<title>ABSTRACT</title>We conducted a preliminary test of the coupled system of an elemental analyzer and the automated graphitization equipment Ionplus AGE3 (EA-AGE3 method) for accelerator mass spectrometry radiocarbon (AMS ¹⁴C) measurements of CaCO₃ samples, by comparing with the conventional method where the samples are hydrolyzed in phosphoric acid and resulting CO₂ gas is manually graphitized in a vacuum line (HPA method). The samples used in the test were the IAEA C2 travertine, fossil and modern corals from the Ryukyu Islands and the Ogasawara Islands, respectively (both are located in the northwestern subtropical Pacific). Results indicate that, relative to the HPA method, the EA-AGE3 method tends to cause an increase of ~0.4–0.5 pMC with more widely scattered data. This is presumably due to ¹⁴C contamination in the EA (the most likely cause seems to be a memory effect of ¹⁴C); this effect could be reduced by careful optimization of conditions and procedures in the EA process. The ¹⁴C data of pre-bomb annual bands (1931–1949 AD) in the modern Ogasawara coral obtained by the HPA method were used to estimate the marine reservoir ¹⁴C-age correction (ΔR) of this region; it ranges from –109 yr to –28 yr with the mean value with standard deviation of –81 ± 29 yr.