A poorly understood ocean underlies the seafloor, trapped in fractures and pore spaces of detrital grains and volcanic structures and bound both loosely and tightly by clay and other hydrous minerals. Movement of this ocean is especially significant in active tectonic environments, and huge volumes of fluid move rapidly through subduction systems. A significant part of postcruise studies from Leg 170 focused on this subsurface fluid flow, including structural fabrics and fluid-flow indicators at the interface of subduction. Shipboard results (Kimura, Silver, Blum, et al., 1997) demonstrated that the Costa Rica margin does not undergo frontal accretion and that the bulk of the small, deformed sedimentary wedge at the toe of the margin was not formed by scraping off the incoming sediment from the Cocos plate. Shipboard results also supported the intense fall in heat flow on the Cocos plate (Langseth and Silver, 1996), with major implications for fluid flow. Here I present a synthesis of postcruise science that focused on the structural, chemical, and physical properties relating to fluids at the subduction front and discuss the implications of these results.