清代历朝《大清会典》和《清文献通考》等典章志书中的田土统计数据存在统计口径错误。乾隆以前历次官方统计数据中，仅康熙《大清会典》所载康熙二十四年数据较为准确，其中“田土总数”包括民田与当时已归并州县各卫所的屯田，“屯田数” 为当时尚未归并州县各卫所的屯田数。雍正、 乾隆两朝《大清会典》和《清文献通考》中雍正二年、乾隆十八年、乾隆三十一年的“田土总数”和“屯田数”各省统计口径不一，某些省份的两个数据均未计入已归并州县的屯田，另一些省份的两个数据则重复计算了屯田。利用各省省志中的数据修正统计口径问题，可知清代前期登记土地总数的增长速度比前人估计得更为平稳。

 There were errors with the land statistics in Qaug Huadaau and Qaug lheuxaau Tougkao. Among the official land statistics before the Qianlong period，only the statistics of 1685 were relatively accurate. The “total land data”of that year included the data on civil lands and military lands that had been transformed from the governance of guards and battalions to the governance of counties. The“military land data" of that year included the military lands which were still under the governance of guards and battalions. The contents of the statistics for 1724、1753、1766 were different from province to province. In some provinces both“total land data" and“military land data" did not include the military land transformed to the counties. In some other provinces，both data were included as military land. After emending the official land data by the statistics from provincial gazetteers，we can find a smoother land growth curve than the estimates of earlier scholars.