An investigation into some aspects of the surface chemistry of proteins at the hydroxyapatite/aqueous solution interface has been undertaken. A stoichiometric hydroxyapatite, fluoride and carbonate substituted hydroxyapatites and an amorphous calcium phosphate were prepared and characterised for use as adsorbents. Adsorption studies were performed using, mainly, systems of suspended adsorbent particles and radiochemical methods. Bovine and Human Serum Albumin, β-Casein, and a salivary glycoprotein, isolated from whole saliva, were radioiodinated to high specific activity but low degrees of labelling using modifications of the Chloramine-T or Iodo-Gen methods of iodination that minimised damage to the protein during labelling. The adsorption of Haemoglobin and Myoglobin was monitored by spectrophotometric methods. Care was taken to characterise the labelled and unlabelled preparations to ensure that the labelled protein behaved identically to unlabelled protein in adsorption experiments. Similarly, the necessary preliminary experiments were performed to adequately characterise the experimental system in use and so avoid the spurious results and interpretations that have characterised some earlier work. Special attention was paid to the adsorbent surface area under various experimental conditions, the effect of polymers and polymerisation behaviour of the protein on its adsorption and the effect of time, particularly at low protein concentrations. Care was taken to ensure that sufficient protein was available to achieve maximal adsorption. All the proteins studied showed a high affinity for the hydroxyapatite adsorbents i.e. the concentration adsorption isotherms were always parallel to the concentration axis at low concentrations and adsorption proceeded to saturation even at the lowest concentrations. Dilution studies showed that the concentration required to maintain saturation adsorption was negligibly small but desorption and exchange studies indicated that proteins can have different affinities for hydroxyapatite. Two types of adsorption behaviour were observed at higher concentrations. The majority of the proteins studies displayed an isotherm that remained parallel to the concentration axis with increasing concentration but β-Casein displayed a stepped isotherm which led to higher plateau level of adsorption. The incorporation of fluoride or carbonate in the hydroxyapatite lattice had no effect on the saturation levels of adsorption of any of the proteins at any concentration. The saturation amounts of adsorption of BSA corresponded to monolayer coverage of the adsorbent surface by molecules close packed in a side-on configuration with little, if any, structural alteration upon adsorption. The effect of the variables of pH, ionic strength and temperature were consistent with the known effects of these variables on the size of the BSA molecule. The saturation amounts of adsorption of β-Casein could similarly be accounted for in terms of monolayer coverage of the adsorbent surface. The low plateau level of adsorption corresponded to coverage by molecules lying prone on the surface in a close-packed configuration but the high plateau corresponded to molecules packed in a more vertical configuration. The stepping effect appeared to be related to the micellisation of this protein in solution. The effects of temperature and calcium ion concentration on adsorption were consistent with the effects of these variables on the solution micellisation. Although it is generally accepted that ionic forces are responsible for the binding of proteins by hydroxyapatites, the present work has indicated that non-ionic and entropic effects can also operate and contribute to the driving force for adsorption. In addition, protein carboxyl or phospho group exchange with species in the hydroxyapatite surface cannot be precluded. The effect of some multivalent metal ions on adsorption was studied. Mercury enhanced the adsorption of a second layer of protein with properties different from the layer initially adsorbed. Although calcium enhanced the levels of adsorption of BSA and β-Casein, the increased amounts adsorbed and the non-reversibility of the adsorption did not provide evidence of a second layer and were more consistent with denser packing in the original monolayer. Multilayer adsorption of β-Casein was promoted by the use of colloidal hydroxyapatite and amorphous calcium phosphate particulate species. Stable multilayers were formed, with consequent accretion formation, when the protein could bridge directly between the particulates i.e. at degrees of surface coverage less than saturation. BSA did not display such behaviour. The phenomenon appears to be related to the protein induced stabilisation/destabilisation of suspensions of colloidal species. The interfacial behaviour of the salivary glycoprotein at the hydroxyapatite/aqueous solution interface is similar to that observed for β-Casein. Many other salivary components also destabilised suspensions of colloidal calcium phosphate species and it is postulated that pellicle formation in the oral environment may be related to the phenomenon.