THE DETECTION OF DIFFUSION ARTEFACTS 109 



3. Another procedure which we have under investigation is to 

 link a sugar onto the compound which is being coloured. To se- 

 cure this coupling it is necessary to have a sugar or related com- 

 pound which has been linked with a phenol or an amine in such 

 a way that the resultant compound is competent to react with a 

 diazonium hydroxide. Thus the actual mode of operation is to 

 produce a diazonium hydroxide in the section, possibly a diazo- 

 tised azo dye as suggested in method 2, and then to couple on the 

 sugar compound just mentioned. When the coupling is com- 

 pleted the sugar may be oxidised with periodic acid. This will 

 give aldehyde groups where previously there were a,/3-glycol 

 groups. The aldehyde groups are then treated with reduced 

 fuchsin, which, of course, gives a vigorous colouration. We are 

 at present experimenting in this way with a derivative of inulin. 

 The use of such a polysaccharide is, however, complicated by 

 steric factors, and to what extent the use of such compounds will 

 prove satisfactory cannot at present be predicted. 



The Detection of Diffusion Artefacts 



All the methods for the study of proteins and nucleic acids 

 which have been discussed have involved the formation of dyes 

 in the sections which are attached by stable covalent bondings to 

 the compound which is to be identified. Consequently, there are 

 no problems arising from the actual diffusion of the dye itself, 

 or from the adsorption of the dye by other compounds in the 

 section. All the methods are so designed that each individual re- 

 agent is a small colourless compound, the excess of which can be 

 washed out of the tissue section with comparative ease. Conse- 

 quently, there should be little trouble arising from unwanted 

 colour produced by residual amounts of reagents. The most seri- 

 ous type of diffusion artefact which is likely to arise under these 

 circumstances is that due to diffusion of the protein or nucleic 

 acid element itself. There is a very serious possibility that such 

 diffusion may occur despite all that can be done in the way of 

 fixation. To detect artefacts of this nature, three methods are 

 in use. 



1. To test for the loss of appreciable amounts of components 

 by diffusion, about 0.2 gram of sections is suspended in the vari- 

 ous reagents. Each step in the procedure is then carried out on 



