Our claim, published in this journal, for successful immunodetection of the protein osteocalcin in dinosaur bone has been challenged on the grounds that the findings are inconsistent with the kinetics of decomposition. Here we show that the close association of osteocalcin to the bone mineral vastly enhances its preservation potential relative to the same protein in aqueous solution. We conducted heating experiments (75-95 °C) of modern bone powder and monitored the survival of three different regions of osteocalcin (N-terminal, His₄-Hyp₉; C-terminal, Phe₄₅-Val₄₉; and the mid-region, Pro₁₅-Glu₃₁) with monoclonal antibodies. Extrapolation of our data to 10 °C ambient burial temperatures indicates that preservation of the γ-carboxylated mid-region in fossil bone cannot be excluded on kinetic grounds. Clearly, in situ sequence analysis will be the only method by which the preservation of fossil macromolecules will be unequivocally established. Nevertheless, our findings demonstrate the importance of mineral association to protein survival, as was borne out by an investigation of Holocene (ca. 6 ka) bones. Only in those samples with little recrystallization was the γ-carboxylated mid-region well preserved. These results imply that the future success of ancient biomolecule research largely depends on our understanding the interaction between these materials and their environment throughout diagenesis.