First fossil species of family Hyidae (Arachnida: Pseudoscorpiones) confirms 99 million years of ecological stasis in a Gondwanan lineage

Systematics

Order Pseudoscorpiones de Geer, 1778

Suborder Iocheirata Harvey, 1992

Infraorder Hemictenata Balzan, 1892

Superfamily Neobisioidea Chamberlin, 1930

Family Hyidae Chamberlin, 1930

Genus Hya Chamberlin, 1930

Hya fynni sp. nov.

Figures 2–7, 8A

ZooBank: urn:lsid:zoobank.org:act:01185A07-5675-40AB-A3DC-27E66924D8DF.

MorphoSource: doi: 10.17602/M2/M589333

Hya sp.: Benavides et al. (2019): Table 3; Harvey et al. (2023): 5.

Type Material: Holotype, adult female (CNUB-PSE-MA2016010), Burmese amber.

Locality & Horizon: Hukawng Valley (26°20′N 96°36′E), Kachin, Myanmar; lowermost Cenomanian (ca. 99 Ma), Late Cretaceous.

Preservation: The specimen is preserved in a nearly translucent amber piece. Although the specimen is well preserved, some features are not visible dorsally including, the right legs, base of the left pedipalp and anterior portion of the carapace (eyes and chelicerae). Only the left chela is present which, due to the cut of the amber, cannot be seen directly dorsally, ventrally, or laterally, but only from the dorsolateral or ventrolateral views. Chelicerae are visible in ventral view (Fig. 4B), and the chelicerae, eyes and obscured pedipalpal regions are revealed by the synchrotron scan (Fig. 5).

Etymology: This specific epithet, ‘fynni’ is a patronym honoring the first author’s son, Fynn Röschmann, who was born during the time the description was prepared.

Diagnosis: Hya fynni sp. nov. can be separated from Hya minuta (Tullgren, 1905) and Hya chamberlini Harvey, 1993 by the more proximal position of trichobothrium est on the fixed finger (Figs. 6, 8). In H. minuta and H. chamberlini, trichobothrium est is situated more distally on the fixed finger, and is only ~40 µm proximal from trichobothrium it (Fig. 8B). This contrasts with H. fynni sp. nov. in which trichobothrium est is situated more proximally on the fixed finger, and is ~70 µm proximal from trichobothrium it (Fig. 8A). H. fynni is also larger (chela length with pedicel, ♀: 0.78) than H. minuta (chela length with pedicel, ♀: 0.68–0.86) and H. chamberlini (chela length with pedicel, ♀: 0.56–0.67).

Description: The following description is based on the holotype, the only known specimen. The specimen is definitively adult given that the complete number of trichobothria are present and likely female.

Color (in amber): Uniform brown.

Chelicerae (Fig. 7A): Only visible in ventral view. Two setae on retrolateral surface of chelal hand and two on retrolateral surface of movable finger; galea unbranched, long and acute; teeth barely visible; rallum not visible.

Pedipalps (Figs. 6, 8A): Long and stout; chelal hand 1.45× longer than broad, chelal finger 1.52× longer than hand. Chelal hand and fixed finger with eight trichobothria: et, it and est situated distally on fixed finger; ist, isb, esb and eb positioned proximally on fixed finger; ib located on dorsal surface of chelal hand. Four trichobothria on movable finger: t and st distally, sb and b proximally. Fixed finger with 14 chelal teeth, heterodentate, closely spaced proximally, but widely spaced distally; chelal teeth on movable finger present. Stout setae not visible on pedipalpal femur. Venom apparatus not clearly identifiable but venom teeth visible in both fingers.

Prosoma (Fig. 4): Carapace rectangular, 2.43× longer than broad. Setae, eyes and genitalia not visible.

Opisthosoma (Fig. 4): Tergites and sternites uniform without medial suture. Tergal chaetotaxy: 4: 4: 4: 4: 6: 7: 7: 7: 8: 6 (not clearly visible).

Legs (Figs. 7C–7F): Femora I and II with basi-dorsal mound and two small setae; three slit sensilla visible on left femur II. Arolium shorter than claws.

Dimensions: Body length 1.25; chelicera 0.19/0.1; carapace 0.56/0.23; opisthosoma 0.69/0.42. Pedipalps: trochanter 0.14/0.08, femur 0.39/0.08, patella 0.29/0.1, chela length (without pedicel) 0.75, chela length (with pedicel) 0.78/0.21, hand length 0.32, movable finger length 0.45. Left leg I: trochanter 0.06/0.07, femur 0.14/0.04, patella 0.08/0.05, tibia 0.15/0.03, metatarsus 0.11/0.03, tarsus 0.17/0.02; left leg II: trochanter 0.12/0.09, femur 0.16/0.06, patella 0.08/0.05, tibia 0.15/0.05, metatarsus 0.1/0.04, tarsus 0.19/0.03; left leg III: trochanter 0.1/0.1, femur-patella 0.26/0.06–0.07, tibia 0.2/0.05, metatarsus 0.13/.0.4, tarsus 0.23/0.03; left leg IV: trochanter 0.09/0.06, femur-patella 0.33/0.09–0.15, tibia 0.16/0.07, metatarsus 0.1/0.03, tarsus 0.26/0.03.

Other Inclusions: The specimen is surrounded by plant residues, which obscure parts of the body.

Key to species in the genus Hya

1. Trichobothrium est on fixed finger situated ~70 µm proximal to it…….Hya fynni sp. nov.

2. Trichobothrium est on fixed finger situated ~40 µm proximal to it……………………….2

3. Chela (with pedicel) 0.605–0.80 (♂), 0.68–0.85 (♀) mm in length; 4–5 setae anterior to ib………………………………………………………………Hya minuta (Tullgren, 1905) Chela (with pedicel) 0.48–0.615 (♂), 0.56–0.67 mm (♀) in length; 1 seta anterior to ib (often absent)…………………………………………….….Hya chamberlini Harvey, 1993

Systematic position of H. fynni

The pseudoscorpion family Hyidae was first erected by Chamberlin (1930) based on a single species, Hya heterodentata Chamberlin, 1930 from Luzon Island in the Philippines. Tullgren (1905) originally described Ideobisium minutum Tullgren, 1905 from Java, Indonesia in the family Obisiidae Sundevall, 1833 (=Neobisiidae), but this species was transferred to Hya by Beier (1974), who described a second hyid genus, Indohya from southwestern (Kerala) India. Harvey (1993) synonymized H. heterodentata with H. minuta, thereby making H. minuta the type species of Hyidae; described H. chamberlini from Sri Lanka; placed Hya in its own subfamily Hyinae Harvey, 1993; and described five new species in Indohya and Hyella Harvey, 1993 from northwestern Australia, placing both genera in the subfamily Indohyinae Harvey, 1993. The first phylogenetic analysis of the family, based on 57 morphological characters scored for 14 hyid species, consistently recovered Hyidae and Hya as monophyletic (Harvey & Volschenk, 2007), but found Indohya to be rendered paraphyletic by Hya, and Hyella nested within Indohya. Based on this analysis, Hyella was designated a junior synonym of Indohya, and five new Indohya species were described, including the first family record for Madagascar. Harvey et al. (2023) performed a phylogenetic analysis for all Indohya species from Australia using both morphological and molecular (COI, 28S rDNA and 18S rDNA) data, and added an additional 27 new species, bringing the total number of extant species in the family to 41. The molecular analysis also found that two populations (Brunei and Singapore) identified as H. minuta had large barcode differences, suggesting the presence of cryptic species (Harvey et al., 2023).

The new fossil species can confidently be assigned to the family Hyidae based on the presence of trichobothrium ib on the dorsal margin of the left chelal hand (Fig. 6), and the presence of slit sensilla and two setae on the basi-dorsal mound of the femora of left leg II (Figs 7C, 7D). On the posterior surface of the left pedipalpal femur, a few stout setae are somewhat visible, however, it is impossible to determine with confidence if these represent the same setae diagnostic for the family. The genitalia are not visible and therefore sexing H. fynni is difficult, however, the specimen is large (1.25 mm) and is likely an adult female. In ascribing the fossil to the genus Hya, attention is paid to the chelal teeth, venom apparatus and small setae anterior to trichobothrium ib. As is characteristic of Hya, the venom apparatus is present in both chelal fingers; the chelal teeth are heterodentate and widely spaced; and two small setae are positioned anteriorly to trichobothrium ib. This contrasts with species of Indohya, in which the venom apparatus is present only in the movable finger; the chelal teeth are homodentate and closely spaced; and setae anterior to ib are absent. Although morphologically similar to extant Hya species, H. fynni can be readily separated from H. minuta and H. chamberlini, based on the position of trichobothrium est, which is more proximal in H. fynni (70 µm from it), than in H. minuta and H. chamberlini (40 µm from it) (Fig. 8). H. fynni is also the largest Hya species (total body length, ♀: 1.25; chela length with pedicel, ♀: 0.78), followed by H. minuta (chela length with pedicel, ♀: 0.68–0.86), and H. chamberlini (chela length with pedicel, ♀: 0.56–0.67).

H. fynni is the first described fossil species of the family Hyidae and the third described Burmese fossil of the superfamily Neobisioidea (Fig. 2). Moreover, H. fynni represents the second Cretaceous fossil record for an extant pseudoscorpion genus. The oldest fossil record for an extant pseudoscorpion family is A. henderickxi in the family Feaellidae, which dates to the Upper Triassic (Upper Carinan–Lower Norian: ~227 Ma) (Kolesnikov et al., 2022). Hya fynni now joins Amblyolpium burmiticum (also from Burmese amber) (Cockerell, 1920; Judson, 2009) in being the oldest fossil records for extant pseudoscorpion genera. In their dated molecular phylogeny of Pseudoscorpiones, Benavides et al. (2019) proposed that with the exception of Chthonioidea, which originated in the Paleozoic, all other Pseudoscorpion families were established in the Mesozoic. Within Iocheirata (pseudoscorpions with venom glands), the Neobisioidea families, including Hyidae, were inferred to have originated in the Triassic–Jurassic, whereas those in Panctenata are presumed to have originated somewhat later in the Jurassic–Cretaceous (Benavides et al., 2019). Until now, only eight families were confirmed through fossils to exist by the Cretaceous: Cheiridiidae, Chthoniidae, Feaellidae, Garypinidae and Ideoroncidae from Burmese amber; Cheliferidae from Archingeay amber; and one undescribed species each of Pseudogarypidae Chamberlin, 1923 and Chernetidae Menge, 1855 from the mid-Cretaceous Rhenish Massif and Canadian amber, respectively (Judson, 2009; Ross, 2019, 2023; Geißler et al., 2022). Hya fynni sp. nov. confirms that both Hyidae and its extant genus Hya existed by the Cretaceous (∼99 Ma). With the exception of its slightly larger size and the position of trichobothrium est on the pedipalpal chela, this extinct species appears morphologically identical to its extant relatives, H. minuta and H. chamberlini—an extreme case of morphological stasis over a 99-million-year period. These results also echo previous studies on fossil pseudoscorpions, which have found similar cases of morphological stasis (Geißler et al., 2022; Johnson et al., 2023; Stanczak et al., 2023).

Late rifting of the Burma Terrane?

The BT rifted from Gondwana, drifted northwards and collided with Eurasia (Fig. 1A) (Heine, Müller & Gaina, 2004; Seton et al., 2012; Metcalfe, 2011, 2013; Zhang et al., 2017; Westerweel et al., 2019). The timing of these events, and the position of the BT in relation to other landblocks during its journey, however, is still debated. Various hypotheses have been proposed based on different types of data, including palaeomagnetic, and plant and animal fossils. An “Early Devonian Rifting” hypothesis suggests that the BT rifted from Gondwana in the Early Devonian, and travelled with other continental blocks (South China, West Sumatra, Indochina and East Malaya) as a microcontinent known as “Cathaysialand” in the Permian, opening up the Palaeo-Tethys, until finally reaching Eurasia in the Late Triassic/Early Jurassic (Metcalfe, 2011, 2013). This hypothesis is supported by the fossil fauna, which is said to resemble that of Gondwana in the Devonian, but which already showed differences in the Carboniferous (Metcalfe, 2011, 2013). Alternatively, the “Late Jurassic Rifting” hypothesis suggests that the BT separated from Gondwana in the Late Jurassic and collided with Sundaland either in the Late Cretaceous (Heine, Müller & Gaina, 2004; Seton et al., 2012; Zhang et al., 2017), or even as late as the Eocene or Oligocene (Westerweel et al., 2019). Support for a later separation comes from angiosperms (Poinar, 2018), wasps (Jouault, Perrichot & Nel, 2021), short-tailed whip-scorpions (De Francesco Magnussen et al., 2022), spiders (Wood & Wunderlich, 2023) and ticks (Chitimia-Dobler et al., 2023) preserved in Burmese amber, which rule out a Devonian separation. Instead, they suggest that the BT could not have separated from Gondwana before the Mesozoic (Poinar, 2018; Jouault, Perrichot & Nel, 2021; De Francesco Magnussen et al., 2022; Chitimia-Dobler et al., 2023; Wood & Wunderlich, 2023).

The present-day distribution of Hyidae reflects that of former Gondwanan lineages, and it was already proposed that this family originated on Gondwana and was shaped by Mesozoic vicariance (Harvey, 1993, 1996a, 1996b; Harvey & Volschenk, 2007). Within Indohya, all species are found on landmasses (Madagascar, southwestern India, western Australia) that were once adjacent to each other, and phylogenetic analyses suggest that Indohya was distributed across these contiguous regions in the Early Jurassic, and diversified during Gondwanan breakup by the Late Jurassic–Cretaceous (Harvey, 1996b; Harvey & Volschenk, 2007). Most Indohya species are also considered to be short-range endemics (Harvey et al., 2023), thereby making dispersal over long distances unlikely and providing support for vicariant speciation (Harvey & Volschenk, 2007), but within the genus Hya, the vagility of species varies. H. chamberlini is restricted to Sri Lanka, a proposed former refugium during Deccan Trapp volcanism (e.g., Conti et al., 2002; Loria & Prendini, 2020). This contrasts with H. minuta, the most widespread species in Hyidae, which is found across Southeast Asia in Peninsular Malaysia, and the islands of Sumatra, Java, Borneo, Sulawesi, Luzon and Leyte, but does not reach the Moluccas Islands to the east (Fig. 3). However, Harvey et al. (2023) reported large differences in the barcoding gene COI in populations of H. minuta from Brunei and Singapore, suggesting previously undetected cryptic speciation. The discovery of H. fynni in Burmese amber not only expands the paleo-distribution of Hyidae to another landmass (i.e., the BT), but also confirms that Hya was present on the BT by the Cenomanian (99 Ma) and had diversified by this time. Considering the proposed Late Triassic origin (~210 Ma) for Hyidae, the “Late Jurassic Rifting” hypothesis seems more plausible (Fig. 1A) to explain the current distribution of Hya, and this hypothesis was also preferred in previous studies on Burmese amber pseudoscorpions, including the family Ideoroncidae (Geißler et al., 2022) and tribe Chthoniini (Chthoniinae) (Wriedt et al., 2021). Alternatively, the “Early Devonian Rifting” hypothesis was favored to explain the presence of Burmese ambers fossils in the tribe Tyrannochthoniini (Chthoniinae) based on previously published divergence dates (Benavides et al., 2019; Johnson et al., 2022, 2023). However, in the absence of biogeographical analyses, these remain speculations, and future research should focus on the systematic placement of Burmese fossil pseudoscorpions within dated phylogenies.

Ecological stasis

The Cretaceous was one of the hottest periods, and fossil evidence generally supports a warm climate and tropical/sub-tropical forest with some freshwater habitats on the BT during this time (Grimaldi, Engel & Nascimbene, 2002; Poinar, 2018; Xing et al., 2018). Among arachnids preserved in Burmese amber, all of their modern counterparts include species that can be found today in tropical rainforests (Selden & Ren, 2017). Although amber is generally biased towards preserving corticolous (bark-dwelling) arthropods, many probable humicolous (litter-dwelling) species have also been documented (e.g., Dunlop & de Oliveira Bernardi, 2014; Müller et al., 2020; De Francesco Magnussen et al., 2022). Based on the ecology of extant relatives, seven Burmese amber pseudoscorpion species including, the chthoniids B. muelleri, B. kachinae, P. burmiticus and W. plausus, the feaellid P. peetersae, and the ideoroncids P. gracilis and P. compactus, were likely humicolous, whereas the cheiridiid P. judsoni was possibly corticolous like many of its Recent relatives (Geißler et al., 2022; Johnson et al., 2023). The ecology of the garypinids, E. acutum and A. burmiticum, is less certain given that extant relatives are found in more variable microhabitats that include both tree bark and leaf litter.

Extant species of Hyidae are found in equatorial or low-latitudinal (<30°N/S) regions. Hya is exclusively found in tropical rainforests, whereas Indoyhya inhabits both monsoon rainforest and gullies/caves in the semi-arid Pilbara region of Western Australia. Within Indohya, species are classified as humicolous or hypogean (cave-dwelling), with most hypogean species exhibiting troglomorphic features, such as elongated appendages and eye loss (Harvey et al., 2023). Hya minuta and H. chamberlini are both humicolous species, and have been recorded in the soil or under leaf litter across their distribution (Harvey & Volschenk, 2007). Hya fynni is nearly indistinguishable from these modern species, and lacks troglomorphic characters. Given all available evidence, the most parsimonious situation would assume that H. fynni, like all extant Hya species, was living in humicolous environments in a tropical rainforest on the BT. This also implies that the ecology of Hya has not changed in the last 99 million years, and marks an exceptional case not only of morphological stasis but also ecological niche stasis of an arthropod lineage over an extensive period. This also contrasts with data available for some other arachnid lineages, such as the araneoid spider fauna (Wunderlich, 2008; Magalhaes et al., 2020) that has seen major changes since the Cretaceous, and even some other niche-conserved arachnids such as Ricinulei (Botero-Trujillo et al., 2022).