Review ArticleThe geological record and phylogeny of the Myriapoda
Introduction
The fossil record and evolution of the Myriapoda were last reviewed in depth by Shear, in 1997. Since that time, new discoveries and new descriptions of important fossils, as well as recent broad-scale analyses of arthropod phylogeny and ingroup relationships of the extant myriapod taxa, have made desirable an updated overview of the field. In this article we first assess developments in our understanding of myriapod phylogeny, as a framework for a discussion of the fossils. We discuss fossil chilopods in a taxonomic framework, since there are only 5 extant orders and all known fossils except for the Devonian Devonobius can be confidently included in extant orders. Diplopod fossils are taken up chronologically, since no Palaeozoic representatives can be assigned with confidence to any of the 16 extant orders.
In the larger context of the Euarthropoda, myriapods have been difficult to place. Traditionally they have been regarded as closely allied to the Hexapoda, either as the hexapod sister group (i.e., Bäcker et al., 2008, Bitsch and Bitsch, 2004), or with the hexapods nested within a paraphyletic Myriapoda (i.e., Kraus, 2001, Willmann, 2003). But recent work utilizing both morphological and molecular characters (see Edgecombe and Giribet, 2002) has presented strong evidence that Myriapoda is not to be included in a clade with Hexapoda, now regarded as more closely related to Crustacea (Dohle, 2001). So do some myriapods belong near the base of the arthropod tree (Strausfeld et al., 2006), are they a part of the Mandibulata along with Tetraconata or Pancrustacea (Crustacea + Hexapoda; Harzsch et al., 2005), or do they form a taxon Paradoxopoda/Myriochelata with the Chelicerata (Negrisolo et al., 2004, Mayer and Whitington, 2009)? Is Myriapoda monophyletic, paraphyletic, or polyphyletic? Are the myriapod classes themselves monophyletic? And within those classes, how are the included orders related? The evidence available to be brought to bear on these questions is variable. Much morphological and diverse molecular data have been gathered for the Chilopoda (Edgecombe and Giribet, 2004, Giribet and Edgecombe, 2006), for example, and the internal phylogeny of the class now seems stable, while in the case of the Diplopoda, morphological analyses rely on relatively few characters, and only one kind of molecular marker (three nuclear coding genes) has been surveyed for a fairly restricted number of species, such that the analyses are sensitive to sampling breadth (Sierwald et al., 2001) as well as to parameters of analysis (Regier and Shultz, 2001, Regier et al., 2005, Sierwald and Bond, 2007). Exemplars of the Pauropoda and Symphyla have been included in few analyses, but in general data for these two classes are sparse, and nothing has been done regarding their internal phylogenies.
The myriapod fossil record is uneven both in a chronological and taxonomic sense. A small number of Silurian and Devonian fossils are available and their mode of preservation, as impressions in fine-grained sediments, as organically preserved cuticle, or embedded in translucent chert often allows for detailed morphological observations (Almond, 1985, Shear and Bonamo, 1988, Shear et al., 1998, Anderson and Trewin, 2003, Wilson, 2005a, Wilson and Anderson, 2004, Wilson et al., 2005). The Carboniferous holds perhaps the richest trove of myriapod fossils, albeit taxonomically biased toward the Diplopoda, and sampled from only limited habitats (Hoffman, 1969, Hannibal, 1987, Shear, 1997). In contrast, only a few myriapod fossils are known from the Permian, and none of them have been described in detail (Hannibal, 2006, Wilson, 2006a). The entire Mesozoic is similarly bereft, with but a handful of fossil myriapods (Edgecombe et al., 2009, Cockerell, 1917; Wilson, 2001, Wilson, 2003). Cenozoic myriapod fossils are almost entirely limited to a few amber Lagerstätten and are assignable to extant taxa (i.e., Santiago-Blay and Poinar, 1992).
The taxonomic distribution of myriapod fossils is strongly biased toward the Diplopoda, likely due to their greater fossilization potential; except for one small group, their cuticle is reinforced with calcium carbonate and is quite robust. However, since the cuticle is almost always consumed by the animal after moulting (to recycle the calcium), cast cuticles, which potentially enable one animal to leave behind more than a single fossil remnant, are unlikely to be available for fossilization. Further, diplopods are ecologically bound to an environment where rapid decay of organic matter, including arthropod cuticles, is facilitated by abundant bacteria and fungi. So despite their strong, mineralized cuticles, millipedes are likely to become fossils only when carcasses are quickly washed into a basin of deposition. The preservation potential of chilopods is even less. Their cuticles are relatively thin and unmineralized and their habitats similarly inimical to the survival of undecayed bodies. Fossils of symphylans and pauropods, with cuticle like that of centipedes, a soil and litter habitat, and mostly minute size, are unknown except for a few examples from Tertiary ambers (Scheller and Wunderlich, 2001, Scheller and Wunderlich, 2004).
The usefulness of the fossil record in myriapod phylogenetics is further complicated by the fact that the characters used to construct trees of living myriapods are rarely available in the fossils, with the result that a number of extinct taxa simply cannot be placed, either systematically or phylogenetically. Nevertheless, we believe that significant information can be derived from them, especially as regards the timing of cladogenesis, and in some cases studies of fossils can provide additional evidence in support of phylogenetic trees. Study of fossil myriapods is also essential to our understanding of the process of terrestrialization in arthropods, since the earliest trace fossils as well as the earliest body fossils of such creatures are of myriapods. Fossil millipedes also provide the earliest evidence of air-breathing (the Silurian Pneumodesmus newmani Wilson and Anderson has spiracles; Wilson and Anderson, 2004) and of chemical defense against predators (Devonian xyloiulideans have ozopores; Wilson, 2006a).
Section snippets
Myriapod phylogeny
Here we briefly review a variety of competing hypotheses regarding myriapod phylogeny, as set out above. Our conclusions are that the data favour a position for Myriapoda within the more inclusive Mandibulata, myriapod monophyly, and monophyly of each of the myriapod classes. Although the traditional systematic arrangement with Chilopoda sister to Progoneata, and the division of Progoneata into Dignatha (Diplopoda + Pauropoda) and Symphyla has been challenged, we find no strong basis for
Interrelationships of Chilopoda
From the perspective of morphology, the relationships between the 5 extant orders of Chilopoda have widespread consensus. The relationships depicted in Fig. 1 conform to groupings recognized in early classifications (Pocock, 1902; Verhoeff, 1902–1925), were depicted in pre-Hennigian phylogenetic diagrams (Prunescu, 1965, Shinohara, 1970), were likewise retrieved when cladistic argumentation was applied to the problem (Dohle, 1985, Dohle, 1990, Shear and Bonamo, 1988, Borucki, 1996), and are
Interrelationships of Diplopoda
As indicated in Section 1, the phylogeny of millipedes is unsettled when compared to that of centipedes. Phylogenetic problems in the Diplopoda are inherently more complex, since 16 extant orders of millipedes (including 144 families; Shelley, 2003) are currently recognized, as opposed to only 5 orders in centipedes. Fewer morphological characters have been studied, and entire potentially valuable character systems, such as the mouthparts, female external genitalia, and legs have either not
Pauropoda and Symphyla
No reliable internal phylogenies for these orders have been proposed, and they are taxonomically understudied, despite the herculean efforts of Ulf Scheller, virtually the only active specialist apart from pauropod taxonomist Yasunori Hagino. Although symphylans and pauropods are known as fossils only from the Baltic (Scheller and Wunderlich, 2001, Scheller and Wunderlich, 2004) and Dominican ambers (Poinar and Edwards 1995), specimens which can be placed in extant families and genera, the
Kampecarida
Kampecarids are enigmatic but relatively common myriapod-like fossils from Late Silurian and Early Devonian deposits in Britain. They were examined in an unpublished thesis by J.E. Almond (with an overview in Almond, 1985), upon which Shear (1997) based some tentative conclusions and a reconstruction. The kampecarid head may have been diplopod-like, with antennae and mandibles, and may either have been covered by two separate plates or followed by a legless collum. At least some of the trunk
Fossil stem-group myriapods?
The problem of identifying a fossil stem-group myriapod was addressed by Edgecombe (2004). While conceding that the presence of stem-group myriapods in the Cambrian is phylogenetically sound, and is indeed strengthened by the discovery of apparent crown group representatives of its sister group, Tetraconata, in the early Cambrian (Harvey and Butterfield, 2008), Edgecombe (2004) found that a variety of candidate fossils from the Lower and Upper Cambrian had in common with Myriapoda only a series
Acknowledgements
Günther Bechly kindly provided the photographs of chilopods from the Cretaceous of Brazil. Markus Koch prepared and photographed the Craterostigmus maxillipedes in Fig. 2C. Joe Hannibal provided photographs of Paleozoic diplopods. Work on this article by WAS was supported by a grant from the National Science Foundation of the United States (DEB05-29715) to WAS, Petra Sierwald and Jason Bond, and by the Faculty Professional Development Committee of Hampden-Sydney College.
References (138)
- et al.
A forgotten homology supporting the monophyly of Tracheata: the subcoxa of insects and myriapods re-visited
Zoologischer Anzeiger
(2008) - et al.
The endoskeletal structures in arthropods: cytology, morphology and evolution
Arthropod Structure & Development
(2002) - et al.
Parental care in Dicellophilus carniolensis (C.L. Koch, 1847): new behavioural evidence with implications for the higher phylogeny of centipedes (Chilopoda)
Zoologischer Anzeiger
(2002) Arthropod phylogeny: an overview from the perspectives of morphology, molecular data and the fossil record
Arthropod Structure & Development
(2010)- et al.
Anamorphosis in millipedes (Diplopoda) - the present state of knowledge, with some developmental and phylogenetic considerations
Zoological Journal of the Linnean Society
(1993) - et al.
Common design in a unique midline neuropil in the brains of arthropods
Arthropod Structure & Development
(2002) - et al.
Further use of nearly complete 28S and 18S rRNA genes to classify Ecdysozoa: 37 more arthropods and a kinorhynch
Molecular Phylogenetics and Evolution
(2006) - et al.
Ecdysozoan phylogeny and Bayesian inference: first use of nearly complete 28S and 18S rRNA gene sequences to classify the arthropods and their kin
Molecular Phylogenetics and Evolution
(2004) - et al.
The fine structure of the eyes of some bristly millipedes (Penicillata, Diplopoda): additional support for the homology of mandibulate ommatidia
Arthropod Structure and Development
(2007) The centipedes (Chilopoda) of the Mazon Creek