This thesis proceeds from the contention that the relationship between text and archaeology, which comprises protohistoric study, is poorly understood. The archaeology of Later Iron Age religious ritual is employed as a forum for examination of this relationship. Whilst text is often privileged in protohistoric study, the Later Iron Age textual data themselves have not been adequately examined. The first aim of this thesis is therefore to evaluate comments on the rites and beliefs of peoples described as Keltoi or Galli in Classical texts of Later Iron Age date. This evaluation forms the basis for exploitation of the potential of the texts and for analysis of existing archaeological approaches to this material. The second part of this work examines three site categories from the current Later Iron Age ritual corpus. These are: water sources, wells and shafts, and rectilinear enclosures. The archaeological criteria on which cult identities are advanced in each case are examined, as is the nature and extent of the use of textual data in informing the ritual identities afforded these loci. It is concluded that ritual identities are not assigned primarily on the basis of Later Iron Age material evidence, but are heavily predicated on text-led presuppositions of the nature of 'Celtic' religion, and on retrospective appeals employing the post-Conquest archaeological record. The validity of ritual identities assigned on such bases is questioned, and it is argued that reliance on these processes has meant that the underlying dynamic of Later Iron Age religion has been little addressed. It is suggested that the contextual integration of textual and archaeologial data, and greater chronological control over both data sets, are prerequisites for methodological progress in the study of Later Iron Age religious ritual.