The hypothesis that intercellular adhesion can be subdivided into two separable phenomena - an initial recognition event and a subsequent stabilization - is supported by the use of a cell binding assay that provides a quantitative measure of inter-cellular binding strengths. Radioactive single cells are brought into contact with cell monolayers at 4° C in sealed compartments. The compartments are inverted and a centrifugal force is then applied to dislodge the probe cells from the monolayers. By varying the speed of centrifugation, the force maintaining associations between embryonic chicken neural retina cells was determined to be on the order of 10-5 dyne. Topographic specificities of single neural retina cells for retinal monolayers from various regions of the retina were detected with this assay and corresponded to those observed in more traditional assays at 37°C. Also observed were two time- and temperature-dependent stabilization processes in which the force required for dislodgment increased. One of the stabilization processes was sensitive to dinitrophenol and was inactive at 4° C; the second was still active in metabolically blocked cells. The metabolic-dependent process resulted in interactions at least 13 times as strong as the initial binding. The metabolic-independent process resulted in about a 2-fold increase in binding strength and had a temperature dependence similar to that of membrane diffusional phenomena.