Abstract The objective of the present study was to estimate genetic trends for birth weight (BW), weaning weight (WW), weaning staple length (WSL), yearling liveweight (YLW), and greasy fleece weight (YGFW), staple length (YSL), mean fibre diameter (YFD), fibre diameter standard deviation (YFDSD) and fibre diameter variation (YFDCV) at first shearing, which are essential to design a more efficient selection breeding programme for Chinese superfine Merino sheep. Data was collected on progeny from eight birth years (2001–2008) in the Chinese superfine Merino breeding flock maintained at Xinjiang Gongnaisi Fine Wool Sheep Breeding Farm. Different appropriate models for investigation of each trait using multivariate analysis were applied. (Co)variance components were estimated by average information restricted maximum likelihood (AIREML) and breeding values of individuals were predicted with Best Linear Unbiased Prediction (BLUP) methodology under multi-trait animal models. Genetic trends of studied traits were estimated by regressing mean of breeding values on birth year. Genetic trends were negative and significant for YFD and YFDSD (p<0.01) being 0.059 and 0.005μm/year respectively. The genetic changes for other traits were insignificant and maintained at the level of base year. Our results herein form a practical basis for designing optimal breeding schemes in Chinese superfine Merino.