In northeastern China, chilling damage is the main threat to rice production. The selection and breeding of cold tolerant rice varieties is an important goal. Research into the differing physiologies of rice strains grown under low temperature conditions can give guidance for the breeding of cold-tolerant varieties. A better understanding is needed of the physiological variation exhibited by rice during different growth stages, under exposure to low temperature stress. We selected fifteen major rice varieties that are grown in the central northeast of China for analysis under controlled environmental conditions. Following the climatic characteristics of the region and the temperature demands of rice at different developmental stages, we simulated natural temperature variation with an artificial weather box; set to provide four daily temperature gradients. A control variety was planted in the natural environment. We examined the relationship between the cold tolerance coefficient (the ratio of yields under low temperature compared with the control), and levels within the leaves of peroxidase (POD), malondialdehyde (MDA), proline (Pro), and chlorophyll, during the seedling, booting, heading and grain filling stages of development. This relationship has been shown to be related to the cold tolerance of the plants. Using the weighted average of subordinate function values (D value), of these four physiological indices, the relative cold tolerance shown by each variety was determined. We found that during each growth stage the POD, proline and chlorophyll levels decreased. This was in contrast to MDA levels; these were observed to increase. The subordinate function changes in the four indices are significantly correlated with the cold tolerance coefficient (P<0.01). At seedling, booting, heading and grain filling stages, we were able to use the above four physiological indicators to create an index of subordinate functions. From large to small these were Pro; MDA; POD; chlorophyll; MDA; chlorophyll; Pro; POD; chlorophyll; MDA; POD; Pro and MDA; chlorophyll; Pro; POD. Quantitative analysis of the cold tolerance characters of each variety was conducted using these subordinate functions; and the contribution of each index to the cold tolerance of each variety was analyzed. The D value of the four indices not only has a significant correlation with the cold tolerance coefficient but is also higher than the result of each index. In conclusion, the D value can be used to effectively evaluate and classify the low temperature tolerance of rice. It can provide an new method of screening for low temperature tolerance in rice breeding. The changes in the physiological indices observed, when plants were subjected to different periods of low temperature stress, accurately reflect the cold tolerance of rice, and can be used to effectively classify rice varieties according to their cold tolerance.