The use of Next Generation Sequencing (NGS) techniques to identify microsatellite markers has replaced more time intensive methods such as molecular cloning. The main advantage of NGS over traditional methods of identifying microsatellite markers is the generation of many more sequences with less effort. It is possible to design primers from unenriched DNA, thereby further reducing the workload and also allowing the use of SSRs that are difficult to enrich (e.g., TA/AT and TAA/ATT). We present microsatellite primer pairs that may be used for phylogeographic analysis as well as to infer the geographical origin of traded material of Catha edulis, which contains two amphetamines that are controlled substances in many counties. We used data from two partial 454 pyrosequencing runs that generated about 2000 sequences containing microsatellites (3% of all sequences) as well as flanking regions sufficient for primer design. Using 23 samples of C. edulis we identified 27 single-copy markers that were broadly amplified across the sampled individuals; 18 showed polymorphism information content (PIC) higher than 0.5. The genetic structure in wild individuals is concordant with their geographic origins; wild samples from northern Kenya are more closely related to Ethiopian samples than are other wild samples from Kenya. The geographic differences in allele frequencies indicate that microsatellite analysis can be used to determine the geographic source of cultivated and wild collected material. © 2013 Elsevier Ltd.