The work of Arthur J. Boucot (1924-2017) is foundational to Devonian biogeography. Cited over 1000 times and mentioned in excess of 105, 000 times, his classification of the Old World, Eastern Americas, and Malvinokaffric realms have become the framework for Devonian biogeographic and faunal studies. This study is the first to test and quantify the biotic area classification using the distributional data of trilobite, brachiopod, and fish taxa across the Lower, Middle, and Late Devonian. The biotic similarity analysis of 676 genera (ca. 30, 000 occurrences) of Late Silurian (Ludlow-Pridoli) and Devonian taxa represents the largest quantification of distributional data undertaken in Devonian biogeography. The Temporal Area Approach (TAAp), applied here for the first time, failed to recover in the area phenogram many of the realms and regions used by Devonian biogeographers, excepting the Mavinokaffric realm, across the entirety of the Devonian. The Old World and Western Gondwana realms are found to be artificial and in need of revision. Without natural areas, bioregionalisations (area classifications) become unstable and unusable, resulting in authors creating further artificial regions and provinces. This study offers a better solution, namely to identify artificial areas in existing area classifications using TAAp and biotic similarity analysis and revise them. Area classifications are modelled on taxonomic process and the movement toward revision is necessary to break the cycle of reinvention of realms and regions within bioregionalisation.