The knowledge on earthworm taxonomic studies in the Philippines over the past decade dramatically increased and the results indicate an astonishingly high diversity and high local endemicity of earthworms in the country. The earthworm fauna of the Philippines is composed of pheretimoid species of the family Megascolecidae. A molecular phylogenetic study was done in attempt to infer phylogenetic relationships among the pheretimoid species in Mindanao and associated islands. Gene markers used include the mitochondrial cytochrome c oxidase subunit I (COI) and 16S rRNA, and the nuclear 28S rRNA and protein-coding histone H3 genes. Despite having limited taxa and limited genes included in the analyses, the combined data set generated a phylogeny more or less consistent with morphology-based expectation. Sims and Easton's classification scheme published in 1972 does not reflect the phylogeny with respect to the genus Amynthas and the subgenus Parapheretima; the characters they used to define these taxa are homoplasious. The species groupings in Pheretima based on the location of spermathecae are partially reflected the pheretimoid phylogeny. Particularly, members of the Pheretima sangirensis group formed a clade supported with 0.99 posterior probability with the inclusion of P. (Parapheretima) pandanensis (also with spermathecal pores at 7/8), the monothecate Pheretima vergrandis with spermathecal pore at 7/8, and an athecate conspecific of Pheretima apoensis. The members of the Pheretima urceolata group also form a weakly supported clade with the inclusion of the monothecate P. (Parapheretima) boaensis (with spermathecal pores at 5/6) and the athecate Amynthas dinagatensis. Moreover, loss of spermatheca or fusion of two spermathecae into one can occur in pheretimoid evolution, such as in the case of the monothecate P. vergrandis and the athecate conspecific of P. apoensis. Several of the branch nodes of the tree based on combined data set have support values that are very weak and formed polytomies, which is most likely due to insufficient data. The results could have improved if more data were available. The addition of genes with divergence rates slower than that of COI and 16S but faster than that of 28S and H3, such as 12S and ITS2, may also improve the resolution. Further molecular work including more taxa is needed to be able to establish a more robust system of classification of the pheretimoid species and come up with a better-resolved phylogeny. A consolidated phylogenetic study in the East Asian archipelagos will provide insights into the evolutionary, biogeographic, and ecological processes involved in island radiations of soil-dwelling animals.