In the current literature, Saharan green-colored stone beads are identified with the term émeraude garamantique, while several archaeological investigations on Garamantian contexts referred to them as amazonite. Moreover, according to the hypothesis proposed by Monod, authors identify the only source of amazonite to be the pegmatite swarm at Eghei Zuma (northern Tibesti, Libya). To check this assumption, we characterized the composition of green-colored stone beads found at Fewet, a Garamantian site (2nd century BC – 1st century AD) in the Libyan Sahara. We performed electron microprobe analyses on stone beads from Fewet and ethnographic contexts of Mali, Mauritania, and Sudan. Furthermore, lacking a regional database on amazonite to compare the geochemical properties of archaeological beads, we analyzed amazonite crystals from several African pegmatites, including samples from Eghei Zuma. The results show that green-colored stone beads from Fewet consist of serpentinite and amazonite. The K/Rb versus Rb diagram reveals that the source of the amazonite from Fewet could not be the pegmatite at Eghei Zuma. The raw material for amazonite beads traded in Garamantian times originated from Egypt and Mauritania. We discuss the archaeological implications of our results, which concerns the directions of ancient trade routes for elite goods in Garamantian times.