiation field (spectrum of secondaries, dose distribution) it creates inside the body. 3) Lack of simplicity. The Commission's effort to simplify the system by defining only a few values for the radiation and tissue weighting factors, useful perhaps (although not entirely obvious) in the field of regulations and measurements, is voided by the complication that the two different approaches create in other fields, such as dose calculations and shielding design. It must be noted that at least in the case of high-energy accelerators the current approach to radiation protection is widely based on the use of fluence-to-dose conversion coefficients: present day computers allow to calculate the latter with any practical level of accuracy even if the coefficients are based on complex physical models. Thus, simplicity is welcome in the sense of having to deal with a single, coherent system, not necessarily as mathematical simplicity. Again, thank you for the opportunity to offer these comments. If you would like to discuss these comments further, please do not hesitate to contact us. The accelerator community within the United States and in fact, world wide, is very interested in the recommendations established by the ICRP and welcomes the opportunity to participate in this process. Sincerely yours, W. R. Casey - Brookhaven National Laboratory Donald Cossairt - Fermi National Laboratory Alberto Fasso - Stanford Linear Accelerartor Center Nolan Hertel - Georgia Institute of Technology Sayed Rokni - Stanford Linear Accelerartor Center Working Group - Accelerator Section, Health Physics Society