Abstract: Parameters and indexes derived from different models of photosynthesis-light response showed distant values,but these models were still the major methods to judge the feature of plant photosynthesis so far.In this paper,four light response models including the rectangle hyperbola,the non-rectangle hyperbola and two kinds of index curves were used to fit the light response data of evergreen poplar clone.Parameters and indexes from different models were compared.Methods of calculating light saturation point(LSP) were also studied in this paper.Results showed that the main characteristic parameters,including maximum net photosynthetic rate(P’_max),apparent quantum efficiency(a),dark respiration rate(R_d),which were obtained from the models of hyperbola,were higher than those from two index curve models.And parameters estimated from rectangle hyperbola were higher than those of the non-rectangle hyperbola.While the parameters derived from two kinds of index models appeared no difference.As for calculation methods of LSP,the LSP obtained from the straight line fitted by photosynthetic photon flux density under 200 μmo·lm^-2·s^-1 combined with other models was underestimated and this method seems unreliable.It is also unreasonable that LSP was estimated through the maximum gross photosynthesis rate(P_max).However,LSP could be effectively estimated through P’_max by eliminating the influence of R_d,and the estimation percentage of each models was basically fixed.The percentages for calculating LSP through P’_max in rectangle hyperbola,non-rectangle hyperbola and index curves were suggested to be(78±1)%,(82±1)% and(96±1)%,respectively.Beside,light response data of Eucalyptus grandis×E.urophylla were used to validate those conclusions.