An optical technique for measuring the thickness of thin films has been adapted and evaluated for studying the structure of the adhesion of cells to glass in tissue culture. This technique, which is termed interference reflection microscopy, has been used to study embryonic chick heart fibroblasts. These findings have been observed: in normal culture medium the closest approach of the cell surface to substrate in its adhesions is ca. 100 A, much of the cell surface lying farther away; chemical treatments which bring the cell surface to near its charge reversal point reduce the closest approach of adhesions to <50 A, probably to <30 A; chemical treatments which increase surface charge increase the nearest approach of cell and substrate in adhesions from ca. 100 A; high osmotic concentration of a non-polar substance, i.e. sucrose, does not affect the distance between cell and substrate in the adhesions. In addition, optical evidence indicates that there is no extracellular material between cell and glass in the adhesions. When cells de-adhere from glass, they appear not to leave fragments behind. The adhesive sites in these fibroblasts appear to be confined to the edge of the side of the cell facing the substrate and to the pseudopods. The significance of this is discussed in relation to the phenomenon of contact inhibition. Evidence is presented that the mechanism of cell adhesion does not involve calcium atoms binding cells to substrate by combining with carboxyl groups on cell surface, substrate, and with a cement substance. Osmium tetroxide fixation results in a final separation of 100 to 200 A between cell and substrate: there are reasons for thinking that this fairly close approach to the condition in life is produced as an artefact. The results can be accounted for only in terms of the action of electrostatic repulsive forces and an attractive force, probably the van der Waals-London forces. Biological arguments suggest that these results are equally applicable for cell-to-cell adhesions.