There are many reports in the literature, some of them considered in a recent series of reviews (Schultz, Day & Sinnhuber, 1962), demonstrating that feeding with rancid fat can in some circumstances have growth-depressing or other unfavourable effects on experimental animals. Thus the practice has grown up of characterizing the fatty fraction of animal feeding-stuffs by simple procedures (in the past these have been the determination of free fatty acids and peroxide value) and rejecting materials that gave values above certain limits.However, there seems no evidence that free fatty acids as such are harmful to animals (though they may reflect the existence of mouldiness in certain classes of materials); moreover, the permitted levels of 'peroxide' have been only a small fraction of the high levels found harmful in diets under experimental conditions. It is known that the presence of oxidizing fat in the diet may cause severe losses of easily oxidizable vitamins, particularly E and A, when these are not provided in stabilized form. However, there is little direct experimental evidence for harmful effects arising from the use of fats, or fat-containing ingredients, within the range of oxidation likely to be encountered in commercial samples and in well-balanced rations prepared with the precautions now available to the compounder. Practical people, nevertheless, have the impression from field observations that poor results are often seen with rations containing unsaturated fat that have been stored for prolonged periods .In a previous paper (Carpenter, Lea & Parr, 1963) we have reported the last of a series of experiments with herring meal (containing approximately 17 % lipid, that is, the chloroform-methanol extract), which had been allowed to oxidize during storage before it was given at a level of about 18 % to chicks. There was no evidence in these short-term experiments with generous vitamin supplementation of any direct toxic effect of the oxidized fish oil; the slight depressing effect on weight increase observed with the stored, compared with the fresh, meal accompanied the non-lipid rather than the lipid fraction on separation. Small losses of available lysine had previously been shown to result from an interaction between the protein and lipid oxidation products in stored herring meals (Lea, Parr & Carpenter, 1958, 1960. Moreover, addition to the chicks' diet of 2-5 yo of herring oil pre-oxidized to a peroxide value of 142 ,umoles/g available at https://www.cambridge.org/core/terms. https://doi