This study assesses the distribution and level of genetic diversity of accessible date palm resources, provides new insights on the geographic origins and genetic history of the cultivated component of this species, and confirms the existence of at least two domestication origins. Furthermore, the strong genetic structure clearly established here is a prerequisite for any breeding programme exploiting the effective polymorphism related to each gene pool.

Since the publication of this paper, it has become apparent that an error was made in the legend to Fig. 3 and the colours referring to occidental and oriental are the wrong way round. The authors apologise for this error, and a correct version of the legend to Fig. 3 is given below.

The genus Phoenix (Arecaceae) comprises 14 species distributed from Cape Verde Islands to SE Asia. It includes the economically important species Phoenix dactylifera. The paucity of differential morphological and anatomical useful characters, and interspecific hybridization, make identification of Phoenix species difficult. In this context, the development of reliable DNA markers for species and hybrid identification would be of great utility. Previous studies identified a 12 bp polymorphic chloroplast minisatellite in the trnG (GCC)-trnfM (CAU) spacer, and showed its potential for species identification in Phoenix. In this work, in order to develop an efficient DNA barcode marker for Phoenix, a longer cpDNA region (700 bp) comprising the mentioned minisatellite, and located between the psbZ and trnfM (CAU) genes, was sequenced. One hundred and thirty-six individuals, representing all Phoenix species except P. andamanensis,were analysed. The minisatellite showed 2-7 repetitions of the 12 bp motif, with 1-3 out of seven haplotypes per species. Phoenix reclinata and P. canariensis had species-specific haplotypes. Additional polymorphisms were found in the flanking regions of the minisatellite, including substitutions, indels and homopolymers. All this information allowed us to identify unambiguously eight out of the 13 species, and overall 80% of the individuals sampled. Phoenix rupicola and P. theophrasti had the same haplotype, and so had P. atlantica, P. dactylifera, and P. sylvestris (the “date palm complex” sensu Pintaud et al. 2013). For these species, additional molecular markers will be required for their unambiguous identification. The psbZ-trnfM (CAU) region therefore could be considered as a good basis for the establishment of a DNA barcoding system in Phoenix, and is potentially useful for the identification of the female parent in Phoenix hybrids.