In all vertebrates, estrogen action is mediated by cognate nuclear receptors. In this study, we cloned the different transcripts of the estrogen receptor beta (ERbeta) gene of the common sole, Solea solea. 5'-RLM-RACE (RNA Ligase-Mediated 5'-Rapid Amplification of cDNA Ends) and 3'-RACE analyses revealed three isoforms of different length, called Long, Intermediate and Short isoforms, consisting of 2212, 1531 and 1207 b, respectively. The Long isoform is characterised by an open reading frame (ORF) encoding 589aa, with an estimated molecular weight of 65kDa. Phylogenetic analysis established that it belongs to the teleost ERbeta1 or ERbetaa cluster. The Intermediate isoform encodes a 490-aa protein, which lacks the first 99aa of the Long isoform, but still retains a complete DNA-binding domain (DBD). In the Short variant (363aa-long), all the N-terminal region, down to the two zinc fingers included, is missing, thus crippling DBD. ERbeta transcription was analysed by semiquantitative RT-PCR with specific primers, common to the Long and Intermediate isoforms, in various sole tissues, such as brain, gills, muscle, stomach, intestine, spleen, head kidney, kidney, liver and gonads. This analysis revealed that ERbeta displays a widespread or ubiquitous pattern of transcription, with the highest levels being found in the gonads and liver.