222 Prof. A. B. Macallum. On the Nature of the [Mar. 3, 
of silver, seemed to indicate very distinctly that the coloured silver compound 
produced in solutions of proteids is due to the latter, it being supposed that 
an “albumenate” of silver obtains which in light forms a compound, 
argentous “albumenate,” analogous to the argentous haloid salt of the 
photographic plate.* 
All these facts make it possible to understand how it came about that there 
was any discussion as to the nature of the silver reaction in tissues, and why 
it was accepted without much question that proteids took a part in this 
reaction. The author has thought the subject worthy of further investigation, 
on the ground that if the silver reaction is due to the presence of few or 
many organic compounds, as well as to the presence of haloids, the accurate 
localisation of chlorine, bromine, and iodine in tissues is possible only under 
great difficulties, whereas if the reaction is due to haloid salts, or can be 
applied so as to demonstrate the presence of haloid salts only, the cyto- 
chemist has a means of determining the presence of not only chlorine, 
bromine, and iodine, but also to a certain extent of sodium and potassium in 
tissues. 
It has for several years been the view of the author that the reaction of 
proteids with nitrate of silver in sunlight is due to the presence of chlorides 
only, and that if proteids could be thoroughly freed from chlorides, the forme1 
would give no reduction compound with the silver salt. The demonstration 
of this, when attempted, was beset with difficulties, due partly to the fact that 
chlorides, and particularly the chloride of sodium, are present everywhere, 
and, therefore, contaminating, as they do more or less, every reagent and 
preparation, the absolute removal of chlorides appeared to be impossible of 
accomplishment, and in part, also, to the extreme sensitiveness of the reaction 
which demonstrates the presence of the haloid salts. Silver nitrate will 
demonstrate the presence of one part of chlorine in 1,000,000 parts of water, 
but I have found, also, that if sunlight is allowed to act on the preparation, 
one part of sodium chloride may be detected in 1,000,000 parts of water—that 
is, 1 part of chlorine in 1,600,000. It is manifest that this test is exceedingly 
delicate, and consequently the reaction, if properly sought for, could be 
obtained in all fluids, and particularly with colloids, which are very tenacious 
of the inorganic salts with which they are associated. Because of this, it 
appeared hopeless to attempt to free proteids from haloids, and thus demon- 
strate the inactivity of proteids towards salts of silver. 
The results of researches recently carried out regarding the detection and 
localisation of potassium salts in animal and vegetable tissues, having 
indicated how important it is to determine whether anything else than 
* Meldola, op. cit., pp. 116 to 119 and 342 to 352. 
