The Alu family of Short interspersed repeats (SINEs) account for a significant portion of the "junk DNA" within the mammalian genome. Thousands of copies of Alu subfamilies are scattered essentially randomly through the human genome. The relative abundance of these repeats provides a rich fossil record of primate and human history. Since Alus have no known functionality, a great deal of ambiguity surrounds their amplification and evolution. Herein, using parsimony approach, we investigate the popular belief that the vast majority of Alu amplification is derived from a small subset of Alu master genes. Our results indicate that the classification of Alu subfamilies reported in literature is incomplete. Furthermore, we summarize some of the factors relevant to the use of the Alu family of repeats in bioinformatic assays.