The fifth-century manuscript known as the Codex Alexandrinus contains the entirety of the Greek Old and New Testaments and is a landmark in the history of the Bible. Though Alexandrinus represents a primary witness to the biblical text, no modern palaeographical or codicological analysis of the manuscript has been performed: the most in-depth studies of the codex pre-date the great papyrological finds of the 20th century. By executing both palaeographical and codicological analyses of the manuscript in light of a modern understanding of the textual history of the Bible and of Hellenistic Greek, and by additionally introducing statistical analysis into what has traditionally been a subjective field of research, this dissertation processes the textual and paratextual data of the manuscript to a degree previously unattained. The focus of the analysis is on the Gospels, which are quite unique in Alexandrinus: they stand at the headwaters of the Byzantine text form; they contain the earliest extant implementation of the Old Greek chaptering system; and the interaction between the unit delimitation and the Eusebian Apparatus in the Gospels is unique among the great uncial manuscripts. However, the analyses extend to both the Old and New Testaments to provide a context in which to study the Gospels. Among the discoveries made in this dissertation, this study overturns the view that a single scribe was responsible for copying the canonical New Testament books and demonstrates that the orthography of the Gospels can no longer be used to argue for the Egyptian provenance of the codex. As the first in-depth study of unit delimitation in the Gospels of Alexandrinus, this work reveals the complex relationship between the paragraphing system, the chaptering structure, and the incorporation of the Eusebian Apparatus from a separate exemplar. Finally, one result of the examination of the Eusebian Apparatus introduces the cascading error as a newly identified category of scribal error.