AbstractCurrently, biological classification and determination of different categories are all based on empirical knowledge,which is obtained relying on morphological and molecular characters. For these methods they lacks of absolutely quantitative criteria ground on intrinsically scientific principles. In fact, accurate science classification must depend on the correct description of biology evolution rules.In this article a new theoretical approach was proposed, in which two characteristic constants were gained from biological common heredity and variation information theory equation, when it is at the maximum information states, corresponding to symmetric and asymmetric variation states. They are common composition ratios, Pg =0.61, and Pg=0.70. By analyzing the common composition ratios of compounds among oleoresins, two pine subgenus:Subgen.Strobus (Sweet) Held and Subgen. Pinus could be integrated into one class, Genus pinus, excellently, when Pg = 0.61.These two pine subgenus could be classified into two groups clearly,when Pg = 0.70.The results is somewhat different from that achieved by means of classical classification relying on morphological characters. On the other hand, the evolution relationship of two subgenus was analyzed based on characteristic sequences of samples, it indicated that white pine origin from pinus tabuliformis. The two constants should be used as the classification constants of some biological categories of plants.