The parameters estimated from traditional A/Ci curve analysis are dependent upon some underlying assumptions that substomatal CO2 concentration (Ci) equals the chloroplast CO2 concentration (Cc) and the Ci value at which the A/Ci curve switches between Rubisco- and electron transport-limited portions of the curve (Ci-t) is set to a constant. However, the assumptions reduced the accuracy of parameter estimation significantly without taking the influence of Ci-t value and mesophyll conductance (gm) on parameters into account. Based on the analysis of Larix gmelinii's A/Ci curves, it showed the Ci-t value varied significantly, ranging from 24 Pa to 72 Pa and averaging 38 Pa. t-test demonstrated there were significant differences in parameters respectively estimated from A/Ci and A/Cc curve analysis (p<0.01). Compared with the maximum ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) carboxylation rate (Vcmax), the maximum electron transport rate (Jmax) and Jmax/Vcmax estimated from A/Cc curve analysis which considers the effects of gm limit and simultaneously fits parameters with the whole A/Cc curve, mean Vcmax estimated from A/Ci curve analysis (Vcmax-Ci) was underestimated by 37.49%; mean Jmax estimated from A/Ci curve analysis (Jmax-Ci) was overestimated by 17.8% and (Jmax-Ci)/(Vcmax-Ci) was overestimated by 24.2%. However, there was a significant linear relationship between Vcmax estimated from A/Ci curve analysis and Vcmax estimated from A/Cc curve analysis, so was it Jmax (p<0.05). and W. Zeng ... [et al.].