Rothen and Mathot (1) and later Vroman and Adams (2) have shown that the antibody-antigen reaction will take place at a liquid-solid interface, and that the results of this interaction can be detected by the use of an ellipsometer. This optical instrument measures the adsorption and rotation of plane polarized light incident on the surface in question and has a limiting sensitivity of much less than a monolayer of adsorbed protein. I have used this technique to study the adsorption of protein from a solution onto various surfaces, such as Nb, Ta, Ni, Au, In, and Si, and further to observe the kinetics of the subsequent reaction with antibodies (Giaver, I., in preparation). These studies confirm earlier findings that the adsorption of a globular protein such as bovine serum albumin (BSA)1 and ovalbumin (OA) results in a monolayer about 30Å thick while the reaction with appropriate antibodies increases the thickness of the layer to approximately 100Å.