A ranked set sampling procedure with unequal samples and unequal replications () RSSUR is proposed and used to estimate the population mean. This estimator is then compared with the estimators based on ranked set sampling with unequal samples and equal replications (RSSU), median ranked set sampling (MRSS) and ranked set sampling (RSS). It is shown that the relative precisions of the proposed estimator are higher than the estimators based on RSSU, MRSS and RSS for unimodal symmetric and moderately asymmetrical distributions. An example of estimating the mean diameter at breast height of longleaf-pine trees on the Wade Tract in Thomas County, Georgia, is presented.