SUMMARY:Evidence is presented that a specific gelatinolytic enzyme besides collagenase is present in purified Clostridium histolyticum collagenase preparations. This means that collagenase itself need not be able to digest gelatin but it does not exclude the possibility.The purest collagenase available to date (DeBellis, Mandl, MacLennan & Howes, 1954) is free of non-specific proteolytic activity and active only against collagen and collagen degradation products (azocoll, gelatin). The question whether these activities are due to the same enzyme, i.e. whether collagenase will break down gelatin as well as collagen is still unsolved. It is possible that a series of reactions catalysed by distinct enzymes is involved but that our analytical procedures are inadequate for the detection of the primary effect of collagenase. This may involve simple breaking of hydrogen bonds as in urea treatment which results in denatured collagen susceptible to enzymes such as trypsin. However, the entire collagenase system differs from such proteinases by its specificity and complete lack of reaction with other proteins such as casein or haemoglobin as well as numerous synthetic substrates tested in this laboratory (Mandl, Zipper & Ferguson, 1957).The presence of a specific gelatinase was first suspected when fractionation of crude collagenase (DeBellis el aZ. 1954) invariably showed the fraction with maximum collagenase activity preceding the corresponding azocoll maximum, If the collagenase were the only enzyme in this region capable of attacking azocoll, the maxima should coincide. Furthermore, the ratio of collagenase to azocoll activity is not the same in different preparations and the concentrations of certain ions (Co, Ni) needed to inhibit activity of any one preparation differ with the substrate hydrolysed (see Table 1).Ogle & Logan (1956) have also advanced evidence for the presence of a distinct gelatinase in the collagenase system. Gelatin or azocoll attack by the preparations investigated could not be ascribed to other known proteinases (MacLennan, Mandl & Howes, 1958;DeBellis et al. 1954) since the fractions were free of S-proteinase and y-(or e-) proteinase would not be active in the absence of cysteine.In the belief that serological tests would be more conclusive the following set of experiments was set up: varying dilutions of six sera kindly supplied by Professor Oakley of Leeds University and used for the detection of other specific enzymes (MacLennan et a2. 1958; Oakley & Warrack, 1950) were incubated for half-hour periods with equal volumes of 0.1 yo solutions of an enzyme fraction rich in collagenase and free of 6-proteinase prepared according