This paper presents data to support the presence of (1) intra-annual signals in the chemical composition (δ18O and Sr/Ca) of the skeletons of sclerosponges from the Bahamas and (2) variable rates of skeletal accretion. These conclusions are based on data obtained by using a microsampling method for the stable oxygen and carbon isotopes in which material was extracted at a resolution of one sample every 34 ¿m and a laser microprobe which obtained trace element data every 20 ¿m (Sr, Mg, and Pb). An age model was established using a combination of changes in the concentration of Pb, the change in the δ13C of the skeleton of the sclerosponges, and U/Th isotopic measurements. These methods yield a mean growth rate of 220 ¿m/yr but suggest that the growth rate in this particular sclerosponge was not constant. The calculated growth rate is within error identical to that determined by U/Th methods. The variable growth rate was confirmed through spectral analysis of the δ18O and Sr/Ca data that showed peaks corresponding to the annual cycle in these parameters as well as peaks corresponding to growth rates of approximately 128, 212, 270, and 400 ¿m/yr. The presence of these additional frequencies suggests a growth rate between approximately 100 and 300 ¿m/yr. These conclusions were supported by modeling of oxygen isotopic data measured on a scleractinian coral as well as model isotope data generated on synthetic time series. These findings have important implications for the use of sclerosponges as proxies of paleoclimate because they emphasize the need for a precise yearly chronology in order that proxy data can be compared with climatic variables.