The goal of our research program is to develop competitive and sandwich fluoroimmunoassays with high sensitivity and fast response time, that do not require external reagents. Our approach to this problem is to employ an optical immunoassay based on total internal reflection fluorescence (TIRF). Specifically, monoclonal antibodies are immobilized on a planar waveguide. Total internal reflection of light in the planar waveguide sets up an evanescent field which extends about 2000 angstroms from the interface. In the competitive immunoassay, a fluorescent label is coupled to a small synthetic antigen which is packaged with the antibody. In the absence of analyte, the fluorescently labeled antigen binds to the antibody and is excited by the evanescent field. Upon the addition of analyte, the fluorescently labeled antigen molecules are displaced by unlabeled antigen molecules and diffuse out of the evanescent field. In the sandwich assay, a primary or `capture' antibody is immobilized on the planar waveguide, and a secondary or `tracer' antibody (which is labeled with a fluorescent dye) is added to the bulk solution. In the absence of analyte, the tracer antibody remains in solution and very little fluorescence is observed. However, upon addition of analyte, a `molecular sandwich' is formed on the waveguide, composed of: (1) the capture antibody; (2) the analyte; and (3) the tracer antibody. Once this sandwich forms, the tracer antibody is within the evanescent field and fluoresces. Fluorescence emission is detected by a charged- coupled device (CCD). Using this approach, we have developed a prototype immunosensor for the detection of human chorionic gonadotropin (hCG). This device meets our design goals and exhibits a sensitivity of 0.1 - 1 pmolar.