The Protura is a group of mysterious soil-dwelling micro-arthropods (usually 0.5-2.0 mm in length), first described by Silvestri in 1907 . Traditionally, it was regarded as a basal hexapod group, but it owns many unique and primitive morphological characteristics compared with other hexapods. For example, they lack antennae and wings, the foretarsus are enlarged with many sensilla serving the role of antennae, eyes and tentorium are absent, they have anamorphic post-embryonic development, and they have 12 abdominal segments (instead of 11) . The proturan spermatozoan has a variable number of doublet microtubules (9-16), with no accessory or central microtubules. It is different from those of other hexapods, but similar to the sperm of sea spider (Arthropoda: Pycnogonida). This probably reflects a high diversification rate, or a lengthy evolution [3–5]. Historically, there were many controversies about the relationship of proturans to other hexapods, and their evolutionary position in the Arthropoda [2, 3, 6–9]. This is because proturans are understudied, being so small and rare, making them difficult to collect, identify, culture and experiment on [2, 10, 11].
The higher-level phylogeny of the major arthropod groups (Chelicerata, Myriapoda, Crustacea and Hexapoda) continues to be a matter of debate despite extensive research based on phylogenetic analysis and genetic data [12–14]. Almost all molecular analyses strongly support the Pancrustacea hypothesis: crustaceans, instead of myriapods, are the closest relatives of the hexapods [15–18]. The Hexapoda (Insecta s. lat.), which includes four groups, Protura, Collembola, Diplura and Insecta (Insecta s. str.), was traditionally considered a monophyletic lineage based on the synapomorphies of body segments, six legs on the thorax, and adaptation to the terrestrial environment. The monophyly of the Insecta has been well established by morphological and molecular studies [8, 10, 17, 18], but the monophyly of the Hexapoda is less certain [17, 19]. Three basal hexapod groups (Protura, Collembola and Diplura) show many different features from insects according to morphology [10, 20] and ultrastructure of spermatozoa . The mitogenomic data of basal hexapod collembolans and diplurans reject the monophyly of Hexapoda, and suggest that some crustaceans are more closely related to the Insecta than Collembola and Diplura [17, 19, 21]. However, recent studies based on EST data and nuclear genes (18S and 28S ribosomal RNA genes, nuclear protein-coding sequences) support the monophyly of the Hexapoda [12, 13, 18].
The arthropod mitochondrial genome is a single circular DNA molecule encoding 13 proteins, 22 transfer RNAs (tRNAs), two ribosomal RNAs (rRNAs), and one A+T-rich region for the control of replication and transcription of the mtDNA. It is used extensively for studying phylogenetic relationships at various taxonomic levels. Unlike nuclear molecular markers, mtDNA is of maternal inheritance, and does not experience intermolecular genetic recombination. In addition, the mitochondrial gene order can provide additional phylogenetic information, since rearrangements appear to be generally rare events, and most mitochondrial gene arrangements often remain unchanged over a long evolutionary period . Mitogenomic data also strongly support the Pancrustacea hypothesis [14, 17, 23], especially with the evidence of the gene order [16, 24]. The gene trnL2 (UUR) is located between rrnL and nad1 in the ancestral arthropod ground pattern, but is translocated to the position between cox1 and cox2 in Pancrustacea . It has been considered a distinctive synapomorphic character for crustaceans and hexapods. The mitochondrial genomes of basal hexapod Collembola  and Diplura  also agree with the "cox1-trnL2-cox2" pattern. So far, no mitochondrial genome information is available for the Protura. This has impeded comprehensive discussions on the evolution of the arthropod mitochondrial genome, and the validity of using mtDNA to study the phylogeny of the Hexapoda [27–29].
In this study, we sequenced the complete mitochondrial genome of Sinentomon erythranum (Protura: Sinentomata: Sinentomidae), to describe the molecular features of the proturan mitochondrial genome, to judge how these evolved, and to see if it has any phylogenetic information, which may help resolve the discrepancy on the monophyly of the Hexapoda between mitochondrial and nuclear DNA markers.