The effect of six fixatives on the areas of rabbit neurons and rabbit and rat cerebral slices

SUMMARY Cell bodies of cerebral neurons from rabbits were isolated by hand, transferred to a microscope slide in a ‘199’ medium, and the projected areas of their cytoplasm and nuclei were measured. In sixty‐four cells there was a strong correlation between the projected areas of the cytoplasm and the nuclei (r=0.66, P < 0001), and the ratio of the projected areas was 11.6. The medium was then replaced by the following fixatives: formalin (10% v/v), Bouin's, Carnoy's, Susas, glutaraldehyde (5% v/v), and osmium tetroxide (1% w/v). Cerebral slices were obtained from the grey and white matter of rabbit and rat, and were also measured before and after treatment with similar fixatives. Relative to the unfixed areas, glutaraldehyde and osmium tetroxide had no significant effect on the projected areas of isolated cells, Carnoy's fixative shrunk the areas of the cytoplasm by means of 16.26%, Bouin's by a mean of 49%, and Susas by a mean of 65%. The shrinkage of the cytoplasm and the nuclei was not significantly different from that of the nuclei for each of these three fixatives individually, but with formalin the mean shrinkage of the cytoplasm was 46% while the nuclei did not shrink significantly. Using the same fixatives the effect on the areas of the cerebral slides from rabbit and rat were as follows: glutaraldehyde and osmium tetroxide caused no change in area; Carnoy's, formalin and Bouin's fixative diminished the areas by a mean of 10–20%, and Susa's by a mean of 35%. It was concluded that a particular fixative often caused a different degree of shrinkage to the cytoplasm, nuclei and cerebral slice. In general, the lower the osmotic pressure of the fixative, the less shrinkage it induced.