The extraordinary sensitiveness of silver bromide in gelatin seemed the more remarkable since it was known that silver bromide in collodion is only moderately sensitive. The explanation was sought for in various directions, but as the result of numerous investigations it appears that the chief cause of the difference is the presence of different modifications of silver bromide. From a consideration of the work already done on the subject, Vogel suspected that silver bromide precipitated in an aqueous colloidal liquid would have notably different properties from silver bromide precipitated in an alcoholic colloidal solution. Silver bromide was prepared in many different ways. Emulsions were made in bromide solutions containing gelatin or collodion (the former aqueous, the latter alcoholic), some with the aid of heat, others without. Part of the emulsion was then poured upon plates kept at a moderate temperature and dried. The remainder was boiled or treated with ammonia before being applied to the plates. He also precipitated silver bromide in dilute gelatin or collodion solutions, allowed it to settle completely, washed the precipitate, and mixed it with a new portion of gelatin or collodion before applying it to the plates. Finally he precipitated pure silver bromide, in the absence of all colloids, by means of pure aqueous or alcoholic solutions of bromides and attempted to bring this upon plates, using gelatin or collodion as a cement. The result of all these experiments is that there are essentially two modifications of silver bromide, the one being obtained by precipitation in aqueous, the other in alcoholic solutions. The first, on account of the position of the maximum of sensitiveness for the solar spectrum, he calls blue sensitive, the other, for the same reason, indigo sensitive. It is of no consequence whether the aqueous or alcoholic solution in which the silver bromide is formed contains gelatin or collodion, or whether the precipitation is effected with excess of bromide or of silver nitrate. It makes no difference whether the solution is hot or cold, or whether the silver bromide is treated with ammonia or whether it is boiled or not. The only necessary condition is that in precipitating indigo sensitive silver bromide the solutions must contain at least 96 per cent of alcohol. From aqueous alcoholic solutions blue sensitive silver bromide is precipitated.