The

$n\times n$ fuzzy system of linear equations (FSLE) has been first studied by

*M. Friedman*,

*M. Ma* and

*A. Kandel* [Fuzzy Sets Syst. 96, No. 1, 201–209 (1998); comment and reply ibid. 140, 559–561 (2003;

Zbl 0929.15004)]. In the paper under review, the general

$m\times n$ FSLE is considered, and its weak solution is expressed by the generalized inverse of the coefficient matrix. Conditions for the existence of a fuzzy solution are derived and a numerical procedure for calculating the solution is designed. In case

$m>n$ the least-squares solution is found by applying an orthogonal matrix to the coefficient matrix.