No general hypothesis has been advanced to explain the underlying cause of the appearance of Heinz bodies in erythrocytes. The initial observation of Riess (1882) on a case of potassium chlorate poisoning was followed by that of Heinz (1890) who administered phenylhydrazine to animals. Since then Heinz bodies have been induced by a great variety of substances, principally aromatic and aliphatic nitro-compounds, aromatic amines and organic (notably azo) dyestuffs, a miscellaneous group of organic nitrogen-free compounds (e.g., naphthalene, 2 :4 dichlorphenol, ascorbic acid), and a number of inorganic substances (e.g., sodium nitrite, sodium sulphite, potassium dichromate, arsine). Heinz bodies are also found normally in cats (Hecht and Wingler, 1952) and their appearance has been observed after splenectomy in animals (Schilling, 1921 ; Freifeld, Schiiowa, and Ludwinow sky, 1937) and in human subjects (Zadek and Burg, 1930; Selwyn, 1955). The substance of Heinz bodies has been thought by many to originate from haemoglobin rather than the cell stroma, and recently Beaven and White (1954) have demonstrated that it is spectroscopically similar to insoluble green haemoglobin which is described as being com plex, containing haems and other iron-containing pigments bound to denatured globin . Associated with the appearance of Heinz bodies, though not invariably, are anaemia, methaemoglobinaemia, and verdohaemoglobinaemia : these and other topics have been exhaustively reviewed by Fertman and Fertman (1955). The use of azo dyes for the colouring of food has led to some preoccupation regarding the possible carcinogenic and toxic effects of the dyes or of the products resulting from reduction-fission at the azo linkage. The current need for more information on dye toxicity is emphasized in the report on colouring matters issued by the Food Standards Committee of the Ministry of Food (1954) wherein reference is made to the work of Hecht and Wingler (1952). These authors point out that although fission products such as aniline, toluidine, or xylidine may not have been proved to be carcinogenic, their other known toxic properties rule out the use of the parent dye in food. A suitable dye might be assessed from this point of view by estimation of the degree of methaem oglobinaemia produced, but Hecht and Wingler (loc.cit.) advocate instead the examination of the blood for Heinz bodies, as this procedure is simpler and more sensitive. These authors reported that no Heinz bodies were produced by azo dyes of the type / 1St1St_t> (R = sulphonated aromatic \_/ ~ component) when the aniline component was sulphonated. This finding accords well with the generally accepted view that sulphonation has a detoxicating effect, and enhances the value of the Heinz body test as an indication of the presence of, for example, free amines. In a later paper, Wingler (1953) found that methylation at the para position on the aniline component of benzene-azo-naphthol 6 and 6 : 8 sulphonates likewise removes Heinz body-forming activity. In the work described in this paper the activity of eight azo dyes (Table 1) has been observed in order to compare the effect of substituent groups in the benzene-azo-naphthalene nucleus. The dyes were administered in the food since an ultimate object has been to assess the significance of Heinz body formation in the problem of food colours for human consumption.