A method of securing serial sections for electron microscopy is described. Serial sections present certain anomalies of interpretation of a nature such that a complete and detailed three-dimensional reconstruction of the sectioned tissue cannot be made. These anomalies are discussed, as well as those which have been encountered in the interpretation of single sections. Observations of the following kinds have been made in an attempt to elucidate the interpretation of single and serial sections: differing methods of mounting adjacent sections, observation of the same section by high-angle stereoscopy, and examination of sections which have been shadowed prior to and subsequent to electron microscopy. It is found that the appearance of sections is independent of the choice of side to be placed against the formvar films. Stereoscopy shows that the appearance of fine structures is strongly dependent upon the direction of the penetrating electron beam with respect to the plane of the structures. Stereoscopy, combined with shadowing, shows quantitatively that extensive sublimation of polymer occurs upon normal exposure in the electron microscope. Observation of sections shadowed prior to electron microscopy indicates that varying amounts of material are removed between sections by the action of microtomy; i.e., it is probable that the sum of the thicknesses of several serial sections is considerably less than the total thickness of material removed from the block. It is believed that this effect, combined with the effect of sublimation, aids in explaining the failure of adjacent sections to exhibit continuity in their detailed structures.