The wall of the conidium of Erysiphe cichoracearum was composed of two layers. The inner layer was probably chitinous and possessed a few spines, whereas the outer layer was three or four times thicker and gelatinous. The end wall of the spore often was incompletely covered with this gelatinous layer. The cytoplasm of the dormant conidium had electron-dark and electron-light areas. It contained mainly spherical or elliptical mitochondria, spherical lipid bodies, which were usually next to the conidial wall, a limited amount of endoplasmic reticulum, and occasionally storage bodies, perhaps glycogen. The chondriosomes appeared to reproduce by division. A large proportion of the volume of conidia was occupied either by a few large vacuoles or several small ones. The vacuoles occupied the central region of the conidium and contained varying amounts of heterogeneous material which could be storage and (or) waste substances and probably most of the water in the spore. Although neutral red did not stain the vacuole, the tonoplast appeared typical in electron micrographs. The nucleus was about equidistant from the ends of the mature conidium and contained a lateral granule which was frequently opposite to a prominent large nucleolus. Crystal violet, sudan black B, Giemsa, and Feulgen stains revealed that two or three unknown bodies were present in the nucleolus. An unidentified body was found also in the nucleoplasm. The Feulgen reaction was positive after fixation with glutaraldehyde but not after the fixatives of Helly and Farmer. When the last two fixatives were used, the nucleoplasm and chromatin were withdrawn a considerable distance from the nuclear membrane. Thus the fixative not only changed the morphology of the nucleus but also its chemical structure.