Early evidence for the importance of legumes in nutrition is limited by difficulties in the identification and interpretation of pulse remains from archaeological contexts. The main aims of this work are to seek in a detailed survey of legume seed micromorphology, criteria that can be used i) for the identification of species and, ii) as evidence of cultivation. Assessment of the role of SEM in archaeobotany is a supplementary aim. Archaeological evidence of pulses in the Old World is first reviewed, followed by the rationale for the selection of members of the tribe Vicieae as the species researched. The second part of the thesis is a background study of the Vicieae and its four major genera Vicia, Lathyrus, Lens and Pisum. Brief accounts of their taxonomy, seed morphology and geographical distribution are followed by an examination of traditional Old World systems of pulse agronomy. Chapter three concerns the material and methods of research. The results are described qualitatively in chapter four, and following multivariate analysis in chapters five and six. The seventh chapter describes some miscellaneous procedures. The eighth chapter is a discussion of the results. Identifications down to species level using seed micromorphology are restricted to certain taxa, contrary to some published reports; but allocation of specimens to a higher taxonomic rank, using an array of criteria, is usually more realistic. Most useful characters are found in features in the hilar region. Clear evidence of cultivation has only been observed in the genus Pisum. Using SEM, similar data may be recorded equally in fresh seeds and fragments of charred archaeological specimens. It is concluded that seeds of the Vicieae usually do not exhibit sufficient morphological differences for species identification. Evidence for cultivation may rather be sought in biochemical changes that facilitate control of the germinationin seeds of cultigens. Future research into pulse biochemistry and agronomy is anticipated.