728 human genes were divided to four groups according to the GC contents of their coding sequences (from GC＜0.43 to GC＞0.58). Examination of synonymous-codon bias in the 4 groups show that NTG (N represents any base of T, A, C, G) is most favored and NCG is most avoided in all four groups. Statistical correlation analysis of GC content in genetic environment with C3/G3 content (the C or G in the 3rd position of codons) and the codon bias in the four groups suggest that C-ending codons are special preferred. This is in favor of accurate translation. Frequency of each amino acid in the four groups was also examined.