RATIONALE While several studies have described the genetic diversity of C.arabica, none has included the whole range of Ethiopian accessions together with a large representation of the varieties cultivated worldwide. None included recently surveyed South Sudanese populations. Over the last years, WCR has built a large database that actually includes all the dimensions of genetic diversity of the species: from Ethiopian, Yemeni and South Sudanese accessions to a large representation of old and new cultivated varieties. The results of the detailed analysis of this database (SSR) are presented here. METHODS The data base represents i) One core collection established in 2014 including mainly Ethiopian accessions (FAO and ORSTOM surveys), ii) Populations of wild arabicas surveyed in South Sudan in 2014 and iii) a large representation of cultivated varieties worldwide. More than 2000 entries of this database were genotyped using a set of9 SSR markers. Multivariate analysis (PCoA) were run in order to decipher the underlying genetic diversity. RESULTS For the first time, wild C. arabica populations of South Sudan are shown to bring new genetic diversity as compared to Ethiopian wild arabicas. A structuration of the Ethiopian accessions surveyed in the 60's (FAO and Orstom) is unraveled. The traditional Bourbon/Typica varieties are genetically related to the Ethiopian cluster east of the Ethiopian coffee area. This study gives a new light on the history of C. arabica movement around the world. While the genetic diversity of cultivated varieties around the world is confirmed to be relatively low, it is still possible to authenticate them through fingerprinting. Most varieties show a residual segregation and not fully fixed homozygous lines. Consequences for varieties authentication are discussed. CONCLUSIONS & PERSPECTIVES This study is the first of its kind with SSR on a wide range of Arabica accessions and varieties. It gives us a new vision of the genetic diversity of the