Research Article
Research Article
A new compression fossil, Eotriadomeroides abjunctus Huber, gen. & sp. nov. (Hymenoptera, Mymaridae), in Eocene shale from the Kishenehn Formation, USA
expand article infoJohn T. Huber, Dale E. Greenwalt§
‡ Natural Resources Canada c/o Canadian National Collection of Insects, Arachnids and Nematodes, Ottawa, Canada
§ Department of Paleobiology, National Museum of Natural History MRC 121, Smithsonian Institution, Washington, United States of America
Open Access


A new fossil genus and species of fairyfly, Eotriadomeroides abjunctus Huber & Greenwalt, gen. and sp. nov. (Hymenoptera: Chalcidoidea: Mymaridae), is described and illustrated from a female preserved as a compression fossil in middle Eocene shale from the Kishenehn Formation, Montana, USA. It is compared to extant species of Neotriadomerus Huber, known only from Australia, and Triadomerus Yoshimoto, a Cretaceous amber fossil from Canada. It is suggested that these three genera, classified together in Triadomerini, likely the most ancestral lineage of Mymaridae, are evidence of the Middle or perhaps Late Jurassic origin of the family.


Chalcidoidea, Eocene compression fossil, Mymaridae


Parasitoid wasps of the family Mymaridae (Hymenoptera), almost all parasitic in eggs of other insects, are common and widespread, occurring on all continents except Antarctica, from 81°49'N (Hazen Camp, Canada) to 54°57'S (Bahía Aguirre, Tierra del Fuego, Argentina). They also occur on most islands, even those farthest from continents, e.g., the Hawaiian Islands, St. Helena, and French Polynesia, or those with harsh climates, e.g., Greenland, Iceland, Auckland Islands, Campbell Island and South Georgia; the latter three more than 50°S. Mymaridae are also one of the two best represented families of Chalcidoidea in the fossil record (the other is Baeomorphidae), represented almost entirely by inclusions in amber (Yoshimoto 1975; Poinar and Huber 2011; Engel et al. 2013). Very few Mymaridae are described from compression fossils (Huber and Greenwalt 2011). The fossil records for the family are from the Cretaceous to the Pleistocene, a duration of at least 100 my, though with a large time gap of about 40 my between the Cretaceous and Tertiary fossils records. A compression fossil specimen from the Kishenehn Formation in Montana, USA, representing a new genus and species, is described here.


Huber and Greenwalt (2011) described the methods of collecting and photographing insect compression fossils from the Kishenehn Formation. The fossil described below was collected in accordance with the United States Forest Service Special Use Permit HUN465. Greenwalt et al. (2015) described the taphonomy of the Kishenehn Formation.

Measurements, in millimeters, were taken from the photographs as accurately as possible and converted into micrometers (μm). Given that the end points of a structure were not always clear, their measurements were rounded to the nearest 5 μm and should be treated as approximate only. Length/width ratios of the antennal segments were calculated from the millimeter measurements, not from the rounded-off micrometer measurements.

Abbreviations used

fu = funicle segment, mps = multiporous plate sensilla. The specimen is deposited in:

NMNH Department of Paleobiology, National Museum of Natural History, Washington, DC, USA.


Eotriadomeroides Huber, gen. nov.

Figs 1–2, 3–5, 6, 7, 8

Type species

Eotriadomeroides abjunctus Huber, here designated.


Female. Antenna with funicle 8-segmented and clava 1-segmented (Figs 25); fore wing with venation extending almost to wing apex, with postmarginal vein as wide as marginal vein or parastigma and ~2.7× as long as parastigma + marginal + stigmal veins (Fig. 7); tarsi 5-segmented (Fig. 8); fore wing microtrichia apparently extending to base of parastigma; hind wing relatively narrow, with acute apex; ovipositor extending ventral to mesosoma almost to level of head and not exserted posterior to apex of gaster (Fig. 1). Other details are apparently the same as for Neotriadomerus Huber, morphologically the genus most similar to Eotriadomeroides.

Figures 1, 2. 

Eotriadomeroides abjunctus Huber, holotype female 1 habitus (except most legs not visible) 2 pronotum + head + partly disarticulated antennae.

Figures 3–5. 

Eotriadomeroides abjunctus Huber, holotype female 3 head + part of antennae 4 left antenna (pedicel–fu5) 5 fu8 + clava of both antennae.

Figures 6, 7. 

Eotriadomeroides abjunctus Huber, holotype female 6 mesosoma 7 wings + metasoma.

Figures 8, 9. 

Eotriadomeroides abjunctus Huber, holotype female 8 tibiae and tarsi of one? pair of legs 9 shale piece containing holotype (circled) of Eotriadomeroides abjunctus Huber.

Male. Unknown.

Derivation of genus name

From the Greek, eos, meaning early + Triadomerus (a compound word derived from Greek, tries, meaning three, and meros, meaning part, referring to the 3-segmented clava) + the suffix -oides, meaning like, resembling. Eotriadomeroides (gender masculine) is therefore an “early Triadomerus-like” genus, referring to its geological age (the Eocene) and morphological similarity to the two other, evidently related genera: Neotriadomerus (with all its species extant) and Triadomerus (with its single species extinct).


Genera of Mymaridae are usually divided formally into subgenera if females of different species within a given genus have either a 1- or 2-segmented clava, or either a 2- or 3-segmented clava, and the other morphological features are essentially identical. So far, no genus is known to have its included species with either a 1-segmented or a 3-segmented clava but none with a 2-segmented clava. Only one genus (Anaphes Haliday) possibly has its included species with a 1-, 2-, or 3-segmented clava but so far Anaphes species with 3-segmented clava have yet been described and named. Examination of the clava of Eotriadomeroides does not suggest it is 2- or 3-segmented but rather that it is clearly 1-segmented, i.e., entire (Fig. 5). For comparison, the species of Eoanaphes Huber and Eoeustochus Huber from the same formation and apparently with the same quality of preservation, are clearly 3-segmented whereas those of Gonatocerus Nees are just as clearly 1-segmented (Huber and Greenwalt 2011). If the clava of E. abjunctus were 2- or 3-segmented then it could be classified as a subgenus of Neotriadomerus, given that all other features, except relative lengths of postmarginal vein to the rest of the venation, are almost the same in both taxa. Eotriadomeroides would then key to Neotriadomerus in the key to Cretaceous genera of Mymaridae (Poinar and Huber 2011). Another possibility would be to treat E. abjunctus as a subgenus within Triadomerus Yoshimoto, described from amber from Cedar Lake, Manitoba (Yoshimoto 1975), which is only about 1000 km away from the type locality (the Kishenehn Basin, Montana) of E. abjunctus. According to McAlpine and Martin (1969) the actual source of the Cedar Lake amber is more likely to be upstream, along the Saskatoon River either near Saskatoon, Saskatchewan, or Medicine Hat, Alberta, respectively about 650 km and ~280 km from the type locality of E. abjunctus as determined from the present day configuration of the localities (essentially unchanged from 46 my years ago). Triadomerus does not have the ovipositor extending anteriorly ventral to the mesosoma and it has a relatively short postmarginal vein compared to length of stigma + marginal + parastigmal veins, so we treat E. abjunctus as belonging to a new genus, different from both Neotriadomerus and Triadomerus, both of which have a 3-segmented female clava and are known, respectively, from seven extant and one extinct species. Eotriadomeroides is best classified in Triadomerini (Huber 2017) but exact relationships among the genera still need resolution.

Eotriadomeroides abjunctus Huber, sp. nov.

Figs 1–2, 3–5, 6, 7, 8

Material examined

Holotype female (NMNH), on 18 × 14 × 0.15 cm piece of oil shale (Fig. 9), labelled “Holotype Eotriadomeroides abjunctus Huber. USNM # PAL 620738”. The circle/square scratched onto the surface of the shape indicates the holotype location. The specimen was collected in 2012 at locality #43946, Park site, Kishenehn Formation, Montana, USA.


Eotriadomeroides abjunctus is the only described species in the genus. Its diagnosis is therefore the same as for the generic description. Comparing it with species of morphologically similar genera, it differs from all the described species of Neotriadomerus (Huber 2017) as follows: clava 1-segmented (clava 3-segmented in Neotriadomerus species); postmarginal vein ~2.7× as long as parastigma + marginal vein + stigmal veins (postmarginal vein at most 0.90× as long in Neotriadomerus species); hind wing narrow and apically acute (hind wing wide and apically blunt in Neotriadomerus species). The apparent absence of a straight setal line extending from apical margin of fore wing about halfway towards the parastigma + marginal veins (Fig. 7) is an additional feature that may separate E. abjunctus from Neotriadomerus but the wing surface of E. abjunctus is not clear enough to be sure if the setal line is absent. Eotriadomeroides abjunctus differs from Triadomerus bulbosus Yoshimoto by the clava 1-segmented (3-segmented in T. bulbosus), ovipositor extending ventral to mesosoma as far as head (ovipositor not extending anteriorly ventral to mesosoma in T. bulbosus), and relatively longer postmarginal vein (relatively shorter in T. bulbosus).


Female. Color. Vertex, antenna except radicle, dorsum of body, except for scutellum, and ovipositor sheaths dark brown or almost black; face, radicle, scutellum, and mesosoma and metasoma ventrally apparently lighter brown (Fig. 1). Total body length ~2850. Head. Head length ~205, head width ~600; mid ocellus diameter ~35. Antenna (Figs 25). Three (possibly 4) mps are visible on the right clava and one on fu8 of the left antenna (Fig. 5); the mps that most likely should occur on the remaining funicle segments are not visible. Length/width measurements: range (ratios) of antennal segments: radicle? ~85/~12 (2.08), scape excluding radicle ~230/~90 (2.53), pedicel ~75/~50 (1.47), fu1 ~170/~45 (3.85), fu2 ~160/~40 (3.83), fu3 ~150/~40 (3.67), fu4 ~150/~45 (3.33), fu5 ~140/~40 (3.08), fu6 ~150/~40 (3.60), fu7 ~135/40 (4.00), fu8 ~125/~150 (2.57), clava ~325/~85 (3.76). Mesosoma. Mesosoma length ~900, metanotum with dorsellum almost certainly triangular (Fig. 6). Wings. Fore wing (Fig. 7) with microtrichia uniformly covering entire surface, apparently to base of parastigma and apparently with one row of a few microtrichia posterior to apex of submarginal vein; fore wing length/width ~1930/~560, length/width 3.50, longest marginal setae ~80; hind wing length ~ 1150, width ~45, longest marginal setae ~115, with wing apex acute. Legs. Tarsi 5-segmented, the tarsomeres becoming shorter towards apex of tarsus (legs segments mostly unrecognizable except two tibiae in part and two tarsi visible, with the end points of basal tarsomeres unclear). Metasoma. Petiole (Fig. 6) evidently short; gaster with terga apparently about equal in length. Metasoma length ~1875; ovipositor length ~2640, with sheaths extending anteriorly ventral to mesosoma to level of pronotum.

Derivation of species name

From the Latin abjunctus, meaning disunited or separated, refers both to the strongly disjunct geographic distribution of this 40 my old fossil from extant members of Neotriadomerus, the most similar looking genus, and to the fact that some of the fossil’s appendages are broken into parts (the antennae) or are separated from the body (the legs).


Amorim and Greenwalt (2020) described Synneuron (Diptera: Canthyloscelidae) from the wings of two fossil specimens, one from the Kishenehn Formation and one from the Koonwarra Fossil Bed in Australia. Given their strongly disjunct (perhaps worldwide) distribution, they suggested the Cretaceous as the minimum age for Synneuron, but, more likely, based on their phylogenetic analysis of the tribe it is classified in, they proposed the Middle Jurassic as the minimum age of the genus. Synneuron has two extant species, one Palaearctic and the other Nearctic. The case of Eotriadomeroides and Neotriadomerus may be similar. Although they are classified in different genera, they, together with the Cretaceous genus Triadomerus, are best classified in the same lineage, the Triadomerini (Huber 2017), with only the species of Neotriadomerus extant. This tribe is likely the most ancestral lineage within Mymaridae, based on its morphology, and its included genera were possibly worldwide in distribution early in the existence of Mymaridae as a recognizable taxon, just as the present day Mymaridae are worldwide. And, as with Synneuron and its relatives, the Triadomerini may also have originated as a Pangaean clade. More likely, however, Triadomerini originated more recently, in the middle Jurassic, as estimated by Peters et al. (2018). Regardless of the actual age of origin, Mymaridae are small wasps easily capable of being dispersed aerially for long distances, as evidenced by their current existence on remote islands. So, if they did originate well after the breakup of Pangaea, dispersal worldwide from one or other of those putative centres of origin, either Laurasia or, much more likely, Gondwana would certainly have been possible.

Chronologically, Eotriadomeroides (43–46 my), falls almost midway between Triadomerus (70–90 my) and Neotriadomerus (present day). Evidently, Triadomerini is an ancient lineage that occurs continuously throughout much of the geological history of Mymaridae as currently understood. A related lineage within Triadomerinae, the Aresconini, contains extant species in three genera (Huber 2017), and one extinct species in one genus, Myanmymar, from 100 my Burmese amber (Poinar and Huber 2011).One hopes that the large time gaps will eventually be filled as more fossils deposits containing Mymaridae are discovered and better evidence for the age of origin of the family will be found.

The middle Jurassic and early Cretaceous had gymnosperm-dominant environments worldwide, which changed to angiosperm-dominant environments in the later Cretaceous (Wing 2000). This change in flora, presumably accompanied by a similar change in fauna (including perhaps the hosts of Mymaridae), may partly explain why two of the genera (Eotriadomeroides and Triadomerus) of Triadomerini are extinct and only one genus (Neotriadomerus) is extant and found only in the southern Hemisphere, which is where many of the archaic taxa of extant Hymenoptera seem to occur.

The piece of shale that contained Eotriadomeroides also contained other synimpressions, as follows: 2 Aphididae and 22 Corixidae (Hemiptera), 16 Chaoboridae, 1 Culicidae and 3 other flies (Diptera), 1 Chalcididae, 1 Chalcidoidea, 1 Formicidae and 1 other wasps (Hymenoptera), 1 Thysanoptera, and 1 plant (Cupressoideae). These insects together suggest they occurred in moist habitat near water.


We thank the United States Forest Service for allowing D. Greenwalt to collect compression fossils from the Kishenehn Formation under the auspices of United States Forest Service Special Use Permit HUN465. We gratefully acknowledge the help of Jennifer Read (Canadian National Collection of Insects, Arachnids and Nematodes, Ottawa, Canada) for labelling the figures and compiling them into plates.


  • Amorim DS, Greenwalt DE (2020) Cretaceous and Eocene fossils of the rare extant genus Synneuron Lundstrom (Diptera: Canthyloscledidae): evidence of a true Pangean clade. Cladistics 36(4): 413–423.
  • Engel JT, McKellar RC, Huber JT (2013) A fossil species of the primitive mymarid genus Borneomymar (Hymenoptera: Mymaridae) in Eocene Baltic amber. Novitates Paleoentomologicae 5: 1–8.
  • Greenwalt DE, Rose TR, Siljestrom SM, Goreva YS, Constenius KN, Wingerath JG (2015) Taphonomy of the fossil insects of the middle Eocene Kishenehn Formation. Acta Palaeontologica Polonica 60(4): 931–947.
  • Huber JT (2017) Eustochomorpha Girault, Neotriadomerus, gen. n., and Proarescon, gen. n. (Hymenoptera: Mymaridae), early extant lineages in evolution of the family. Journal of Hymenoptera Research 57: 1–87.
  • Huber JT, Greenwalt D (2011) Compression fossil Mymaridae (Hymenoptera) from Kishenehn oil shales, with description of two new genera and review of Tertiary amber genera. ZooKeys 130: 473–494.
  • Peters RS, Niehuis O, Gunkel S, Bläser M, Mayer C, Podsiadlowski L, Kozlov A, Donath A, van Noort S, Liu S, Zhou X, Misof B, Heraty B, Krogmann L (2018) Transcriptome sequence-based phylogeny of chalcidoid wasps (Hymenoptera: Chalcidoidea) reveals a history of rapid radiations, convergence, and evolutionary success. Molecular Phylogeny and Evolution 120: 286–296.
  • Poinar Jr G, Huber JT (2011) A new genus of fossil Mymaridae (Hymenoptera) from Cretaceous amber and key to Cretaceous mymarid genera. ZooKeys 130: 461–472.
  • Wing SL (2000) Evolution and expansion of flowering plants. 209–232. In: Gastaldo RA, DiMichele WA (Eds) Phanerozoic Terrestrial Ecosystems. Paleontological Society, Pittsburgh, 308 pp.
login to comment