M. C. Lea—Photobromide and Photoiodide of Silver. 493 
out in a few days or weeks, then could no longer be developed, 
but the film could receive a fresh image. This seemed an 
unanswerable proof of the physical nature of the latent image 
at least on silver iodide. The argument was: If the produc- 
tion of this latent image is the result of chemical action involv- 
ing the loss of iodine by the silver salt, how then is this 
iodine recovered when the image fades out? If it is formed of 
subiodide, where does this latter substance get back its iodine 
to return to the normal form, as it unquestionably does? 
No answer could be given then or after, and this experi- 
ment, repeated and confirmed by others, has always seemed 
the strongest support of the physical theory. When, however, 
it appears that silver iodide can take up iodine and hold it, the 
course matters follow becomes evident. By the action of light 
a very small quantity of subiodide is formed, and combines 
with the normal to form photoiodide. The iodine set free 
evidently does not pass off but remains combined with neigh- 
boring molecules of AgI, and in the dark gradually re-combines 
with the photoiodide re-converting it to normal AgI. In this 
retention the lower tension of iodine as compared with bromine 
and chlorine no doubt plays its part. 
In thus explaining away the fading out of the latent image 
on silver iodide, the last argument in favor of the physical 
theory is destroyed, while the chain of proof supporting this 
new explanation, that the latent image consists of normal haloid 
combined with its own subsalt, remains unbroken. 
Reactions—When ammonia is poured over purple photo- 
iodide of silver, the color quickly passes to a salmon and then, 
even after some days, seems to undergo no further change. Of 
separation of metallic silver as in the case of the corresponding 
chloride and bromide, there is no trace. 
In sodium hyposulphite it dissolves slowly, leaving a slight 
but distinct residue. 
Dilute nitric acid mixed with dilute solution of KI slowly 
but completely converts it to normal silver iodide. 
Light acts slowly upon it, changing the color to greenish- 
oray. 
I have already spoken of the remarkable manner in which 
light acts reversely by development on photobromide. On 
photoiodide the effect is usually the same as on photochloride, 
viz: the action is direct. But occasionally it reverses and the 
exposed part comes out lighter in development than the part 
that has not been exposed. In this respect the behavior of the 
. iodine compound is intermediate between that of the chlorine 
and the bromine. 
