Karst vegetation plays an important role in the karst ecosystem carbon cycle and the global carbon balance. However, most of the estimations on biomass and carbon sequestration patterns for many karst forest vegetation ecosystems were questionable. The lack of accurate characterization of carbon storage in tree, understory, forest floor and organic soil layers in karst forest vegetation ecosystems has hindered the effort to estimate their contribution to the global carbon stocks. In this study, we investigated changes in biomass and carbon storage of karst vegetation along successional stages of a chronosequence following human disturbance in Mashan county, southwest China. We sampled nine plots, each of 20 m×50 m, three of which represented successional stages of young, mid-mature and old forest, respectively. We found that community biomass increased significantly along the successional stages, from 48.17 t/hm² in young forest to 113.47 t/hm² in mid-mature forest to 242.59 t/hm² in old forest (P <0.05). The biomass of the understory vegetation in different successional stages is very low, accounting for 4.32%, 1.47% and 0.47% of the total biomass in communities of young, mid-mature, and old forsts, respectively. The mean biomass carbon storage in old forests (123.26 t/hm²) was significantly higher than that in young forest (26.32 t/hm²) and than that in mid-mature forest (57.78 t/hm²) (P <0.05), and it was significantly higher in mid-mature forests than that in young forests (P <0.05). Carbon storages in tree and herb layers were significantly different among the three forest types (P <0.05). The differences in carbon storage in shrub layer, however, was not significantly different among the three forest types (P > 0.05). Carbon storage in forest floor litter in mid-mature forest (0.20 t/hm²) was significantly lower than that in young forest (1.19 t/hm²) and in old forest (1.40 t/hm²). Soil carbon storage was significantly different between old forest (113.43 t/hm²) and young forest (198.44 t/hm²), but the difference in soil carbon was not significant between old forests and mid-mature forests (167.39 t/hm²). In all three forest types, carbon storage in soil decreased as depth increased. Overall, carbon storage in old forest ecosystems (236.69 t/hm²) was only slightly higher than that in young forest (224.76 t/hm²) and mid-mature forest (225.17 t/hm²), but these differences were not statistically significant (P >0.05). In contrast, soil carbon storage was highest in the young forest (198.44 t/hm²), followed by mid-mature forest (167.39 t/hm²) and old forest (113.43 t/hm²). The ratios of plant vs. soil carbon storage in various secondary forest ecosystems increased significantly along the successional stages. The percentages of plant and soil carbon storage were 11.71% and 88.29%, respectively, in young forest ecosystems, 25.66% and 74.34% in mid-mature forest, and 47.92% and 52.08% in old forest. It appeared that soil carbon was converted to plant carbon as forest moving forward along the successional chronosequence in the karst region. Our results suggested that the carbon storage capacity of the karst forest ecosystems decreased due to the high bare rock cover, limited soil cover, shallow soil layer, and low biomass. Our results also indicated that the carbon sequestration and carbon allocation in the karst ecosystems were influenced strongly by karst forest vegetation succession. More long-term monitoring and research are needed to further evaluate biomass and carbon accumulation of the karst forest vegetation succession.