Application of organic manure increases methane emissions from rice paddy fields by increasing soil organic matter. Soil organic carbon is one of the key substrates that affect methane emission from paddy soils. Different methods of researching organic carbon have been used by different researchers which have led to different conclusions regarding correlations between organic carbon and methane emission from paddy soils. To define how methane emission is influenced by fractions of soil organic carbon, a paddy field experiment with application of different exogenous organic carbon (rice straw+chemical fertilizer, chicken manure+chemical fertilizer and pig manure+chemical fertilizer) was used to monitor and analyze methane emission, changes in organic carbon fractions and their correlation. A week after application of organic manure, rice straw +chemical fertilizer (RS) and chicken manure+chemical fertilizer (CM) showed emission peaks of 221.6 and 128.2 mg·m^-2·h^-1, respectively. The methane emission was mainly concentrated before the heading stage. The amount of methane emitted in the growing season following the chemical fertilizer (CF) treatment was 296.0 kg/hm² and following the pig manure+chemical fertilizer (PM), CM, and RS treatments were 340.7, 493.6 and 794.8 kg/hm², respectively. There was no significant difference in methane emissions between the PM and CF treatments, while emissions from the CM and RS treatments were 1.67 times (P < 0.05) and 2.69 times (P < 0.05) higher than from the CF treatment, respectively. Amounts of methane emission and the content of oxidizable organic carbon in the paddy soil followed the same order: RS > CM > PM > CF and no fertilizers. The content of organic C fractions followed the order: fraction 1 (organic C oxidizable by 33 mmol/L KMnO₄) > fraction 2 (the difference in C oxidizable by 167 mmol/L and that by 33 mmol/L KMnO₄) > fraction 3 (the difference in C oxidizable by 333 mmol/L and that by 167 mmol/L KMnO₄). Fraction 1 was the highest proportion of the easily oxidizable organic carbon at 42.7%-65%; followed by fraction 2 at 23.3%-48.9%; and fraction 3 was the smallest at 2.7%-17.1%. Fractions 1 and 2 in the tilling stage were higher than in the other stages in rice growth, and were a minimum in the mature rice. Fraction 1 in the rice growing season showed the same trend with methane emission as the RS and CM treatments, being obviously higher than the CF and no fertilizer treatments. Results of path analysis showed that, for fraction 1, the direct path coefficient and correlation coefficient were extremely significant at 1.0381 and 0.6709, respectively, whereas the direct path coefficients of fraction 2 and fraction 3 had negative values. Among all the organic carbon fractions, fraction 1was directly related to methane emission, while the other organic carbon fractions were only indirectly related to methane emission through fraction 1 during the later growth stages of rice, and the emission amounts were lower. Taken together, our results show that fraction 1 of the oxidizable organic carbon was the primary substrate of methane emission. Effective measures causing a decline in fraction 1 as a proportion of oxidation organic carbon in fertilizer resources and soil are possible key technologies for mitigating methane emissions from paddy fields.