Veronal is a weak acid (diethylbarbituric) with a dissociation constant of approximately 10~8 (ref. 2); and Michaelis5 introduced its sodium salt as a buffer for use between pH. 7 and 9. Sodium acetate is effective in buffers on the acid side of neutrality (see refs. 4 and 6; Fig. 1). Michaelis2 mixed sodium acetate and sodium veronal in one buffer for his interests in enzymology so that he could get a range of pH from 2-9 without variation in ionic strength. The graph in Fig. 1 shows that sodium acetate in the acetate/veronal buffer of Michaelis has no influence on the H at which fixatives for electron .microscopy are generally held (7-2-7-5). The acetate is effective in the range between pH about 3-6. It is therefore not obvious why acetate should be included in buffers used for making up fixing solutions intended to be used at pll 7-3-7-5. Veronal seems to have little, if any, capacity to buffer between pH. 3 and 6-5; and if the fixing solution held at a pH. slightly on the alkaline side of neutrality were for some reason to absorb a small quantity of acid, its pH. might fall quite considerably and make the solution definitely acidic. If, however, acetate were present in the same solution, it could take over the buffering from veronal when the pTL fell to just on the acidic side of neutrality and thus keep the solution nearer neutrality. It is difficult to say whether so much acidity is produced during fixation of minute pieces of tissue that acetate is brought into play as a buffer. Experiments were therefore made on fixation by buffered osmium tetroxide prepared exactly according to the formula given by Palade1 except that sodium acetate was omitted from the buffering solution. Pancreatic exocrine cells, convoluted tubule cells of the kidney and testes of the mouse were used as test-objects. The characteristic appearances of the cell inclusions that have been related with acidity do not seem to be obvious in the electron micrographs, if the tissue is embedded in a suitable embedding medium, for example, 'Epikote 812'. Comparable micrographs have also been obtained after embedding in partially prepolymerized methacrylate. The preservation particularly of membranous structures considerably deteriorates if monomer methacrylate is used. An experiment was also done to find out whether the presence of sodium acetate in Palade Js fluid was helpful for better preservation of tissues, inasmuch as it increased the tonicity of the fixing solution. A neutral salt, potassium nitrate, was used as a substitute for sodium acetate in the veronal stock solution, so that the final fixing solution was nearly isotonic with Palade's fluid (depression of freezing point 0-308 C). But the quality of preservation visualized in the micrographs did not seem to be markedly different from that produced by fixation without the addition of potassium nitrate to the veronal solution. Fig. 1. The graph indicates the effective range of buffering of sodium acetate (M/7) (aa) and sodium veronal (M/7) (66) in separate solutions and together in one solution (M/7) (cc), at the concentration used In Michaelis's buffer. 5 ml. of the stock solution of the salt (or salts) and a ml. of N/10 hydrochloric acid are diluted with 20-a ml. of distilled water to make up the buffering solutionSimilar experiments were also made on the exocrine pancreas and convoluted tubule cells of the kidney of the mouse by fixation in potassium permanganate buffered by veronal buffer, instead of veronal/acetate as used by Luffc3. The appearance of the various cell inclusions was consistent with what was already known about these cells. A detailed account of these experiments will be published elsewhere7. However, the quality of fixation is judged by the quality of preservation seen in electron micrographs; and the micrographs of the pancreatic exocrine cells and the convoluted tubule cells of the kidney look the same whether fixation is done by a carefully buffered solution of osmium tetroxide held at pH 7-3-7-5 or by solution of osmium tetroxide simply turned alkaline by a small addition of potassium acetate (pH. 7-2-7-5). These results therefore seem to suggest that a solution of osmium tetroxide alkalized in this way might well be substituted for buffered solutions in the investigation of some other cells8.Furthermore, it seems doubtful whether it is always necessarily profitable to use buffered or alkalized fixing solutions for electron microscopy. Adequate preservation of the pancreatic exocrine cells and the convoluted tubule cells of the kidney has been obtained by fixation in simple unbuffered and unbalanced solutions of osmium tetroxide and subsequent embedding in 'Epikote 812' or partially prepolymerized methacrylate7*8'10. Electron micrographs of the tube feet of the sea-urchin, Echinocyamus pusillus, which are only about 35[i in diameter11, fixed by simple distilled water solutions of osmium tetroxide or potassium permanganate and embedded in 'Araldite'12 or 'Epikote 812* do not seem to be deficient in criteria that are generally correlated with satisfactory fixation. It did not seem to make any"appreciable difference in the quality of preservation whether isotonic or hypotonic fixing solutions were used, if methacrylate (partially prepolymerized) was used for embedding. All the fixing solutions mentioned here that have been used for fixation of the pancreas and kidney of the mouse are greatly hypotonic to the body-fluids of mammals, as is also Palade's fluid, which is so widely used in electron microscopy. But this does not seem to have effected the preservation in the micrographs. These results seem to cast doubts on the importance of invariable use of isotonic fixing solutions; this conclusion is substantiated by the experiments on the tube feet of Echinocyamus.This work was done during the tenure of a senior studentship of the 1851 Exhibition and a research fellowship at New College, Oxford. The work on Echinocyamus is being carried out in collaboration with Dr. D. Nichols.