S U M M A R YMitochondria1 structure in yeast cells under various physiological conditions has been studied by high voltage electron microscopy of sections that are 0.5 to 2-0 pm thick. Such thick sections of the yeast Candida utilis had a small number of long, branched tubular mitochondria per cell. The mitochondria extended into cell buds and unseparated daughter cells. It was apparent from parallel studies with thin sections that most of the rounded mitochondrial profiles viewed in thin sections should not be interpreted as being numerous small individual mitochondria. Attempts to study thick sections of the yeasts Saccharomyces cerevisiae and Schizossaccharomyces pombe were frustrated by poor contrast.
I N T R O D U C T I O NCopper deficiency in mice results in the appearance of giant mitochondria in the liver (Suzuki, 1969;Dallman & Goodman, 1970). Electron microscopy of thin (about 75 nm) sections of Candida utilis apparently showed, in addition, that copper deficiency caused replacement of numerous rounded mitochondria by a smaller number of much enlarged mitochondria, or giant mitochondria (Davison, Downie & Garland, 1972;Keyhani, 1973). However, yeast mitochondria can be tubular (Yotsuyanagi, 1955 ; Kawakami, 1961), and tubular mitochondria repeatedly crossing the plane of a thin section would appear as numerous individual mitochondrial profiles. Furthermore, if these tubular mitochondria then swelled, their appearance in thin section would reveal fewer but much larger mitochondrial profiles. To settle whether the typical mitochondrial morphology usually deduced from thin sections is in error, two techniques are available. One is serial sectioning and reconstruction by model building, and was used to demonstrate mitochondrial morphology in yeasts by Keddie & Barajas (1969), who reported that there were only one to four large branched mitochondria per cell. The other technique is high voltage electron microscopy (h.v.e.m.) of thick sections. We have used this latter method for studying the mitochondrial morphology of the yeast C. utilis. It is concluded that inspection of single thin sections leads to quite erroneous interpretations of mitochondrial size and number. Examination of thick sections by h.v.e.m. provides a simpler and faster method of inspecting mitochondrial morphology than does serial sectioning, and permits many samples to be studied. We describe our experience with this technique. A preliminary report has been published (Davison & Garland, 1974). We also describe the effects of copper and other nutrient limitations on the electron microscopic appearance of C. utilis.