Abstract 1. (1) Hydrogen peroxide and silver nitrate, and possibly other soluble noble metal salts acting on a photographic plate after exposure and before development increase the developability of the light-exposed portion above its normal developability and in greater proportion than that of the unexposed portion. This we have called intensification of the latent image. 2. (2) Hydrogen peroxide and the salts just mentioned seem to be unique among the numerous substances we have used for producing intensification of the latent image, in that they are the only ones which give a strong intensification when the concentration is low enough not to produce objectionable fog. This effect is, however, approximately equivalent in degree to that produced by a brief flash exposure of the plate to light before or after the principal exposure. 3. (3) With the same concentration of hydrogen peroxide and time of treatment, the latent image on the fast plates which we have used was affected to a greater extent than on medium-speed plates, while on the process and desensitized plates used it was not affected at all. Eastman portrait and N. C. films behaved like the process plates. 4. (4) The latent image on the process plates, desensitized plates, and the film just mentioned could be intensified to a certain extent by increasing the hydrogen peroxide concentration, or by lengthening the time of treatment over that used for the fast plates. 5. (5) It was possible to produce no more than a slight latent fog with hydrogen peroxide, i.e., with no light exposure, on desensitized plates, no matter what concentration or time of treatment were used. 6. (6) Water caused an apparent intensification of the latent image. This is explained as partly due to swelling of the gelatin, allowing more rapid penetration of developer. 7. (7) The ratio of gamma of the peroxide-treated to the untreated plates usually increased with increasing development time. 8. (8) Intensification of latent image in most cases increased somewhat with increasing development time when measured as a density effect and decreased when measured by H. and D. speed difference. 9. (9) Experiments have shown that as little as I part hydrogen peroxide in 10,000,000 of water can be determined to within 20 to 30 per cent. by the latent image intensification method, when it is known that other intensifying agents and oxidizing substances are not present. 10. (10) The hypothesis, previously stated by Wightman, Trivelli and Sheppard, that light produces not only a developable but also an undevelopable latent image, and that the peroxide action makes this latter image developable in producing latent image intensification is supported by the present work. An extension of this hypothesis has been found necessary to explain the action on the latent image of fast, medium and slow plates, namely, that there is less undevelopable image formed by light in the medium and slow plates and desensitized high-speed plates and it is for this reason that the same concentration of peroxide and same time of treatment give different degrees of intensification on these plates of different speed.