In the literature it has been suggested that the protective layer, deposited along the wall between xylem parenchyma and vessels, is involved in tylose formation as part of an antipathogenic response. Yet, in a number of cases, the presence of a protective layer is not related with tylose development. It is proposed here, that the protective layer primarily acts as a buffer against hydrostatic oscillations in the vessels. As the hydrostatic pressure in the vessels becomes less negative, the xylem cells will increasingly withdraw water from the apoplast. Once the hydrostatic pressure has surpassed a certain limit, the protective layer is unable to withstand the osmotic pressure of the parenchyma cells and the latter will bulge into the vessels giving rise to the formation of tyloses.