NON-COAGULANT PRIMARY FIXATIVES 47 



ev idence, but it has been shown that osmium tetroxide can in fact 

 join certain double-bonded ring-compounds together.^"- ®^ The 

 double bond is lost in the process. An extra supply of oxygen is 

 required if this reaction is to occur. 



Although reactions of this kind are probably concerned in the 

 fixation of proteins by osmium tetroxide, yet there are indications 

 that others also play a part. Tryptophane and histidine are not the 

 only free amino-acids that form a dark precipitate with osmium 

 tetroxide: cysteine (p. 36) behaves in the same way,^ yet its side- 

 group possesses no double bond. 



There is no reason to believe that tryptophane and histidine 

 play any important part in the colouring of proteins by dyes, yet 

 it is characteristic of the proteins of tissues fixed by osmium 

 tetroxide that they show a remarkable lack of affinity for acid 

 dyes.^^ Since these dyes attach themselves to the -NHo groups of 

 the side-chains of basic amino-acids (especially lysine, p. 19), 

 there is strong reason to believe that osmium tetroxide somehow 

 blocks these groups. The mechanism of such blocking is, how- 

 ever, unknown. Some proteins that contain a lot of lysine and 

 arginine darken readily with osmium tetroxide, but the prota- 

 mines, which contain a particularly high proportion of arginine, 

 seem scarcely to darken at room-temperature.^ It would be 

 unjustifiable to assume that darkening is an invariable con- 

 comitant of reaction. 



Reactions with nucleic acids. Does not precipitate DNA from 

 solution."^ 



Reactions with lipids. Osmium tetroxide blackens unsaturated 

 lipids of all kinds, but does not attack saturated ones.^ It is there- 

 fore clear that it reacts at the double bonds. Nearly all lipids in 

 organisms are mixtures containing some unsaturated components, 

 and blackening is therefore almost invariable, though it may be 

 very slow. 



Osmium tetroxide is soluble in lipids. It may therefore enter 

 a saturated lipid and subsequently be reduced to black osmium 

 dioxide (p. 45) when the tissue is placed in ethanol. It follows that 

 the blackening of a lipid droplet does not necessarily prove that 

 the lipid was unsaturated. 



