Epithelial barrier function studies often attribute alterations in barrier function to induced changes in tight junctional (TJ) complexes. The occurrence of spontaneous and cytokine-induced, focal cell detachment in cell layers of the human gingival epithelial cell line, Gie-3B11, highlights the danger of this assumption without confirmatory experimentation. Gie-3B11 cell layers manifest morphological polarity, TJ complexes and barrier function after confluence but fail to then maintain a stable epithelial barrier. Transepithelial electrical resistance rises to over 100 ohms x cm² a few days after seeding cell layers at a confluent density, but then spontaneously declines, with simultaneous, inverse changes in transepithelial ¹⁴C-D-mannitol diffusion rates. This barrier decline correlates with the appearance of focal cell detachment/hole formation in cell layers. Both barrier compromise (decreased electrical resistance; increased ¹⁴C-D-mannitol leak) and hole formation are accelerated and exaggerated by exposing cell layers to proinflammatory cytokines. Both are inhibited by increasing the basal-lateral medium compartment volume, suggesting that cell layers are secreting factor(s) across their basal-lateral surfaces that are causal to hole formation. The molecular mechanism of cell death/detachment here is not as significant as the implications of hole formation for the correct interpretation of barrier function studies. Barrier changes in any epithelial model should be attributed to induced changes in TJ complexes only after thorough investigation.