Abstract: A progeny test plays an important role in forest breeding programs, but there still exist some problems in half-sib progeny test in calculating the estimates of genetic parameters and some statistics, especially when the experimental data are unbalanced. HalfsibMS is a new R-based package developed for calculating genetic parameters with balanced or unbalanced data in half-sib progeny test at multiple sites in forest trees. For the fixed-effect model, the package used the methods of restricted linear model to estimate the general combining abilities of parents and other fixed effects. For the random effect model, the package applied variance analysis method of mixed linear model to compute the genetic variance components and further the heritabilities. Moreover, HalfsibMS can calculate standard error of each parameter estimate and statistics of significance tests. For multiple traits, the new tool can also provide the genetic correlation coefficient between any two traits at the levels of family, site×family and block×family effects. This package can be downloaded from the site of http://fgbio.njfu.edu.cn/tong/halfsibms/halfsibms.html free.