The need for controlled drug delivery has driven the development of a range of novel materials with tuned degradation, deformation, or diffusion profiles. The key functionality of these materials is their ability to release the desired chemical, often a drug, at an appointed time. Examples include hydrogels, [1][2][3] poly(glycolic acid)/poly(lactic acid) (PGA/PLA), [4] microfluidics, [5] and reversible-hydrophilicity surfaces.[6] These materials systems interact with the drug by electrostatic interactions, covalent grafting with a cleavable linker, physical entrapment, or large diffusion barriers. The difficulty with these systems is that they are large relative to a cell (order of 1-10 mm), and rely upon ensemble averaging to provide a uniform response, which may not be applicable at the cellular scale. With the emerging importance of chemical stimulation of single cells within 2D cultures for systems in biology, regenerative medicine, and neural networks, significant research is on-going to create artificial systems that can mimic these release profiles at the micrometer to submicrometer scale. Microfabricated inorganic systems, like micropipette perfusion [7][8][9] and chip-based chemical delivery, [10][11][12] can have subcellular spatial resolution, but are difficult to scale to more than one cell or into 2D arrays. Typically, micropipette perfusion has no true seal, and thus continually releases some signal by diffusion, even in the ''off'' state. Microfluidic gating technologies such as soft channel deformation, [5] mechanical check valves, [13] and bulk fluid injection/withdrawal [10,14] have a relatively large device footprint, and are difficult to design into 2D arrays rather than in an array of 1D channels. Microfluidic systems also rely on additional flow channels to supply and modulate fluid flow, further complicating device design and fabrication. Here we demonstrate a simple architecture for creating arrays of small reservoirs that contain chemical signals. The challenge for these and other pore-like delivery systems (such as pore-track membranes or anodic alumina) is how to produce a gating mechanism that is self-contained within the sub-micrometer pores without large control structures. Lipid bilayers present one potential candidate for such a gating device.