110 CYTOCHEMISTRY OF PROTEINS 



this considerable quantity of suspended sections, after which the 

 sections are centrifuged down and the fluid phase then tested 

 for protein and nucleic acid. In this way the gross loss of mate- 

 rial from sections by diffusion may be detected. 



2. Where the components of the tissue which are being stained 

 can be obtained in a soluble condition, a very useful method is to 

 carry out the cytochemical reaction with this solution of the 

 components. Then a tissue section is exposed to the coloured 

 soluble component to see whether it adsorbs upon the section. 

 This reveals the sites of affinity within the section for the coloured 

 component. In cases where the distribution of sites of affinity 

 coincides with distribution of sites of the compound as revealed 

 by a cytochemical method, further enquiry is clearly indicated. 



3. Perhaps the most useful method of all is the use of superim- 

 posed sections, somewhat along the lines used in the case of 

 alkaline phosphatase. Let us consider as an example the dis- 

 tribution of tyrosine in a section. A section is placed on a slide 

 and treated with a blocking agent so that no residual tyrosine 

 groups remain in it. Then upon this is placed the section which 

 is to be examined, and the cytochemical reaction is carried out 

 with the superimposed sections. If any diffusion artefact occurs, 

 it should be revealed by the appearance of colour for tyrosine in 

 the underlying blocked section. 



Methods for the Electron Microscope 



In 1948 Mr. Walker and Miss Hanson, of Professor Randall's 

 Biophysics Research Unit at King's College, London, asked me 

 whether it would be possible to develop methods of the types 

 discussed above for use with the electron microscope. The physi- 

 cal problem involved is markedly different from that encountered 

 with the light microscope. In the light microscope perception of 

 an object is based upon non-transmission of light as a result 

 either of scattering or absorption, or both. If scattering is re- 

 duced to a minimum, and the component which it is wished to ob- 

 serve has an absorption spectrum markedly different from that 

 of any other component present, quite accurate estimations may 

 be made even of small absorptions of light. For example, an 

 absorption of 5 percent of the light passing through an object 

 may be made with an accuracy of better than 10 percent. On 



