Abstract
The conventions at present used in the classification of blue-green algae frequently prove unsatisfactory. A solution is suggested which requires the simultaneous use of two different approaches. When a binomial is essential the flora of Geitler (1932) should be adequate for most purposes, but any long term attempt to sort out the present chaos will require the use of numerical taxonomy. This dual approach will necessitate various practical changes in current methods of naming and reporting on blue-green algae.