Archaeologists working in the northern Maya lowlands have faced persistent problems in establishing chronological precision and accuracy. In particular, it has proven difficult to create multi-phase chronologies for the Late and Terminal Classic periods. Investigators at Xkipche, a small Puuc site southwest of Uxmal, have employed both seriation and the typological approach to ceramic chronology. The results of the ceramic seriation suggest great persistence from the second century until a.d. 1100, a continuity that is not supported by the Type-Variety approach to chronology. This report begins by reviewing the ceramic data, and then turns to another archaeological material, obsidian. Procurement patterns and production technology are discussed for the 182 obsidian artifacts collected during the first five seasons of the Projekt Xkipche. These data are compared with similar information gleaned from other sites in the northern Maya lowlands. Results of this obsidian analysis conflict with the absolute chronology proposed for the Xkipche ceramic sequence. Finally, a compromise ceramic chronology is proposed, one that is consistent with both obsidian and architectural data. This proposed chronology divides the seemingly monolithic Cehpech ceramic complex into three phases: Early Cehpech (a.d. 550–700); Late Cehpech (a.d. 700–900/950); and, Terminal Cehpech/Sotuta (a.d. 900/950–950/1000).