472 



Messrs. S. E. Sheppard and C. E. K. Mees. [Dec. 20, 



On the Theory of Development. In the ferrous fluoride developer the 



+ + 



reducing agent seems most probably the ferrous ion, Fe, the concen- 

 tration of which is extremely high. With ferrous citrate, a complex 

 ferro-citratanion. Now, on the diffusion theory the great difference in 

 velocity observed is difficult to explain ; in particular, the mobile 

 ferrous ion should have proved the most efficient reducer. Possibly 

 we still have to do with the influence of free ferric ions on the 

 reduction velocity, although their original concentration in the ferrous 

 fluoride solutions was extremely small, while those formed by the 



reaction Fe + Ag z Fe + Ag (metal), should be immediately removed 

 by the ionic equilibria 



and 



Fe + 3F ^ FeF 3 

 Fe + 6F ^ FeF 6 . 



However, the action of bromides shows that the actual chemical 

 reaction can, under certain circumstances, influence the total reaction 

 velocity. Although, in general, ionic reactions, such as we must 

 conceive development to be, occur with practically immeasurable 

 velocity, yet it is possible to follow some in which the ionic concen- 

 tration is very low.* In the development of silver bromide, the 

 concentration cannot pass the solubility product of AgBr, so that the 

 maximum concentration of silver ions is very low. Further, the 

 unknown nature and influence on the reaction of the so-called latent 

 image have not been taken into account. 



Bodenstein,f in a criticism of the Nernst-Bruner theory, has 

 indicated the presence of a high temperature coefficient as a criterion 

 for distinguishing whether the speed of reactions in heterogeneous 

 systems is conditioned by the diffusion process or the chemical reaction. 

 A study of the temperature-influence, of the influence of soluble 

 bromides, the reversibility of the reaction, and of the microscopy of 

 development and exposure has been made, in continuance of the 

 present work, as well as experiments on the nature and destruction of 

 the " latent image," and it is hoped to publish these shortly. 



The work given here shows that development can be studied 

 quantitatively, and brings it in line with general physico-chemical. 



Summary of Results. (1) An accurate quantitative method of 

 attacking photographic questions is described, and it is shown that if 

 T be the optical transparency of the silver deposited in photographic 

 processes, the D = - log T is accurately proportional to the mass of 

 silver. 



* Cf. Dontian and Le Rossignol, "Velocity of Reaction between Potassium 

 Ferricyanide and Iodide in Neutral Solution," ' J. C. S. Trans.,' p. 703, 1903. 

 f M. Bodenstein, ' Zeit. f. Physik. Chem.,' 49, H. 1, p. 42, 1904. 



