On the basis of a comparative study of 178 strains of cyanobacteria, representative of this group of prokaryotes, revised definitions of many genera are proposed. Revisions are designed to permit the generic identification of cultures, often difficult through use of the field-based system of phycological classification. The differential characters proposed are both constant and readily determinable in cultured material. The 22 genera recognized are placed in five sections, each distinguished by a particular pattern of structure and development. Generic descriptions are accompanied by strain histories, brief accounts of strain properties, and illustrations; one or more reference strains are proposed for each genus. The collection on which this analysis was based has been deposited in the American Type Culture Collection, where strains will be listed under the generic designations proposed here.
I N T R O D U C T I O NThe cyanobacteria constitute one of the largest sub-groups of Gram-negative prokaryotes. As a result of their traditional assignment to the algae, the classification of these organisms was developed by phycologists, working under the provisions of the Botanical Code (Stafleu et al., 1972). Almost entirely on the basis of observations on field materials, about 150 genera and well over lo00 species have been described. The discriminatory properties, both generic and specific, are either structural or ecological, these being virtually the only characters determinable in the field. Types are represented by herbarium specimens or, failing these, by descriptions and illustrations; cultures are not recognized as valid type materials under the Botanical Code.The attempt to identify cyanobacteria in culture through this field-based system of classification leads to many difficulties and ambiguities. The limited and necessarily provisional taxonomic goal of the present article is to redefine certain cyanobacterial genera in such a way that simple and clear-cut generic assignments can be made for cultures. It is based on our experience over the past decade with pure strains representative of all major sub-groups of cyanobacteria. As far as possible, we have attempted to maintain the system of generic nomenclature and the generic definitions now used by phycologists (Bourrelly, 1970;Geitler, 1932;Desikachary, 1959). However, when the discriminatory characters that nominally distinguish two genera are either not determinable on cultures or within the range of variation of a single strain, the existing genera have been combined. Some of the proposed generic definitions include discriminatory characters that have not hitherto received taxo-