In the process of preparing and analyzing the metabonomics samples, there are lots of influential factors such as sample individual differences, environmental factors and human factors. Here we report a study on 3 healthy male SD rats of different samples carried out by different operators (n=4). It was a two-way experiment designed to explore the impact factors and lay the foundation for future experiments. The cluster analysis was employed in comparing 12 samples. The results showed that the differences of rats between the sample groups had a certain influence on the metabonomics data, and there was relatively little variability between the 4 operators. That is to say, whilst the sample are rather different from one another, a given sample is generally quite similar to itself in metabonomics study.