Review of problems of correlations of structure and function in cells

Many of the correlations in cytology between structure and function compare the metabolism of subcellular fractions with the electron microscopy of the same tissues. Conclusions from subcellular fractionation have generally been derived on the assumption that killing the animal or plant, homogenisation, centrifugation and extraction of the tissues, do not affect the activities of the biochemical structures examined nor their distribution in the subcellular fractions; this assumption has never been tested. There is insufficient literature on the effects of reagents used in histology, immunocytochemistry and electron microscopy, on chemical activities, their distribution, or their visibility in sections. The artifacts of electron microscopy, such as shrinkage and precipitation, are widely recognised but not taken into account sufficiently when structures are examined. No attempts have been made to explain the two dimensional geometry of many 'unit' membranes, nor how a cytoskeleton can cause intracellular movements. Particular examples, such as the alleged location of oxidative phosphorylation along the mitochondrial cristae, the presence of biochemical ribosomal and lysosomal activities without the appearance of ribosomes and lysosomes, and intracellular movements in the believed presence of a cytoskeleton, are given to illustrate the difficulties of widely believed correlations between structure and function in cells.