A study of recent literature on haze-and foam-forming substances in beer suggests the following picture:Traces of true protein, apparently above 150,000 in molecular weight (MW), exist in beer. Apart from these, the main complex nitrogenous substances are present in compounds in the 5,000-70,000 MW region. Nevertheless, these are not proteins, but amino-acid complexes (proteoses) coupled with non-nitrogenous substances.As is well known, haze and haze precursors are proteoses coupled with polyphenols. These two species can be joined by hydrogen bonds to give compounds which are readily dissociated. However, a much firmer, covalent, linkage can be formed. This type of linkage arises from oxidation of polyphenols to quinones, which in turn can oxidize and couple with several groupings on proteoses. One of the most probable points of attack is the sulphydryl group of cystine, and the relatively high percentage of cystine in haze indicates that this occurs. It seems likely that this reaction, to form acidic complexes, occurs mainly in the initial stages of mashing and boiling. In beer, further coupling between pairs or triplets of polyphenolic or quinolic groupings can take the products into the very high molecular weight region of hazes.Correspondingly, the foam-enhancing proteoses tend to be lower in cystine content and to show less combination with polyphenols. However, they too are largely in the form of complexes and the evidence suggests that here the proteoses are mainly combined with glucose polymers, a reaction which takes place during kilning and possibly during boiling. These complexes are less acidic than the poly phenolic ones and tend to be rather lower in molecular weight.