“…They abandoned the use of muscle protein and of purified depot fat as reference compounds and substituted values determined statistically by regressing the calorific value and C content of a variety of ash-free tissues of the body on their N content. T h e 'protein' base-line was then taken to be the mean C content and calorific value of the complex mixture of compounds which has a N content of 16.0%, and the 'fat' base-line the C content and calorific value of that mixture which contains no N. T h e approach is probably sounder than the approach by Kohler ( 1900-I), Armsby (1903) and Brouwer (1965) but, as shown in Table 4, the various factors proposed, despite their variation, do not make very much difference to the final estimate of energy retention. When N retention is very high, Blaxter & Rook's factors predict energy retention to be about 2% lower than that predicted by Armsby.…”