Common Family Of Arixeniidae

7.43. Reconstruction of Gerarus danielsi (Geraridae), from the Late Carboniferous of Mazon Creek, Illinois. Gerarids were Paleozoic orthopterids and a stem group to the Titanoptera. Redrawn from Burnham (1983).

7.44. The caloneurodeans, such as Paleuthygramma tenuicornis (Paleuthygrammatidae) from the Permian of Russia, were enigmatic relatives of early Orthoptera and Phasmatodea that became extinct probably at the end of the Permian. Length 24 mm; redrawn from Carpenter (1992).


Little is known of this extinct Paleozoic order of poly-neopteran insects (Figure 7.44). The order is noteworthy for the secondary loss of the anal area in the hind wing, the unbranched and nearly parallel (or fused in a few groups) veins CuA and CuP in both the fore- and hind wings, and, where known, unsegmented cerci. Other features of the order include the strongly convex and concave wing veins; five-segmented tarsi; long, multisegmented antennae; and fore-and hind wings having a similar shape, venation, and texture. Presently there are nine families, many with a single genus, and the systematics of the group requires considerable revision. Nothing is known of caloneurodean biology aside from the fact that they were terrestrial, apparently primitively resembling cursorial orthopterids. The caloneurodeans are known only from the Late Carboniferous and Permian.


Almost anyone can immediately recognize an adult earwig with its stout, terminal forceps (Figure 7.45) and the short, tegminous forewings. Less conspicuous are the three-segmented tarsi; very distinctive, greatly expanded hind wing vannus (Figure 7.50), with a dramatically reduced remigium and unique folding mechanism; a prognathous head (lacking a gula); absence of ocelli; a subgenital plate formed by an enlarged seventh sternum in females; and a vestigial ovipositor. The uniquely expanded anal fan of the hind wing may eventually prove to be independently derived from that of other poly-neopterans, in which case Dermaptera would be classified elsewhere. Earwigs are overflowing with unique features and the chronology of these is even apparent in the fossil record.

Dermaptera are distributed globally except Antarctica and the extreme Arctic, but most of the nearly 1,900 described species occur in tropical to warm-temperate habitats (e.g., Steinmann, 1986, 1989a,b,c, 1993) (Figure 7.46). Earwigs typically live in riparian habitats, in crevices, in leaf litter, or under the bark of trees. Most species are nocturnal and omnivorous; only a very few species are strictly herbivorous or carnivorous (Chopard, 1938). The cercal forceps are used to capture prey and are employed in mating and in folding the hind wings under the tegmina (Kleinow, 1966). Female earwigs exhibit extended maternal care over the eggs and early instar nymphs, carefully cleaning them to protect from invasive fungi (Herter, 1943; Rentz and Kevan, 1991). After two molts, however, nymphs must fend for themselves, or they will be eaten by the mother.

Earwigs have a long geological history. The oldest fossils to date are tegmina from the Late Triassic-Early Jurassic of England and Australia (e.g., Jarzembowski, 1999). As already

7.45. The modification of cerci into forceps is the most recognizable trait of the earwigs (order Dermaptera). The forceps are used to defend, to capture prey, and to assist in folding the fanlike wings under the short tegmina. Not to the same scale.

discussed and what should be of little surprise, some traits believed by neontologists to be diagnostic for an earwig do not apply to the earliest earwigs. Basal, extinct members of the order all had five-segmented tarsi, well-developed ovipositors, veined tegmina, and long, multisegmented cerci. Traditionally, primitive earwigs from the Late Jurassic and Early Cretaceous were classified in the suborder Archi-dermaptera (e.g., Bei-Bienko, 1936; Vishniakova, 1980; Carpenter, 1992), a paraphyletic stem group to modern Dermaptera. Archidermaptera in a restricted sense contains the Jurassic-Cretaceous families Protodiplatyidae, Turanovi-idae, and Dermapteridae, which comprise the basalmost lineage of the order (Willmann, 1990a; Haas and Kukalova-Peck, 2001; Engel, 2003c) (Figure 7.47). Archidermapterans are a sister group to the Pandermaptera, which comprise two further suborders: Eodermaptera, for the Jurassic-Cretaceous families Semenoviolidae and Turanodermatidae, and the Neodermaptera, which contains all the modern lineages. Eodermapterans share with Neodermaptera the derived development of unsegmented, forcipate cerci but primitively retain venation in their tegmina, presence of ocelli, and pen-tamerous tarsi. The Neodermaptera have three-segmented tarsi, no ocelli, and lost venation in their tegmina. They first appear in the Early Cretaceous (e.g., Popham, 1990; Engel et al., 2002) but may have originated in the latest Jurassic since there is a putative, undescribed pygidicranoid from the Jurassic of central Asia (Rasnitsyn and Quicke, 2002). Certainly, definitive neodermapterans (Figure 7.48) and recognizable pygidicranids are known by the mid-Cretaceous (Grimaldi et al., 2002; Engel and Grimaldi, 2004c) (Figure 7.49). Tertiary earwigs, mostly of the Forficulidae, are preserved in several deposits, and a relatively unexplored diversity of species is known in Baltic (Burr, 1911), French (Nel etal., 2002c); and Dominican ambers (Figure 7.51).

The internal phylogeny and classification of Neo-dermaptera has been in constant flux, with dramatically different arrangements of families and superfamilies by contemporaneous authors. Recent phylogenetic work has begun to shed light on relationships within the suborder (e.g., Haas, 1995; Haas and Kukalova-Peck, 2001; Colgan et al., 2003) (Figure 7.51). Within Neodermaptera, the infraorder Proto-dermaptera, including the superfamilies Pygidicranoidea and Karschielloidea, is basal but perhaps paraphyletic (e.g., Haas, 1995; Haas and Kukalova-Peck, 2001). Unlike all other earwigs, the protodermapterans have ventral cervical scle-rites of equal size, carinae on the femora, and a segmented pygidium. Almost all Cretaceous records of Dermaptera are protodermapterans. The infraorder Epidermaptera, including all other Neodermaptera, is characterized by the enlargement of the posterior ventral cervical sclerite, rounded femora, and fusion of the three pygidial sclerites (Popham, 1985).

7.46. Representative Recent earwigs. Assembled from Genera Insectorum.

7.47. An archidermapteran, Microdiplatys campodeiformis (Pro-todiplatyidae), from the Late Jurassic of Karatau in Kazakhstan. Archi-dermaptera lacked the cercal forceps and primitively retained some veins in the forewings, though the tegminous forewings were short as in modern earwigs. PIN 2904/441; length, excluding cerci, 10 mm. -►

7.48. (right) A neodermapteran earwig from the Early Cretaceous of Brazil. AMNH; length 5 mm.

7.47. An archidermapteran, Microdiplatys campodeiformis (Pro-todiplatyidae), from the Late Jurassic of Karatau in Kazakhstan. Archi-dermaptera lacked the cercal forceps and primitively retained some veins in the forewings, though the tegminous forewings were short as in modern earwigs. PIN 2904/441; length, excluding cerci, 10 mm. -►

7.48. (right) A neodermapteran earwig from the Early Cretaceous of Brazil. AMNH; length 5 mm.

Although two additional suborders (or infraorders) have been traditionally recognized for the parasitic earwigs, both are actually derived Epidermaptera. The families Hemimeri-dae and Arixeniidae are ectoparasites with a suite of paedo-morphic characters (i.e., the retention of nymphal traits in the adult). The Hemimeridae is perhaps closely related to Apachyidae (Klass, 2001b), while Arixeniidae is related to, if not derived from within, Spongiphoridae (Popham, 1985). Hemimerids live on murid rodents in Africa. Two genera are recognized: Hemimerus, which consists of nine species living on Cricetomys rats, and Areomerus, which consists of two species living on Beamys rats (Rehn and Rehn, 1936; Nakata and Maa, 1974). The five species of arixeniids live in the roosts of Cheiromeles bats (Molossidae) in southeast Asia, where they feed on secretions from the skin or on other insects invading the roost (Medway, 1958; Nakata and Maa, 1974). Hemimerids were at one time considered a distinct order, called Diploglossata, Dermodermaptera, or Hemime-rina (e.g., Verhoeff, 1902; Brues and Melander, 1915; Popham, 1961). As with many ectoparasitic insects, fossil Hemimeri-dae and Arixeniidae have yet to be discovered, and the families are probably no older than the mid-Tertiary.

Dermaptera are believed by some to stem from the Permian order Protelytroptera owing to tegminous forewings and the large, uniquely formed anal fan (Kukalova-Peck, 1991; Haas and Kukalova-Peck, 2001). Some families have erroneously been placed in Protelytroptera, specifically the Cretaceous Umenocoleidae, which is actually a highly specialized lineage of roaches (Figures 7.70, 7.71). The possibility exists that Permian Protelytroptera were earwig progenitors, but definitive evidence is still required to establish this.


7.50. Phylogenetic relationships among major lineages of earwigs. Significant characters indicated in Table 7.2. Relationships based on Haas and Kukalova-Peck (2001). Thick dark lines are the known extent of fossils, lighter thick lines indicate fossils possibly belonging to those groups.

TABLE 7.2. Significant Characters in Dermaptera Phylogeny"

1. Cerci unsegmented, forcipate

2. Trimerous tarsi

3. Ocelli lost

4. Tegminal veins lost

5. Ovipositor reduced

6. Posterior, ventral cervical sclerite enlarged

7. Three pygidial subsegments fused

8. Reduction to single penal lobe and single virga

9. Expanded regions of anal and intercalary veins distinctly separated a Numbers correspond to those on phylogeny, Figure 7.50.

7.49. The earliest pygidicranoid earwig, Burmapygia resinata (Pygidi-cranidae), in mid-Cretaceous amber from Myanmar. Pygidicranids are among the most primitive of living earwigs. AMNH Bu274; length 6 mm; from Engel and Grimaldi (2004c).

7.51. A forficulid earwig in Miocene amber from the Dominican Republic. The hind wings of earwigs are remarkably distinctive, with most of the veins fused at the base and a large anal fan comprising most of the wing. Morone Collection, M3400; body length 6 mm. Photo: R. Larimer.


Grylloblattodea (or Notoptera) represent an interesting, relict lineage today confined entirely to the Northern Hemisphere. In a class renowned for its overwhelming diversity, the Grylloblattodea, along with Zoraptera and Mantophasmatodea, hold the distinction of being the least diverse of insect orders. Today there are 26 species classified into five genera within a single family - Grylloblatta from the northwestern United States and southwestern Canada; Grylloblattella and Gryl-loblattina from the Russian Far East; Galloisiana from Japan, Korea, China, and Russia; and Namkungia from Korea but which probably makes Galloisiana paraphyletic (Storozhenko, 1997, 1998; Storozhenko and Park, 2002). These soft-bodied, wingless insects are typically found in leaf litter or under stones in cold temperate forests, often at higher elevations, although some blind Asian species have been discovered in caves (Namkung, 1982; Nagashima, 1990). Species are active at cold temperatures, and several studies have indicated optimal temperatures around 1-4°C (Mills and Pepper, 1937; Henson, 1957) (Figure 7.52). Although ice crawlers prefer low temperatures, they are not impervious to freezing. Individuals of Grylloblatta can be killed by ice formation within the body at around-8°C (Morrissey and Edwards, 1979). During winter months when night temperatures drop below freezing, ice crawlers likely survive under the snow-pack and near the soil where temperatures may deviate little from freezing (Atchison, 1979). These insects are omnivorous and typically scavenge dead arthropods but rely on plant material when frozen carcasses become scarce (Pritchard and Scholefield, 1978; Nagashima et al., 1982).

Modern Grylloblattodea have numerous defining features, such as a median, eversible sac on the first abdominal sternum; the loss of ocelli (also in Mantophasmatodea and perhaps shared through common ancestry in these two lineages); and the asymmetrical male genitalia. Unique among all hexapods is the presence of a spina on the metathoracic sternum. Although difficult to place in the greater scheme of insect phylogeny, ice crawlers are probably basal members of Orthopterida and were also thought to be "living fossils" even by their discoverer (Walker, 1914, 1937). Wings are absent in modern grylloblattids and other characteristics are generally primitive: the five-segmented tarsi; long, multisegmented cerci; and an ovipositor composed of three stout pairs of "valvulae" intermediate to the orthopterid ovipositor. The third pair of valvulae are, in fact, the gonoplacs. In Grylloblat-todea the gonoplac is greatly developed and incorporated into the ovipositor, representing an intermediate stage among orthopterids where the function of the reduced, second gonapophyses is assumed by the gonoplac. Gonoplac development in ovipositor construction is a typical orthopterid feature.

The presence of enlarged coxae, which is typical of basal Dictyoptera, has at times been used to support a relationship

7.52. A rock crawler, Grylloblatta (Grylloblattodea). The order today consists of 26 species in the northern parts of the Holarctic Region. They require cool to cold temperatures. Photo: Alex Wild.

of Grylloblattodea, along with Dermaptera and Mantophas-matodea, with this group. This placement has also been supported to some degree by molecular analyses (e.g., Maekawa et al., 1999; Wheeler et al., 2001). However, if this trait is primitive, as suspected by Hennig (1981), then it has little bearing on the placement of these orders. Alternatively, the analyses of Kamp (1973: later reanalyzed by Kuperus and Chapco, 1996) weakly supported a Grylloblattodea + Dermaptera relationship but placed these orders as relatives of the Orthopterida. Ultrastructure of the spermatozoa and embry-ological development also support a relationship to Orthoptera (Ando and Nagashima, 1982; Baccetti, 1982).

Generally believed to have an extensive geological history (e.g., Storozhenko, 1997, 1998; Vrsansky et al., 2001; Rasnitsyn and Quicke, 2002), the Grylloblattodea has essentially taken on the role of "Protorthoptera" in some modern systems of insect classification. In such systems, most protorthopteran families have been transferred to the Grylloblattodea. As previously discussed, many of these Paleozoic and Early Mesozoic families are not related, and any discussion of past "grylloblattodean" diversity based on these is misleading. In fact, this is highlighted by the phylogenetic outlines of such systems, where Grylloblattodea is depicted as giving rise to all other Polyneoptera (e.g., Storozhenko, 1997, 1998; Rasnitsyn and Quicke, 2002). Thus, quite in opposition to other authors, we presently believe the ice crawlers to have been of modest diversity in the beginning as well as today. Some taxa in the fossil record appear to share with modern Grylloblattodea a similar ovipositor construction and may truly represent stem-group ice crawlers. Interestingly, these Jurassic and Permian families, like Blattogryllidae (Jurassic), Megakhosaridae (Jurassic), and Tillyardembiidae (Permian), possessed fully formed wings (Figures 7.53, 7.54). The wing venation of these fossils has not been critically explored for derived features potentially uniting Grylloblattodea with other orders. Unlike Orthopterida, however, the precostal field was hardly developed in these fossils, but they did possess enlarged anal fans in the hind wing, supporting the poly-neopteran position of the order. If Grylloblattodea and Man-tophasmatodea are living sister groups, then the loss of wings might be a defining feature of the combined lineage, and the Mesozoic fossils may represent a stem group to both (further suggesting that the two orders should be united; see discussion

7.53. Although at one time considered an ally of the webspinners, Tillyardembia antennaeplana (Tillyardembiidae), from the Permian of Russia, is now recognized as an early relative of the Grylloblattodea and perhaps Mantophasmatodea. PIN 1700/1177; length 22 mm.

7.54. Blattogryllus karatavicus (Blattogryllidae) from the Late Jurassic of Karatau in Kazakhstan, a stem-group grylloblattodean. Modern Grylloblattodea and their close relatives Mantophasmatodea are wingless, but stem groups to the lineage were fully winged and possessed the anal fan typical of polyneopterans. PIN 2554/227; length 28 mm.

7.54. Blattogryllus karatavicus (Blattogryllidae) from the Late Jurassic of Karatau in Kazakhstan, a stem-group grylloblattodean. Modern Grylloblattodea and their close relatives Mantophasmatodea are wingless, but stem groups to the lineage were fully winged and possessed the anal fan typical of polyneopterans. PIN 2554/227; length 28 mm.

later in this chapter). Other fossil lineages sometimes placed in the Grylloblattodea are related to other orders, such as Lemmatophoridae, which is more closely related to Ple-coptera. Limited available evidence suggests that Grylloblattodea diversity has changed little through geological time, although the loss of wings in the Cretaceous or Early Tertiary represents a significant morphological modification in Recent ice crawlers.


The African rock crawlers are the most recently discovered order of insects (Klass et al., 2002) (Figures 7.55, 7.56), and, not surprisingly, they are already stirring debate. Little information has accumulated or permeated into the literature, and thus any discussion for the moment must remain tentative. Modern Mantophasmatodea occur in xeric, rocky habitats in southern Africa. These insects are apparently aggressive carnivores, pouncing on prey and grasping their victims with the fore- and mid-legs. Rock crawlers tend to be nocturnal, feeding on unsuspecting moths, silverfish, and roaches, but take most small arthropod prey they can catch and subdue. During daylight hours rock crawlers hide among stones and the bases of plants, particularly clumps of grass or spiny shrubs in South Africa's Succuluent Karoo (e.g., Walker, 2003). The 15 species are segregated into three families, although these should perhaps be downgraded to subfamilies of a single family owing to the homogenous habitus of all members of the group. The principal papers for Mantophasmatodea are Zompro (2001), Klass et al. (2002, 2003a,b), Zompro et al. (2002, 2003), and Engel and Grimaldi (2004b).

The group is monophyletic, defined by the combination of the following traits: a loss of ocelli (although, as mentioned, perhaps a trait of Grylloblattodea + Mantophasmatodea); a hypognathous head; loss of the epistomal sulcus, and the unique subgenal sulcus that loops from the posterior mandibular articulation to the anterior tentorial pit and then back to the anterior mandibular articulation; a loss of wings (perhaps also shared with Recent Grylloblattodea); an enlarged pretarsal arolium with a series of long setae; a vomer-like process articulating on the apical margin of the tenth sternum; and unsegmented cerci (modified for clasping in males) (Klass et al., 2002, 2003a). Typical for a poly-neopteran, few traits clearly unite the group with any other order.

Shortly after the description of Mantophasmatodea, Tilgner (2002) highlighted that the group might represent a derived lineage of Caelifera, perhaps near the Proscopiidae. This hypothesis was based on the observation that the cryp-topleuron, diagnostic for Orthoptera, has been secondarily lost in Proscopiidae and that the apparently five-segmented tarsi of Mantophasmatodea are, in fact, synsclerotic (i.e., united to form a single, compound structure) and resemble some "trimerous" Caelifera. Tilgner believed that there was little to truly differentiate Mantophasmatodea from such Orthoptera. The hypognathous head, which is similar to that of many Orthoptera, lends credence to this observation. However, as Klass (2002) indicated, derivation from within Orthoptera is unlikely owing to the absence of defining orthopteran characters, such as the development of a crypto-pleuron and jumping hind legs. Furthermore, the synscle-rotic conditions in Orthoptera and Mantophasmatodea are not homologous. In Orthoptera, the basal three tarsomeres are entirely fused to form a single subsegment of the podite. Conversely, in Mantophasmatodea the three basal segments are still differentiated by distinct, dorsal grooves (Klass et al., 2002, 2003a; Klass, 2002). The possibility, although very unlikely, does exist that Mantophasmatodea may prove to be derived Orthoptera, with several secondary reversals to primitive traits, just as Phasmatodea or Titanoptera may similarly prove to be derived from the Caelifera. None of these seem likely, and it is far more likely that Mantophasmatodea are the living sister group to Grylloblattodea.

The gonoplac is short in Mantophasmatodea but is sclero-tized and more developed than the second valvulae, as in Orthopterida, and this structure possibly acts as the functional ovipositor. This would further support a placement of


7.55. The closely related and relict orders Grylloblattodea (above: Grylloblatta washoa) and Mantophasma-todea (below: Karoophasma bieolouwensis). The former is today a Northern Hemisphere group, while the latter lives in southern Africa. To the same scale, above: length 10.5 mm.

7.56. A male Lobophasma redelinghuysensis (Austrophasmatidae), of the recently described order Mantophasmatodea, from the Western Cape Province's fynbos in South Africa. Photo: M. D. Picker.

Mantophasmatodea near Orthopterida. Mantophasmatodea possess a vomer-like process on the apex of the tenth sternum and therefore somewhat resemble Phasmatodea. However, Klass et al. (2002, 2003a) dismissed this feature as homologous with the vomer of Phasmatodea owing to the articulation of this sclerite along its posterior margin to the tenth sternum, versus the anterior margin in Phasmatodea. Alternatively, it could be interpreted that this inversion is merely a unique alteration of the trait in Mantophasmatodea. Despite the superficial similarity of many Mantophasmatodea to Timema, the former lack a micropylar plate in eggs, although they do possess a circular ridge reminiscent of an operculum (Klass et al., 2002), and the diets of Timema and mantophasmatodeans are completely different. At present there is no justification for a placement of Man-tophasmatodea within Orthopterida, particularly not near Phasmatodea.

Interestingly, like the coelacanth, this group was known as a fossil before living species were known. The mid-Eocene Baltic amber genus Raptophasma was identified for years as an enigmatic insect perhaps allied to walking sticks (e.g., Arillo et al., 1997), but it was not formally described until recently (Zompro, 2001). A second Baltic amber genus, Adicophasma, was identified and described as being more closely allied to the modern species than to Raptophasma because it had stout spines on the legs and body typical of some Recent species (Engel and Grimaldi, 2004b) (Figure 7.57). These Tertiary fossils possessed some of the distinctive traits of the order, such as the peculiar track of the epistomal sulcus and the very large, setose arolium, as well as the fused basal tarsomeres. However, these fossils lack the reduction of the compound eyes seen in living Mantophasmatodea.

Biogeographically, a restriction of Mantophasmatodea to sub-Saharan Africa (Klass et al., 2002, 2003a; Picker et al., 2002) is a tantalizing gondwanan juxtaposition to the Gryl-loblattodea, itself confined to Laurasia. If a Mantophasmatodea + Grylloblattodea relationship is ever conclusively demonstrated, then this would be highly significant and sim-

7.57. Mantophasmatodeans today are restricted to sub-Saharan Africa; at least as recently as the Eocene they were more widespread, as shown by Adicophasma spinosa in Baltic amber. AMNH; length 4.1 mm; from Engel and Grimaldi (2004b).

7.59. Phylloblattid "roachoid" from the Late Triassic of New South Wales, Australia. AMF38257; longest length 62 mm.

7.58. Phylloblatta gallica (Phylloblattidae), from the Late Carboniferous of Commentry, France. Paleozoic insects closely resembling modern roaches were diverse and abundant; they were not true roaches, however, but rather stem-group dictyopterans, or "roachoids." NHM In. 7296; body length 40 mm.

7.59. Phylloblattid "roachoid" from the Late Triassic of New South Wales, Australia. AMF38257; longest length 62 mm.

7.58. Phylloblatta gallica (Phylloblattidae), from the Late Carboniferous of Commentry, France. Paleozoic insects closely resembling modern roaches were diverse and abundant; they were not true roaches, however, but rather stem-group dictyopterans, or "roachoids." NHM In. 7296; body length 40 mm.

ilar to the apparent Laurasian-Gondwanan split in the Ple-coptera. Under such an hypothesis the loss of wings and ocelli would be shared features that evolved in an immediate common ancestor of both orders, and Mesozoic "Grylloblat-todea" fossils could be stem groups to both lineages, presumably with these traits appearing sometime in the Cretaceous or earliest Tertiary.

Few insect lineages have species as disparate as those in the Dictyoptera, comprising the predatory mantises, sapr-ophagous roaches, and the highly social, cellulose-feeding termites. A close relationship of these orders would seem implausible were it not for distinctive structures in the male and female reproductive systems, the proventriculus, and evidence from DNA sequences, as well as several relict, transitional species. Roaches are commonly believed to be ancient insects evolving since the Carboniferous, though in fact fossils of modern families are no older than Cretaceous - an age on a par with the other two orders.

Dictyoptera, in fact, is probably relatively recent, extending to the Jurassic, but for which there is currently very little evidence.

A popular belief in Paleozoic roaches (e.g., Guthrie and Tindal, 1968) is understandable because abundant Carboniferous fossils possessed many of the features of modern roaches, including the tegminous forewings and large, shield-like pronotum (Figures 7.58, 7.59). However, Paleozoic "roachoids" differed from modern roaches in several key respects, most significantly by possession of a large external ovipositor - a very primitive trait appearing before insects even evolved flight. The common ancestor of the lineage that includes the modern families of roaches, termites, and man-tises had a highly reduced ovipositor, as all species have today. This ancestor probably derived from one group of the Paleozoic roachoids, perhaps sometime in the Jurassic (Grimaldi, 1997b). Names have been proposed to distinguish these groups: Order Blattaria for the modern families of roaches; Dictyoptera for the orders Blattaria, Isoptera, and Mantodea and Paleozoic roachoids; and Blattodea or Blat-toptera for the paraphyletic assemblage of Paleozoic roachoids (Hennig, 1981; Grimaldi, 1997b).

+1 0

Post a comment