The movement of arms is having a profound effect on the 50 or so ongoing ethno-political conflicts in the 1990s. A rigorous analysis of their acquisition is lacking. The authors look at the dimensions of the problem and the pattern of sources. They offer five hypotheses about the types of weapons traded. These hypotheses, backed by tables, are examined against the acquisitions by ethno-political groups in 49 conflicts. The findings generally support the evidence that most ethno-political groups use light arms and that importation from neighbouring states is the most frequent form of arms acquisition. The authors concede that data-collection is so far insufficient. Nevertheless, it is useful in pointing to the large problem this trade poses for arms control efforts. Light arms are difficult to track and easy to conceal and cause huge numbers of casualties. They pass easily from one conflict to another. Embargoes are hard to enforce, but regional control efforts should be tried. The conclusions underline the need for further study and the collection of information.