The analysis of individual quantitative relationship among community species， also known as the Species Abundance Distribution （SAD）， is considered to be the key to understanding what determines species diversity. In recent years， numerous SAD models have been proposed on various theoretical grounds， but it is difficult to draw general conclusions about which models provide the best fit to SADs. Previous studies have shown that the screening of SAD model should not only examine the goodness of fit of SAD model with the empirical data of community samples， but also evaluate model's ability to simultaneously explain some macro ecological patterns. The fractal model proposed by Su has good fit to the empirical data of community samples and reveals the general pattern of SAD； that is， N_r / N₁ tends to be 1∶1/2∶1/3… （N_r/N₁， N_r and N₁ represent the number of individuals of the r-th and the first species in descending order）. However， the model has not been given enough attention， and the relevant verification research is also lacking. This paper uses a global community species database to test the model and the general pattern of SAD. The results show that： \mathrm{①} the fractal model has a good fit to the database； \mathrm{②} the general pattern of SAD can be supported by the database. These results might provide a more reliable scientific basis for understanding SAD and exploring the determinant mechanism of species diversity.