Discovery of the centipede family Plutoniumidae (Chilopoda) in Asia: a new species of Theatops from China, and the taxonomic value of spiracle distributions in Scolopendromorpha Author Di, Zhiyong Author Cao, Zhijian Author Wu, Yingliang Author Yin, Shijin Author Edgecombe, Gregory D. Author Li, Wenxin text Zootaxa 2010 2667 51 63 journal article 10.5281/zenodo.276390 df2ba19d-2faa-419c-a99d-98c12d2864df 1175-5326 276390 Theatops Newport, 1844 Theatops Newport, 1844 : 409 . Attems 1930 : 250 –251. Shelley 1997 : 71 –74 (see full synonymy p. 71); Shelley 2002 : 81 . Opisthemega Wood, 1862 : 35 (synonymy by Pocock 1888 : 287 ). Opisthomega Saussure & Humbert, 1872 : 200. Type species: Cryptops postica Say, 1821 , by subsequent monotypy of Newport (1845) (see Shelley 1997 : 72–73). Diagnosis: Plutoniumidae with 9–10 pairs of spiracles, on segments 3, 5, 7 (variable), 8, 10, 12, 14, 16, 18 and 20. Assigned species: T. californiensis Chamberlin, 1902 ; T. chuanensis n. sp. , T. erythrocephalus ( C.L. Koch, 1847 ) , T. phanus ( Chamberlin, 1951 ) , T. spinicaudus ( Wood, 1862 ) . Discussion: The concept of Theatops employed here differs from previous diagnoses in allowing for variability with respect to the absence or presence of a spiracle on segment 7. All previously known species have nine pairs of spiracles but the morphological spectrum of Plutoniumidae is altered by the discovery of T. chuanensis , described below, that corresponds to Theatops in every respect apart from having an extra pair of spiracles, on segment 7. The presence or absence of a spiracle on segment 7 is a character that has traditionally been interpreted as a generic-level feature in Scolopendromorpha ( Attems 1930 ) or even at higher taxonomic levels, such as subfamilies ( Schileyko 1992 ). Examples of genera that are distinguished solely by this character include Scolopocryptops Newport, 1844 (without it) and Dinocryptops Crabill, 1953 (with the spiracle) in the Scolopocryptopidae ; and Otostigmus Porat, 1876 (without it) and Ethmostigmus Pocock, 1898 , and Rhysida Wood, 1862 (with the spiracle) in the Scolopendridae ( Kraepelin 1903 ; Attems 1930 ). Following traditional taxonomic practice, Theatops chuanensis would be regarded as a separate, monotypic genus, diagnosed by the spiracle on segment 7. Here we advance arguments for not following this course. Evidence is increasingly accumulating to indicate that the presence or absence of a spiracle on segment 7 is less reliable as a phylogenetic and taxonomic character than previously thought. In the case of Scolopocryptops versus Dinocryptops , the absence of a spiracle on segment 7 maps onto morphology-based cladograms as a shared primitive character and the group lacking the spiracle ( Scolopocryptops ) is paraphyletic with respect to the one that possesses it ( Dinocryptops ) ( Edgecombe & Koch 2008 ; Koch et al. 2009, 2010). The evolutionary lability of the spiracle on segment 7 in Scolopocryptopinae was noted by Crabill (1955) , who reported an otherwise normal specimen of Scolopocryptops sexspinosus that had a small but evidently functional spiracle on segment 7 (Crabill based his interpretation of function on the presence of tracheae in segment 7 of the specimen). Crabill (1955: 134) , however, interpreted character polarity opposite to that optimised on cladograms: he “assume[d] the presence of seventh somite spiracles to be primitive possessions”, and despite “the loss of seventh somite spiracles… as evidence of parallel evolution in this character”, he maintained their taxonomic value (“That this is a sound generic distinction cannot, at the present state of knowledge, be seriously doubted”). This confidence is undermined by the discovery that Scolopocryptops nigrimaculatus Song et al. , 2004 , may exceptionally show a small segment 7 spiracle on one side of the body only versus its (normal) absence on the other side ( Song et al. 2004 : fig. 3C). With respect to non-monophyly at least, the situation is not dissimilar in Otostigminae. Apart from lacking the autapomorphies of other genera, Otostigmus is diagnosed only by a single character (absence of the segment 7 spiracle), and analyses of molecular data in progress retrieve both Otostigmus and Rhysida as polyphyletic groups. Although genera (or subfamilies: Schileyko 1992 ) distinguished by the presence or absence of a spiracle on segment 7 are easily identified and taxonomically convenient, in the case of Otostigminae they do not reflect phylogeny accurately. Lewis (2004) provided a possibly relevant observation from Otostigminae in Tanzania , wherein Otostigmus tanganjikus Verhoeff, 1941 , and Rhysida intermedia Attems, 1910 , are apparently distinguished by no character apart from the spiracle on segment 7; in these instances it might be anticipated that the similar species pairs are more closely related than either is to most species assigned to the same genus under the standard classification. In this case, the segment 7 spiracle may remain a reliable character at the species level, but likely not at the generic level. In Plutoniumidae, the morphological analyses of Edgecombe & Koch (2008) and Koch et al. (2009, 2010) retrieved Theatops as a paraphyletic group, i.e., Plutonium is nested within Theatops . As in the previous examples in which spiracle distributions are the principal basis for diagnosing genera, paraphyly of one genus is not surprising because Theatops is diagnosed only by primitive characters (the primitive spiracle arrangement, contra Crabill’s inferred ancestral condition). A three-genus classification ( Theatops : lacking spiracles on segment 7; a new genus for T. chuanensis with spiracles on segment 7; and Plutonium : with spiracles on segments 2–20) would increase paraphyly rather than lessen it, i.e., Theatops would be paraphyletic to two genera rather than one. We do not place Plutonium in synonymy under Theatops , which would eliminate non-monophyletic taxa from Plutoniumidae, but we do discard the option of erecting a monotypic genus for T. chuanensis . Knowledge of Theatops chuanensis is limited to a single specimen and it can rightly be questioned whether its segment 7 spiracle could be an abnormality in a population or species that more commonly lacks a spiracle, by comparison to the examples cited above in which Scolopocryptops species only exceptionally have a spiracle on that segment. The likelihood that the segment 7 spiracle is an aberration in T. chuanensis is contradicted by the fact that the spiracles on segment 7 are as large as those on segment 8, are equally developed on both sides of the body, and the two segments have identical proportions of the stigmatopleurites. In contrast, in Scolopocryptops species that only rarely display a segment 7 spiracle, it is either conspicuously smaller than that on other segments ( Crabill 1955 ) or both smaller than that on segment 8 and restricted to only one side of the body ( Song et al. 2004 ).