Okanagan Highlands taxa as stem-group Phasmatodea

A number of fossil insects previously thought by some authors to belong to the Phasmatodea cannot be assigned to the order (Tilgner Reference Tilgner2000; and see recent review by Wang et al. Reference Wang, Béthoux, Bradler, Jacques, Cui and Ren2014). Morphology of the terminalia, however, was decisive in establishing a Phasmatodea affinity for those grouped together in the Susumanioidea, currently considered as stem-group Phasmatodea, which includes insects ranging from the Jurassic of Asia through the Paleocene of Alberta, Canada (e.g., Kevan and Wighton Reference Kevan and Wighton1981; Ren Reference Ren1997; Gorochov Reference Gorochov2000; Tilgner Reference Tilgner2000; Nel and Delfosse Reference Nel and Delfosse2011; Shang et al. Reference Shang, Béthoux and Ren2011; Wang et al. Reference Wang, Béthoux, Bradler, Jacques, Cui and Ren2014). Within this group, species known only by wings are associated with the taxon by concordance of their venation with members that are confidently assigned to it by genitalic characters. The Susumanioidea might, however, be paraphyletic with regard to Euphasmatodea or Euphasmatodea+Timema, and represent stem-group Phasmatodea (Tilgner Reference Tilgner2000; Bradler and Buckley Reference Bradler and Buckley2011; Shang et al. Reference Shang, Béthoux and Ren2011; Wang et al. Reference Wang, Béthoux, Bradler, Jacques, Cui and Ren2014).

The morphology of the E. hichensi forewing is consistent with that found in the majority of Susumanioidea of Gorochov (Reference Gorochov2000), and the subsequently described Renphasma Nel and Delfosse, Adjacivena Shang et al., and Cretophasmomima Kuzmina (see figures in Rice Reference Rice1969; Birket-Smith Reference Birket-Smith1981; Kevan and Wighton Reference Kevan and Wighton1981; Kuzmina Reference Kuzmina1985; Gorochov Reference Gorochov1988, Reference Gorochov2000; Nel and Delfosse Reference Nel and Delfosse2011; Shang et al. Reference Shang, Béthoux and Ren2011; Wang et al. Reference Wang, Béthoux, Bradler, Jacques, Cui and Ren2014). The vein RP has two branches, as in all Susumanioidea except Phasmomimoides Sharov and Adjacivena, which bear three; the RP of Promastacoides was interpreted by Kevan and Wighton (Reference Kevan and Wighton1981) as having five branches, although, as they note, the region where these putatively originate is damaged and obscured, and their identities and relationships are somewhat conjectural. Much of the rest of the wing venation in the Susumanioidea appears rather conservative, except for the branching of CuA+CuPaα and its relationship with CuPaβ (which is distinctive in E. hichensi). The size difference between the holotype and paratype is consistent with sexual dimorphism.

Placement of the McAbee specimens is based on other characters, as they lack well-preserved wing venation. The Euphasmatodea is characterised by derived character states such as the significantly shortened tegmina in flighted forms, basally curved fore femora forming a notch to accommodate the head in daytime catalepsy position, a notably elongated mesothorax and metathorax, and a shortened primary ovipositor (Wedmann et al. Reference Wedmann, Bradler and Rust2007; Bradler Reference Bradler2009). The contrary character states found in species A of forewings fully developed (1), fore femora straight (2), mesothorax and metathorax not notably elongate (3), and the primary ovipositor valves (gonapophyses) elongate (4) are individually rare within Euphasmatodea (Bradler Reference Bradler2009) where they most probably represent derived characters states, and are never found in combination. These traits are most likely plesiomorphic in species A, implying a closer affinity with fossil taxa that share them: a variety of susumanioid Chinese Mesozoic stem-Phasmatodea and the early Eocene Gallophasma longipalpus Nel et al. (Phasmatodea: Gallophasmatidae) (Nel et al. Reference Nel, Delfosse, Robillard and Petrulevičius2010) from French amber, an insect that might not belong to the order at all (Tilgner Reference Tilgner2000; Bradler and Buckley Reference Bradler and Buckley2011). Although Timema shares character states (2) and (3) with the McAbee insects, it is clearly separated from them the by (1) and (4).

Species A and G. longipalpus share character states (1)–(4), and are both ∼53 million years old – a time when forests covered high latitude tectonic land connections between Europe and North America, with associated large-scale biotic interchange including insects (Archibald et al. Reference Archibald, Johnson, Mathewes and Greenwood2011b and references therein). In G. longipalpus, the maxillary palp is extremely extended, about two and a half times head length. The preserved palp of the species A holotype is not unambiguously determined as either labial or maxillary, although the latter appears more plausible, as it is always longer in known Phasmatodea. It is about 1.4 times the head length; while not as extremely extended as in G. longipalpus, this is unusually long. It is possible, but not likely, that the section identified as the portion of the palp distad the bend (Fig. 2A–2C) is in fact a part of the tarsus from the left foreleg that has been disarticulated and shifted to this position postmortem – however, even the remainder is long relative to head length.

Palp morphology is the sole character state that favours a gallophasmatid affinity for species A. Genitalic evidence may be more informative. A further characteristic of the Euphasmatodea is that the usually internal ovipositor is concealed by an operculum, an enlarged subgenital plate formed by abdominal sternum 8 (character state 5) (Wedmann et al. Reference Wedmann, Bradler and Rust2007; Bradler Reference Bradler2009). This is also – most plausibly, we believe – present in species A (Fig. 3), as it is in Susumanioidea. An operculum is clearly absent in Gallophasma, which is preserved in full detail within amber.

Aethephasma megista Ren, Hagiphasma paradoxa Ren, and Orephasma eumorpha Ren (Phasmatodea: Hagiphasmatidae), from the Early Cretaceous Yixian Formation of Liaoning, China (age: Chang et al. Reference Chang, Zhang, Renne and Fang2009) also bear character states (1)–(5). Renphasma sinica Nel and Delfosse (Phasmatodea: Susumaniidae), also from the Yixian Formation, possesses character state (1), and apparently (2) and (3), as evident in Nel and Delfosse’s photograph (2011: figure 1). It is only known from a male specimen. Adjacivena rasnitsyni Shang et al. (Phasmatodea: Susumaniidae) of the Middle Jurassic Jiulongshan Formation, Daohugou, China bears a strong superficial resemblance to species A based on character states (1), (4) and possession of an operculum (5); the foreleg and thorax are not known. Details of wing venation and other character states diagnostic of hagiphasmatid and susumaniid species and genera cannot be compared with species A because of preservation. Comparison at the genus and species level with susumanioids known from body fossils is also problematic: they appear very similar at a higher taxonomic level based on their preserved character states, but, those possibly useful to separate them at the genus and species levels are not available for comparison. For these reasons, although this insect is otherwise rather complete, we leave the species in open nomenclature, awaiting more detail in future fossils.

Distinguishing this and the other new McAbee insect species by age is also not possible. Kevan and Wighton (Reference Kevan and Wighton1981) gave the age of the Paskapoo Formation exposures where the Alberta susumanioid specimens were recovered as late Paleocene, Torrejonian–Tiffanian, i.e., about 60 Ma old, an estimation recently supported by Lerbekmo et al. (Reference Lerbekmo, Heaman, Baadsgaard, Muehlenbachs, Evans and Sweet2008). This would place these sediments roughly seven million years older than McAbee. Grimaldi and Engel (Reference Grimaldi and Engel2005) suggested that the average insect species might persist some 3–10 million years, and Hörnschemeyer et al. (Reference Hörnschemeyer, Wedmann and Poinar2010) list individual cases where extant insect species are indistinguishable from insects as old as the Eocene, including one from the ~53 million-year-old Okanagan Highlands locality at Driftwood Canyon, north of McAbee. Particularly given the size variation assumed for the Republic species, and the damaged and indistinct nature of the McAbee wings and the damaged nature of the Alberta wings, species A (and the other McAbee specimens discussed here) cannot be confidently separated from the Alberta species – nor from the Late Cretaceous species of Canada, Europe, and Siberia (see list of Gorochov Reference Gorochov2000). Therefore, we decline to assign this (or the other McAbee specimens) to a species or genus level taxon. New specimens with more clearly preserved wings are needed.

Species B shares character states (1), (2), and (3) with species A; and ovipositor morphology (4), (5), is not determinable due to preservation. The antennae are much longer than in species A as preserved, and the flagellomeres are robust and long, distinctly different. We strongly suspect that it belongs to a different species than species A.

In F-1392/1393, which we treat as stem-Phasmatodea incertae sedis, character states (1)–(3) agree with species A and B and (4) and (7) are not determinable due to preservation. Antennal morphology is similar to that of PB-3825/F-1099, indicating that it might belong to species B, or be closely related.

Stem-group Phasmatodea in the Eocene

Molecular clock analysis provides a minimum age for the origin and radiation of Euphasmatodea as Eocene, around Okanagan Highlands time (Buckley et al. Reference Buckley, Attanayake and Bradler2009), however, fossils supporting this supposition are extremely rare (Tilgner Reference Tilgner2000; Wedmann et al. Reference Wedmann, Bradler and Rust2007). If this does represent the true time of origin and diversification of Euphasmatodea, it would be consistent with the general pattern of Paleogene diversification of modern phytophagous insect groups during the expansion to dominance of angiosperm forests that began in the Cretaceous (Crane et al. Reference Crane, Friis and Pedersen1995), which was at least partially interrupted by the K-Pg extinction event (e.g., Labandeira et al. Reference Labandeira, Johnson and Wilf2002; Wilf and Johnson Reference Wilf and Johnson2004; Wilf et al. Reference Wilf, Labandeira, Johnson and Ellis2006; Iglesias et al. Reference Iglesias, Wilf, Johnson, Zamuner, Cúneo and Matheos2007; Wappler et al. Reference Wappler, Currano, Wilf, Rust and Labandeira2009), and underwent subsequent recovery and diversification by Okanagan Highlands times (e.g., Grimaldi and Engel Reference Grimaldi and Engel2005; Archibald et al. Reference Archibald, Mathewes and Greenwood2013, Reference Archibald, Kehlmaier and Mathewes2014; Labandeira Reference Labandeira2013 and references therein). Other significant insect groups such as ants and bees apparently originated in the Mesozoic and maintained a small diversity and ecological community presence until a radiation in the early Paleogene (Wilson and Hölldobler Reference Wilson and Hölldobler2005; Cardinal and Danforth Reference Cardinal and Danforth2013).

The more ancient lineages of stem-group stick insects bear morphology indicating the possibility of coevolution with gymnosperms before the establishment of angiosperm-dominated forests, perhaps associated with ginkgos specifically (Wang et al. Reference Wang, Béthoux, Bradler, Jacques, Cui and Ren2014). Ginkgo Linnaeus (Ginkgoaceae) was present in the Paleocene deposits of Alberta, Canada where Susumanioidea is known (Crane and Stockey Reference Crane and Stockey1985), is well represented at McAbee and Republic (e.g., Greenwood et al. Reference Greenwood, Archibald, Mathewes and Moss2005; Moss et al. Reference Moss, Greenwood and Archibald2005), and is relictual today, with wild populations restricted to a region of the Dalou Mountains of southwest China (Tang et al. Reference Tang, Yang, Ohsawa, Yi, Momohara and Su2012) (although a host change is also possible).

The expanded presence of stick insects largely resembling Mesozoic taxa into the early Eocene comes as a surprise, indicating a transitional phase in which they may have persisted during the radiation of crown Phasmatodea. These ancient stick insects are not seen after the early Eocene Okanagan Highlands species, which were replaced by modern forms in a world of changing forest communities, climates, and plant–insect interactions.