The stalks of peritrich protozoa have aroused great interest for many years and for many reasons. Some are contractile, some are not. This attribute of contractility, first observed by Leeuwenhoek in Vorticella and published in this journal in 1676, is one good reason for further studies of a structural, cytochemical and physiological character. This paper is mainly concerned with matters of fine structure which relate not only to the mechanism of contraction in those stalks that behave in this way, but also to wider problems of morphogenesis in ciliates. The early literature of this subject is clouded with optical artifact and one of the first problems to be solved is the precise difference between non-contractile and contractile stalks. Seven families are now included in the suborder Sessilina of the Peritrichida and members of the Epistylididae and the Vorticellidae have been selected for this investigation. This choice has made possible a detailed comparative study of fine structure in the non-contractile stalks of the first group and the contractile ones of the second. All stalks possess longitudinally arranged structures. In the non-contractile stalks these structures are tubular in form and may be observed in the phase-contrast microscope. In the contractile stalks the longitudinal structures are of two main kinds, one of which is confined to the annulus and the other to an inner canal separated from the annulus by a membrane. The annular structures are tubular and numerous in Carchesium and Zoothamnium and transversely striated, while in Vorticella they are composed of unstriated fibres, few in number. The structure within the canal is the main feature that distinguishes the stalks of Vorticellidae from those of Epistylididae. It consists of a long bundle of closely packed fine fibrils and is to be identified with the stalk spasmoneme or myoneme of the older literature. Only one type of fibril has been observed in the spasmonemes and present facts are not consistent with the idea that they contract in the same way as muscles. The spasmoneme is protein in nature with positive indications of the presence of -NH<latex>$_2$</latex>, S-H, and S-S groups. The annular structure in the Vorticellidae and the tubular structures of Epistylididae have cytochemical affinities with the keratin group of proteins. Structurally, they grow out as the stalk develops from an assembly of organelles known collectively as the scopula. In the contractilia the spasmoneme passes into the zooid through a more or less central gap in the scopula and terminates in the form of a circular fan of fibrils on or close to the zooid pellicle. The fine structure of the stalk tubules of the Epistylididae has been investigated in some detail, particularly for one species of Epistylis. In Epistylis and Opercularia the tubules are transversely striated in a manner similar to that described for Carchesium and Zoothamnium. In Campanella each tubule consists of a loose helix of fibrils interlocked with those of neighbouring tubules. The form of attachment of the tubules of Epistylis and Opercularia to the scopula organelles has been determined. Wherever the preparations were of sufficient quality a comparison has been made of scopula organelles and the corresponding and possibly homologous structures of normal cilia known as basal bodies or kinetosomes. There are similarities and, of course, differences. It seems justifiable to regard the scopula organelles as basal bodies modified in the course of the evolution of this Order for the purpose of contributing a degree of structural stability and rigidity to the stalks. No such obvious `origin' for the spasmoneme has been found in the adult organisms. This illustrates the danger and perhaps sterility of attempts to link the genesis of one structure to another on purely morphological grounds. The role of the scopula organelles and in a wider context kinetosomes in the organization and possibly the synthesis of fibrous proteins is discussed.