The Green River Formation

Much is known about the physical setting and geologic evolution of the Eocene Green River Formation, attributable largely to considerable economic interest in exploration of oil-shale layers within this spatiotemporally extensive complex of lake deposits in the western United States [31, 35]. However, the biota of the Green River Formation is less well understood [36], especially the insect fauna [37, 38]. The history of fossil insect studies of Green River strata has had a distinguished legacy extending to the pioneering work of S.H. Scudder in Wyoming during the 1890’s [39], and was soon followed by numerous contributions by T.D.A. Cockerell from the early 1900’s to the late 1920’s [40]. Summaries of the insect fauna have been made by Wilson [41], Grande [42, 43], and Graham [44]; however, these studies were regionally biased towards northern deposits of the Greater Green River Basin and Fossil Basin in southwestern Wyoming. More recently, several studies have focused on the insect fauna of the Piceance Basin in northwestern Colorado [45–52], and these reports provide a broad introduction toward understanding the insects along the southern margin of the basin, adjacent the Colorado River.

The Eocene Green River lake system developed primarily from exposure of erosion-resistant strata during multiple episodes of the Laramide Uplift, allowing widespread and prolonged deposition of organic-rich lacustrine mudstone [53] and associated finely laminated carbonate beds. These lithologies are exposed in various intermontane basins in northeastern Utah, southern Wyoming, and northwestern Colorado. The geological age of the major portion of the Green River Formation localities ranges from very early Eocene to mid middle Eocene on the basis of radiometric dating as well as mammalian biostratigraphy [31]. However, Lake Uinta was the longest lived of the Eocene lakes, spanning more than 17 million years, from late Paleocene to early late Eocene in age [31, 42]. The insect-yielding part of this sequence from the east-central Piceance Basin became substantially infilled with sediments, followed by freshwater flooding as Lake Uinta breached the Douglas Creek Arch [31].

There are three members of the Green River Formation within the Piceance Basin. The Parachute Creek Member represents the youngest member (ca. 47.8 Myr), characterized by considerable development of oil-shale sediments [31, 54]. To our knowledge, the Parachute Creek Member is the source of the insect and plant fossils for all the Piceance Basin localities, including those discussed herein.

The Green River paleoclimate has been determined by taxon-independent leaf physiognomy data using variables such as the size and shape of dicot leaves and by taxon-dependent analyses of nearest-living-relative for assessing the climatic associations of plant micro- and megafossils [55, 56]. Mean annual temperature is estimated as mid-mesothermal, ca. 16°C from leaf margin analysis, summarized in Wilf [56]. The presence of frost-intolerant taxa such as palms indicate mild winters [56].

Phylogenetic Analysis

The relationships and the position of the newly described fossil were assessed on the basis of phylogenetic analyses. The terminal taxa and characters used in the analysis were taken from Guilbert [30] and Schuh et al. [1], with the addition of the newly described fossil. The resulting matrix contained 24 taxa and 50 characters from the adult morphology (Table 1; S1 Table); eleven of these were multistate characters. All characters were coded as unordered (Fitch parsimony) [57] and equally weighted. In all analyses Xylastodoris was used as an outgroup; it belongs to the miroid family Thaumastocoridae.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Character matrix for extant and fossil Tingidae.

Coding of characters see S1 Table.

https://doi.org/10.1371/journal.pone.0133330.t001

The parsimony analysis was conducted with the branch and bound algorithm of PAUP* 4.0b10 [58]. A strict consensus tree was calculated from the resulting most parsimonious trees. To test node stability a bootstrap analysis was done with the branch and bound algorithm and 1000 bootstrap replicates.

For a second methodologically independent approach not using parsimony as a criterion, we analysed the data with a probabilistic Bayesian algorithm using MrBayes 3.2.4 x64 [59, 60]. For this analysis, the datatype "standard" was used, as implemented for morphological data in MrBayes. Two analyses consisting of 2 million generations and two runs with four chains each were produced. One analysis was done with fixed equal rates for changes of all characters, and another analysis used gamma variable rates. The resulting topologies from both analyses were identical. WINCLADA [61] was used to evaluate the character distributions on the trees.

Nomenclatural Acts

The electronic edition of this article conforms to the requirements of the amended International Code of Zoological Nomenclature, and hence the new names contained herein are available under that Code from the electronic edition of this article. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix “http://zoobank.org/”. The LSID for this publication is: urn:lsid:zoobank.org:pub: A0F81E39-57EA-425E-B895-C0AED910339B. The electronic edition of this work was published in a journal with an ISSN, and has been archived and is available from the following digital repositories: PubMed Central, LOCKSS.

Type Species

Gyaclavator kohlsi sp. nov. (by monotypy).

Diagnosis (based on male)

Macropterous. Body oblong, dorsal surface areolate. Head slightly produced in front of the compound eyes, bucculae not extended in front of the head. Antenna with segment I longer than II, II being shortest, III longest, IV strongly petaliform (autapomorphy of the genus). Pronotum with three longitudinal carinae of equal length; surface areolate; posterior margin rectilinear; comparatively short. Scutellum large, triangular. Hemelytra divided into stenocostal, costal, subcostal, discoidal and sutural areas by longitudinal elevated veins; surface covered with areolae, which are larger on costal area; transverse veins on subcostal and discoidal areas; clavus triangular.

Comments

Although the overall preservation of the specimens is spectacular, many characters used in the study of Tingidae cannot be observed. However, sufficient detail can be extracted to provide a robust description and an assessment of the relationships of the new genus and species with phylogenetic analyses. The main autapomorphic feature of Gyaclavator gen. nov. is the specialized and unique antenna with a fourth antennal article that is distinctly petaliform and which justifies the erection of a new genus.

Etymology

The new genus name is a combination from the Latin: Gyas, a name for a mythical giant, and clavator, meaning “club.” This binomial is a reference to the large size of the distiflagellomere. The gender is masculine.

Gyaclavator kohlsi Wappler, Guilbert, Wedmann et Labandeira sp. Nov

(Figs 2 and 3)

urn:lsid:zoobank.org:act:D199D0AF-0A80-46FF-BAEB-1241B1079F0C

Holotype

USNM 578190 (only one slab, no counterpart), deposited in the Department of Paleobiology, National Museum of Natural History, Washington, D.C., U.S.A. A well-preserved adult, ventro-dorsally compressed antennae well exposed (Fig 2); original field numbers shown in parentheses; David Kohls, collector.

Paratypes

USNM 582493, (only one slab, no counterpart); Green River Formation, Parachute Member, uppermost lower Eocene (latest Ypresian); Old Mountain Site, USNM locality 40189, Garfield County, Colorado; USNM 572502 (part and counterpart), Denson Site, USNM locality 41679; USNM 578171 (only one slab, no counterpart), Denson Site, USNM locality 42053. David Kohls, collector.

Etymology

The specific epithet derives from the surname of David Kohls, the collector of the type series, in recognition of his contribution to paleontology by the donation of this specimen to the National Museum of Natural History (USNM), Washington, D.C., U.S.A.

Type Horizon and Age

Green River Formation, Parachute Creek Member. This is equivalent to the (latest) Ypresian Stage of the (latest) early Eocene.

Type Locality

Denson Site, USNM locality 42053, Piceance Basin, Garfield County, Colorado, U.S.A.

Diagnosis

As for the genus (see above).

Description of holotype USNM 578190 (Fig 2)

Male. Macropterous. Body oblong, three times longer than wide; body length from head to tip of abdomen 4.6 mm, length from head to tip of hemelytra 5.5 mm; width of thorax 1.5 mm, width of abdomen 1.4 mm.

Head short with large eyes of nearly round outline, bucculae not extending forward at apex of clypeus, rostrum long, almost reaching abdomen; cephalic spines not evident, as specimen ventrally preserved; antennae as long as body length from head to abdomen, with four antennomeres; first segment stout, almost three times longer than second; second segment shortest, third segment longest, about five times longer than first; fourth segment stout, petaliform, much wider than others, about twice the length of the first, apparently with longitudinal ridge or divided in two adjacent structures, one longer, the other wider.

Pronotum longer than wide, collar narrow, with three apparently ridge-like carinae, narrow paranotum visible on right side, of small round areolae, narrow and/or reflexed onto the pronotum; sulcus present but metasternal laminae not seen; scutellum well developed, half the length of hemelytra. Rostrum almost reaching anterior border of abdomen. Femora rather stout, wider and shorter than tibiae, tarsi not seen.

Hemelytra of macropterous form; much longer than abdomen, slightly and typically widened at anterior part, round at posterior part; areolate, areolae generally small, round and of same size in all areas; divided in the usual areas. Stenocostal area narrow, with one row of areolae on posterior half; costal area narrow, wider than stenocostal area, widened in front and three to four areolae deep, with one row of areolae at the middle, with two rows of areolae directed backwards. (It is uncertain whether the hypocostal area obscures parts of the costal area; if so, the costal area is raised or slightly reflexed, unusual in Cantacaderinae.) Subcostal area much wider, five areolae wide at widest part, divided into several small areas by transverse veins; discoidal area as wide as subcostal area, five to six areolae wide at widest point, divided in four smaller areas by transverse veins; sutural area wide on posterior half; all main veins slightly raised.

Paratype USNM 578171 (Fig 3A, 3C and 3D)

Male. Morphology very similar to holotype with characteristic antennae. Body lenght from head till end of abdomen 4.4 mm, length from head till tip of hemelytra 5.0 mm.

Paratype USNM 572502 (Fig 3B, 3E and 3F)

Male. Morphology very similar to holotype with characteristic antennae. Body lenght from head till end of abdomen 4.5 mm, length from head till apex of hemelytra 5.4 mm. Rostrum long, almost reaching abdomen. Sutural area of macropterous hemelytra overlapping and projecting over apex of abdomen.

Description of paratype USNM 582493 (Fig 3G and 3H)

Male. Morphology very similar to holotype with characteristic antennae. Anterior border of head not preserved, also no antennae preserved. Body length till end of abdomen ca. 4.3 mm, length head to hemelytra 5 mm. Areolae of the hemelytra equal in size, transparent. Anterior pronotal margin imperceptible. Pronotum with three parallel carinae. Stenocostal area uniserate. Costal area 0.3 mm wide, with four rows of rounded areolae. Subcostal area slightly wider than costal area, with six rows of cells in widest part. Discoidal area triangular in shape. Clavi not covered by pronotum.

Phylogenetic Aspects

For the fossil record of Tingidae, many fossil species have ambiguous phylogenetic relationships and exhibit a mosaic of characters typical for different taxa. For example Archeopopovia and Tingicader exhibit a combination of characters of the Phatnomini and Tingini. Other fossil taxa such as Burmacader multivenosus exhibit a combination of characters of the Cantacaderinae and Tinginae. A similar situation exists for Gyaclavator kohlsi. Unfortunately, diagnostic characters of the cephalic tubercles and the pregenital segments are not available for a better resolution of Gyaclavator’s taxonomic status, but Gyaclavator is distinguished from Tinginae + Cantacaderinae by the relatively large and triangular scutellum (Fig 3G), a character shared with members of the Vianaidinae. By contrast, the scutellum is greatly reduced in Tinginae + Cantacaderinae. The bucculae do not extend in front of the head (Fig 2C), which could argue for placement of Gyaclavator within Tinginae. Conversely, the ventral and dorsal presence of a uniseriate stenocostal area (Fig 2H–2I) and occurrence of transverse veins on the subcostal and discoidal areas argue for placement of Gyaclavator in the Cantacaderinae. The phylogenetic analyses of this complex mixture of characters favor slightly its close relationship with Cantacaderinae.

Biogeographical Aspects

The Tingidae, particularly the Phatnomini, were very diverse in the Palearctic region during the middle Eocene. Fossils of Phatnomini are known from Northeastern Europe to Mongolia and from the Lower Cretaceous to the upper Miocene. The earliest fossil record belonging to the Phatnomini is Sinaldocader ponomarenkoi Golub and Popov, 2008 [20], from Baissa in Transbaikalia of Russia. The age of this specimen is mid Lower Cretaceous, likely Hauterivian to Barremian, ranging from ca. 125–135 Myr. Cantacaderinae fossils, or more precisely, Cantacaderini (because there are no fossil records yet for Carldrakeanini and Ceratocaderini), are more recent lineages and have been found from the European Eocene to the Miocene of northern Europe. Similarly, fossils of Tingini are relatively recent, occurring from the Eocene to Miocene, but are more biogeographically widespread, and are present from China via Europe to North America. Today, Tingini are worldwide distributed. Both the Cantacaderini and Phatnomini currently are distributed principally in tropical and subtropical regions of the southern hemisphere, while their fossils to date are mainly known from the Palearctic region. The third monophyletic group of Tingidae, the Vianaidinae, is represented by at least one fossil species in North America, from New Jersey amber of late Turonian age (ca. 93 Myr) [1, 64]. The Vianaidinae are basal to all other Tingidae but their extant distribution is restricted to South America. This could argue for a New World origin, buttressed by the fossil Vianaidinae from North America. The recent discovery of a fossil close to Vianaidinae in Burmese amber from the earliest Cenomanian at ca. 100 Myr [12] contradicts a New World origin for the clade, and instead suggests already a wide distribution during the Cretaceous. This scenario is also supported by the ancient species Golmonia pater Popov, 1989 [11], from the Lower Cretaceous of Mongolia (tribe Golmoniini), which has been considered as basal to Tingidae.

In conclusion, the origin and age of Tingidae are unclear. The oldest known fossil of Tingidae is a member of the Phatnomini, which is older than any fossils of the basal Vianaidinae. Known fossils of the Cantacaderinae are much younger than those of the Phatnomini. These data indicate that the Tingidae could be significantly older than the mid Early Cretaceous and more fossils are needed to shed light on their geographic origin.

If the new species Gyaclavator kohlsi belongs to the Candacaderinae, it would be the first fossil record of this group for the Nearctic region. Presently, no Cantacaderinae are known from the Nearctic region, but they occur in the Neotropics, and one fossil has been recorded from Dominican amber [14]. Consequently, their former occurrence in North America is not unexpected, although such a presence cannot contribute new insights to the questions concerning the biogeographic origin of the Tingidae.

Behavioral Aspects

The dilated fourth antennal segment in Gyaclavator kohlsi n. sp. is probably a rare fossil example of a visual display structure that likely functioned for mate attraction or male-male-competition. This conspicuous feature previously was unknown for the Tingidae. Modified antennae also are found in the genera Copium and Paracopium, whereby the antennal segments can be strongly thickened; otherwise, these two genera differ minimally from Tinginae with only slightly enlarged antennae [e.g., 65] and they have nothing in common with the modifications found in Gyaclavator. The discovery of antennae with a communicative function in tingids is timely, as there has been considerable recent interest in the roles of sexual selection and reproductive competition in the evolution of insect reproductive behavior [66–69]. Adorned body parts such as antennae are mobile features that can be used as intraspecies semaphores for signalling with or without attractants, or with tactile stimulation [70]. Recently described flabellate antennae of a Lower Cretaceous hymenopteran indicate the antiquity of long-range female attractants [71]. In addition, it has been acknowledged that male reproductive success often is limited by access to females [72]; thus, sexual competition can select for secondary sexual characters in males. Consequently, visual displays such as in birds of paradise [73], peacocks [74] and claw waving displays of fiddler crabs [75], are conspicuous signalling structures that function in courtship and mate attraction, and probably are involved in male–male competition.

Speculation about which factors favor the dilation of antennal segments prominently include habitat effects, such as the availability of feeding resources, community composition or other environmental factors. It would be premature to attribute enlargement of the terminal antennal segment in Gyaclavator kohlsi to a particular factor, as sensillae are not preserved on the fossil specimens. However, examples are known of the dilation of particular antennal segments in coreid heteropteran adults, such as a the foliate or dialated third antennal article in the leaffooted bug Chariesterus, and the expanded second and third segments in Chondrocera [76]. In both examples, antennae are used for display and their conspicuousness through aerial waving may be partially associated with the presence of male neighbors and may function in male–male competition. Such displays also may act as a long-range female attractant (see S1 Movie). From these data, we infer that G. kohlsi represents an extreme specialization of antennal structure that is interpreted as convergent evolution of structural traits in unrelated lineages. Furthermore, these data show an unexpected morphological disparity in lace-bug antennal structure, and shed new light on mechanisms of sexual selection involving intra- and intersex reproductive competition during the early Eocene.