An experiment was conducted to study effects of the sources and levels of dietary Zn on the performances and characteristics of serum and feces in weanling pigs. Thirty six, 18 of each sex. 4 wks old. 3-way cross bred (Y x L x D) pigs were assigned to 6 treatments in a 2 x 3 factorial design (Zn sources x levels). The supplemental In sources and levels were ZnO or Zn-methionine chelate (Zn-Met) at 100, 1,000 or 2,000 ppm Zn, respectively. Three pigs were housed in a raised floor pen and 2 pens were assigned to each treatment. Individual pig weight and feed intake of each pen were recorded weekly for 4 wks. ADFI and ADG were significantly (P$lt;0.05) different among the treatments, in which Zn-Met-100 (100 ppm Z.n) were the highest. The lowest ADFI and ADG were shown in ZnO-100 (100 ppm Zn) and Zn-Met-2,000 (2.000 ppm Zn). respectively. There were no significant differences in feed conversion ratio and serum IgG concentrations among the treatments. APase activity in serum and Zn level in serum and feces showed significant (P$lt; 0.01. P$lt; 0.05 and P$lt; 0.01, respectively) differences among the treatments and were significantly (P$lt; 0.01) influenced by the Zn level. The highest APase activity and Zn level in serum were shown in Zn-Met-2000 and the highest Zn level in the feces was shown in ZnO-2000 (2,000 ppm Zn). Effects of Zn source on Zn level in serum and feces were significant (P$lt;0.01 and P$lt;0.02, respectively). Zn-Met treatments had higher serum Zn level and lower fecal Zn levels than In0 treatments. Digestibility of dry matter, crude protein and crude fiber were significantly lower in Zn-Met-2000 than in the other treatments. In conclusion, the availability of Zn from Zn-Met was higher than that from ZnO. Zn-Mct-100 showed equal or better performance and lower Zn excretion than ZnO-2000. However, the use of Zn-Met-2000 had an adverse effect on the performances of weanling pigs.