The build up of the isotopic signal in corals was followed by sampling the newly formed skeleton at a monthly resolution for a period of two years in order to establish the interrelations between the calcification processes and the skeletal isotopic composition. We deployed two underwater sampling schemes, which provide a monitor of the changes in water temperature and δ18O and in the corresponding newly accreted skeleton of undisturbed Porites lutea colonies under natural conditions and four transplanted colonies, which maintained the genetic identity throughout the experiment. The results indicate that δ18O of the newly accreted skeleton does not correlate with ambient temperature although the seasonal temperature variability at the site (winter to summer) is in the order of 6°C and δ18O of seawater is constant throughout the year. In contrast to the newly formed surface skeleton, the isotopic compositions of the deep and older parts of the skeleton show the predicted annual isotopic pattern with highly significant correlation between δ18Os and SST. The transformation between temperature-independent to temperature-dependent isotopic signal occurs several months after the skeleton was formed at the surface. The position of the skeleton in relation to the open sea may generate the difference between δ18Os of the surface skeleton and that of the skeleton previously accreted further down the tissue layer. Our data support the general model of a multi-step skeletogenesis process, where the temperature independent skeleton is entails the first step, the production of skeletal scaffold, and the environmental temperature signature is captured by the next two other steps: the thickening and the periodic abrupt uplift occurring at the depth of the tissue layer. However, re-examination and development of the current isotopic models for coral calcification are required in order to explain the observed different temperature dependency during the growth's sequence.