We report the result of exposing silicone rubber to washed pig platelet suspensions that contained on average 0.018 mg of proteins/ml in solutions. This protein content is sufficiently low to reasonably neglect the protein adhesion to the material. On comparing the measured platelet adhesion from the platelet suspensions with that from blood, we find that when the gas nuclei normally present in the surface roughness of the material are removed the number of adhering platelets is the same. Thus, in the absence of the gas nuclei, the proteins in blood plasma play a negligible role in the platelet adhesion. In contrast, when both the gas nuclei and proteins are present, the maximum platelet adhesion was observed. From this and the above observation, it appears the gas nuclei affect one or more of the proteins, and this brings about an increased platelet adhesion. Finally, the platelet adhesion from the platelet suspensions was reduced after the removal of the gas nuclei. Thus the platelets themselves sense the change in the surface resulting from the removal of the gas nuclei.