Straw input into soil is an agricultural management practice to increase the storage and stabilization of soil organic carbon (SOC). Soil aggregates are major sites for the sequestration of straw C. Scanty information is available on the C stabilization characteristics in aggregate fractions, and how they respond to straw input levels under different soil fertility levels. The objectives of this study were to determine the distribution of straw C in soil aggregate fractions and to investigate how the stabilization of straw C varied with soil fertility and straw input levels. We mixed 13C-labelled straw with Phaeozem and Luvisol soils (FAO classification system) at levels of 1 %, 3 %, 5 % and 10 %, respectively, and then put them into carborundum tubes for in-situ field incubation. Over a 360-day incubation, about 13 %–23 % of straw C accumulated in the macroaggregate (>250 μm), 5 % in the microaggregate (53−250 μm) and less than 2.0 % in the silt and clay fraction (<53 μm). The distribution of straw C in macroaggregate and microaggregate fractions increased with straw input levels. While the organic C content of microaggregate fraction averaged 9.2 g kg−1 and showed no significant difference (P > 0.05) among different soils at 10 % straw input. The organic C content of silt and clay fraction exhibited a logarithmic relationship with SOC content (y=0.69ln(x)-0.07, R2 = 0.58, P < 0.05). These results suggest that C stabilization of macroaggregate and microaggregate fractions positively responds to straw input levels, and that silt and clay fraction C approaches a steady state with high C input level.