12 MICROSCOPIC TECHNIQUES 



substance before serious diffusion could occur. Still, in the case 

 of the precipitation of tissue chloride by silver nitrate solution, Scott 

 and Packer (1939) pointed out that differences in ionic mobilities 

 and the effects of ionic charges at cellular interfaces might easily 

 produce precipitations in regions different from those in which 

 the chloride originally existed. On the other hand, Gersh (1941) 

 claimed that the results he obtained from chloride distribution, 

 using silver nitrate as the reagent, were valid as borne out by 

 related data obtained with entirely different biochemical methods. 

 Regardless of the merits in this particular instance, the dangers 

 indicated by Scott and Packer cannot be ignored, and no way has 

 yet been devised to really eliminate them ; they constitute a funda- 

 mental limitation in the application of the chemical methods of 

 microscopic technique. 



In the special case of the localization of enzymes, the sites of 

 activity may be determined in tissue sections by immersing the 

 sections in a buffered substrate medium containing a reagent which 

 will bind one of the products of the enzymatic action in situ by 

 precipitation. In addition to the four requirements already listed 

 for determinations of the disposition of tissue constituents, it is also 

 necessary that the following be included for enzyme methods: 



5. A reagent which when added to the buffered substrate will 

 react with one of the enzymatic products but not with the substrate 

 or buffer. 



6. A reagent which will also have no untoward effect on the 

 enzyme. 



7. If the enzymatic product which reacts with the reagent is a 

 substance pre-existing in the tissue, either the sites of enzyme action 

 must be different from those of the pre-existing substance, or the 

 increase in the amount of the visualized compound resulting from 

 the enzyme activity must be demonstrable, or, better yet, the sub- 

 stance must be removed in advance by a method which will not 

 take out the enzyme. 



8. A control experiment in which either the substrate is omitted, 

 or a highly effective soluble enzyme inhibitor, such as fluoride, is 

 added (the inhibitor must not react with substrate, buffer, reagent, 

 or products") — the advantage of the inhibitor is that, in some cases, 

 naturally occurring substrate may be present with the enzyme and 

 thus give a false aspect to the nonenzyme control. No such control 



