The reliability for determination of cobalt in biological samples was studied by using various analytical methods such as atomic absorption spectrometry (AAS), absorption photometry(AP), and instrumental neutron activation analysis (INAA). The marine samples used were various kinds of shellfish which were dried at 105°C and then ashed at 500°C in an electric furnace. After the ashes were decomposed with HF, HNO3 and fuming HNO3, followed by evaporation to dryness, the residues were dissolved in dilute HCl and were analysed by AAS and AP. In the AAS method the absorbance was directly measured at 240.7 nm using air-C2H2 flame. In the AP method cobalt was first extracted into xylene with 1-nitroso-2-naphtol reagent, and then the absorbance was measured at 410 nm. In the INAA method, the ash sample and the reference standard were irradiated for 24 h at a thermal neutron flux of 5×1011 n cm-2 s-1 in the TRIGA II reactor at Rikkyo University, and after cooling for 15 days the 1332 keV gamma ray was measured by a Ge(Li) detector coupled with a multichannel pulse height analyser. The analytical results were (in μg Co/g wet tissue) : 0.49 (AAS). 0.22 (AP) and 0.21 (INAA) in Tapes philippinarum; 0.28 (AAS), 0.081 (AP) and 0.077 (INAA) in Mytilus edulis; 1.2 (AAS), 1.1 (AP), and 0.95 (INAA) in Tridacna squamosa, respectively. Values obtained by AP and INAA agreed well each other, whereas those obtained by AAS were higher than the corresponding values obtained by the former two methods. The origine of the higher values was found to be the interference by iron and calcium in the samples. The reliability of AP and INAA were confirmed by the analysis of Bovine Liver (NBS standard reference material) and standard rock samples AGV-1 and JG-1.