New records and new species of springtails ( Collembola : Entomobryidae , Paronellidae ) from lava tubes of the Galápagos Islands ( Ecuador )

The Collembola fauna of the Galápagos Islands is relatively unexplored with only thirty-five reported species. Entomobryoidea, the most diverse superfamily of Collembola, is underrepresented, with only five species reported from the Galápagos. Here we present the findings of the first survey of Collembola from Galápagos lava tube caves, providing a significant update to the total number of entomobryoid Collembola species reported from the Galápagos Islands. Collections made during a March 2014 expedition to study lava tubes of the islands yielded new records for seven species of Entomobryoidea, including four genera not previously reported from the Galápagos Islands: Coecobrya, Entomobrya, Heteromurus, and Salina. As a result, three new species (Entomobrya darwini Katz, Soto-Adames & Taylor, sp. n., Pseudosinella vulcana Katz, Soto-Adames & Taylor, sp. n., and Pseudosinella stewartpecki Katz, Soto-Adames & Taylor, sp. n.) are described and new diagnoses are provided for Heteromurus (Heteromurtrella) nitens Yosii, 1964, Lepidocyrtus nigrosetosus Folsom, 1927 and Pseudosinella intermixta (Folsom, 1924). Lepidocyrtus leleupi Jacquemart, 1976 is synonymized with L. nigrosetosus. An updated checklist of all species within the superfamily Entomobryoidea reported from the Galápagos Islands is provided. Subterranean Biology 17: 77–120 (2016) doi: 10.3897/subtbiol.17.7660 http://subtbiol.pensoft.net Copyright Aron D. Katz et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. RESEARCH ARTICLE Subterranean Biology Published by The International Society for Subterranean Biology A peer-reviewed open-access journal

New records and new species of springtails (Collembola: Entomobryidae, Paronellidae) from lava tubes of the Galápagos Islands (Ecuador)

Introduction
Like the Hawaiian and Canary islands, major islands of the Galápagos archipelago are comprised of shield volcanoes (Fig. 1A).The Galápagos volcanoes are on the Nazca oceanic plate over the Galápagos hotspot, far from plate boundaries (Holden and Dietz 1972, Hey 1977, Toulkeridis 2011).The main Galápagos Islands are located east of the N-S-trending East Pacific Rise and south of the E-W-trending Galápagos Spreading Center and some 1000 km west of the Ecuadorian mainland.Associated with the shield volcanoes are hundreds of relatively small, cinder cones, ash cones, and spatter cones.These islands range in age from 0.0032 to 4 million years, with still active volcanoes on the younger islands (White et al. 1993, Geist et al. 2014, Global Volcanism Program 2015), with even older islands now represented only by seamounts.These geodynamics and the ever-changing volcanic development of the Galápagos Islands has given rise to conditions under which the unique endemic life on these islands has evolved (Darwin 1859).Evidence now suggests that for at least some animal taxa, colonization of the Galápagos archipelago likely took place before many of the present islands had emerged from the ocean (Christie et al. 1992, Torres-Carvajal et al. 2014, Husemann et al. 2015).
The composition of the fauna of the Galápagos Islands is shaped by the volcanic history of the islands, as well as the remoteness of the archipelago from the mainland.This is even more true for the subterranean fauna, comprised of "wrecks of ancient life" (Darwin 1859), living in the relatively stable, cool, humid conditions (Ashmole et al. 1992) of lava tubes on the slopes of the various shield volcanoes that form these islands.These lava tube caves (e.g., Fig. 1B) form from heated basalt flowing down the slopes of volcanoes, forming preferential flow paths that eventually drain, creating the caves.Various smaller cracks and other subterranean voids occur both in the relatively smooth pāhoehoe lava flows and in the more rugged 'a'ā lava (Harris and Rowland 2015), providing innumerable places in which invertebrates may live (Howarth 1991, Peck 2001, Stone et al. 2004, Howarth et al. 2007, Toulkeridis 2011).
Diversity and relationships among subterranean animals, shaped by vicariance and dispersal are a common theme for island lava tube faunas of Hawaii (Howarth 1991, Juan et al. 2010), the Canary Islands (Oromí et al. 1991, Juan et al. 2000, Naranjo Morales and Abreu 2015), Rapa Nui (Easter Island) (Wynne et al. 2014), and the Galápagos Islands (Peck 1990).However, the inaccessibility of much of the Galápagos archipelago, due to a lack of roads and extremely rugged terrain, have restricted biological inventories of subterranean ecosystems in comparison to the Hawaiian (more than 50 terrestrial troglobites, Howarth 1991) and Canary archipelagos (more than 160 species of invertebrates unique to the underground environment, Naranjo Morales and Abreu 2015).Most recent work on Galápagos has been carried out, or summarized, by Stewart B. Peck (Peck and Kukalova-Peck 1986, Peck and Shear 1987a, 1987b, Peck (Santa Cruz Island, Galápagos Islands, Ecuador), which is typical of the morphologies of the lava tube caves sampled.Maps of many of the other caves sampled during this study are available in Addison (2011) and Toulkeridis and Addison (2015).1990, 2001, Peck and Finston 1993) who reports more than 55 eyeless and reduced eye arthropods, many of these from caves or cave entrances (Peck 1990(Peck , 2001)).
Earlier analyses by Peck (1990) and Taylor et al. (2012) suggested that much remains to be learned regarding the subterranean fauna of the Galápagos Islands.These analyses helped build the justification for a multidisciplinary expedition, including the biological collections reported here, studying lava tube caves in the Galápagos Islands in March 2014.This expedition was conducted in association with the 16th International Symposium of Vulcanospeleology (Puerto Ayarro, Santa Cruz, Galápagos Islands, Ecuador) (Toulkeridis and Addison 2015), and built upon earlier field studies led by two of the authors (TT, AA).Here we make a contribution regarding the entomobryoid springtails (Collembola) of the Galápagos Islands as part of the ongoing studies of lava tube caves of the islands by our research team.
Thirty-five species of Collembola have been reported from the Galápagos Islands (Peck 2001); 18 in the order Poduromorpha (Stach 1932, Gama 1986, Najt et al. 1991) and 17 in the order Entomobryomorpha, 12 of which are in the family Isotomidae (Thibaud et al. 1994).The superfamily Entomobryoidea is underrepresented, with only 5 species formally reported from the Galápagos (Folsom 1924, Jacquemart 1976).The ostensible lack of diversity is most certainly due to the limited number of collections where specimens were subsequently identified and published.All 35 reported species were identified and described in six publications from material obtained from just four collections: the Williams Galápagos Expedition 1923-1925(Folsom 1924); the Norwegian Zoological Expedition to the Galápagos Islands 1925, conducted by Alf Wollebaek (Stach 1932); the Belgian Zoological Mission to the Galápagos Islands and Ecuador led by N. and J. Leleup, 1964-1965(Jacquemart 1976, Gama 1986, Najt et al. 1991, Thibaud et al. 1994); and collections made by Guy Coppois 1974-1976(Najt et al. 1991, Thibaud et al. 1994).
Entomobryoidea is the most diverse superfamily of Collembola, representing more than one fourth of all described species worldwide (Bellinger et al. 2015).Species in the genera Pseudosinella, Sinella, Coecobrya, and Trogolaphysa, represent a large proportion of the springtail diversity found in New World caves, but the springtail fauna of Galápagos lava tubes has not been documented and these genera (except Pseudosinella) have not been reported from the islands.The purpose of this study is to describe and document all species in the superfamily Entomobryoidea collected during the recent 2014 bioinventories of Galápagos lava tubes.Here we provide descriptions for three new species endemic to the Galápagos, three new diagnoses, and new records for four genera and seven species.We also include an updated checklist of all species in the superfamily Entomobryoidea reported from the Galápagos Islands.

Methods
Geologic setting on Santa Cruz and Isabela islands.The Santa Cruz shield volcano is subdivided into two main units (Bow 1979), an older Platform Unit with an age of 1.3-1.1 Ma and a younger unit represented by lavas of the Shield Series with ages as young as 30-20 ka (Bow 1979, White et al. 1993, Geist et al. 2014).Sierra Negra on Isabela Island is a young, volcanically active shield volcano with various eruptive centers and lava fields that have been divided into five distinctive age groups, all being younger than 6000 years (Reynolds et al. 1995).Ten eruptions have occurred on Sierra Negra in the last 200 years, with the most recent eruption in October 2005 (Geist et al. 2008).
Specimen collection and preparation.Collections for this study were made at lava tube caves on Santa Cruz and Isabela islands, Galápagos (Fig. 1A) from 8 through 21 March 2014.Bioinventories were conducted in eleven lava tube caves (Table 1): ten on Santa Cruz Island and one on the Sierra Negra volcano on Isabela Island (Galápagos Islands, Ecuador).Caves were selected on the basis of opportunity, distribution across Santa Cruz Island, and coordination with other ongoing research (mapping, geology, soil science).For caves on Santa Cruz Island, handheld meters were used at most sites to collect temperature, humidity and light data in surface, entrance, twilight and dark zone habitats.Individual specimens were collected by hand from ceiling, walls, and floor in terrestrial or drip pool (Fig. 2A) microhabitats throughout the lava tubes, using a paintbrush moistened in alcohol, or with an aspirator.Where accumulated leaf litter deposits were encountered, litter samples were collected into plastic bags and then extracted for 2-4 days using cloth Berlese funnels heated with 25 watt light bulbs.All material was preserved in 70% ethanol.
Individuals sampled were sorted under a dissecting microscope to morphospecies and photographed to record color pattern and body shape prior to slide mounting.All slide-mounted specimens were cleared with Nesbitt's solution and mounted with Hoyer's medium (Mari Mutt 1979) in preparation for light microscopy.Sex was determined by the observation of genital plate morphology.Specimens where genital plate morphology was obscured are listed in material examined sections without a sex determination.Illustrations were hand-drawn under a camera lucida, scanned, with final drawings created using Adobe Illustrator.Chaetotaxy nomenclature.Descriptions of dorsal body chaetotaxy follow the nomenclature established by Szeptycki (1979); the dorsal head chaetotaxy follows Jordana and Baquero (2005) and Soto-Adames (2008).Traditional dorsal chaetotaxy nomenclature for Pseudosinella established by Gisin (1967) is referred to in some descriptions for comparison and simplicity.See Soto-Adames (2010a) for comments and comparisons between nomenclature systems.All descriptions of labial chaetotaxy follow the nomenclature of Chen and Christiansen (1993).

Results
Invertebrate samples were collected from all eleven caves from all caves zones.Specimens of Entomobryoidea were found from eight caves (Santa Cruz Island: Cueva Aguirre, Cueva Cañón, Cueva Cascajo, Cueva Chato 1, Cueva Chato 2, Cueva Gallardo, La Llegada; Isabella Island: La Cueva de Sucre).The lava tube caves varied from dry and largely barren of organic deposits to moist and containing organic deposits, soils, and water pools.The sites examined during this study are on the slopes of two shield volcanoes -Santa Cruz Island (elevation: 964 m) and Sierra Negra on Isabela Island (elevation: 1124 m).These caves are all in lava flows of sufficient age to have become largely covered with vegetation and to have significant settling and collapse resulting in various surface connections and tube segmentation (e.g., Figs 1B, 2B).Measured light levels ranged from 91,900 lux on the surface to 0 lux in the dark zone, air temperatures ranged from 37.3 °C in direct sunlight and 32.8 °C in the shade on the surface to 21.9 °C in the dark zone, soil temperatures ranged from 31.7 °C on the surface to 20.8 °C in the dark zone, and relative humidity ranged from 37.3% on the surface to 93.7% in dark zone, deep-cave habitats (Fig. 3).Most springtails were taken in Berlese leaf  litter and moss samples in the entrance zone of caves (e.g., Figs 2B, 4), with specimens individually hand-collected in deep cave habitats, such as drip pools (Fig. 2A) or isolated woody debris.
Head.Apical pin seta on Ant.IV present (Fig. 5B).Dorsal chaetotaxy of head as in Figure 5C   Remarks.The chaetotaxy of the single individual collected (an adult female), is identical to that described for H. nitens from the Kingdom of Tonga by Yosii (1964).The individual from Galápagos differs from Yosii's (1964) description only in claw morphology, our specimen lacks the inner unpaired ungual tooth (Fig. 6E).This is the first record of the genus Heteromurus from the Galápagos Islands.
The description above is intended to supplement the original description by Yosii (1964) with detailed descriptions and illustrations of dorsal head and body chaetotaxy.Most scales and setae had fallen off our specimen during transport making it difficult to differentiate elements (i.e., scales, sensilla, microsetae).Therefore, the mc on Th.II-Abd.IV are not illustrated.Some sensilla may have also been omitted from the illustrations if not visible on the specimen.This species' peculiar distribution (reported from Tonga and Galápagos) indicates its range extends across the Pacific Ocean and may occur on other Pacific Islands.The difference in number of inner ungual teeth and vast geographical distance separating the populations of Tonga and Galápagos hint at a species level differentiation.However, in view of the general morphological similarity and a lack of sufficient material, we have chosen not to erect a new name for this single individual.

Description. Size. Up to 2.6 mm
Color pattern.Background color white or light orange, with dark purple pigment limited to Ant.III-IV, latero-posterior margin of Th.II, lateral margin of Th.III and meso-and metathoracic coxae.Some individuals have and additional irregular purple spot on the base of the furcula, others have no pigment at all, except for the antennae.The black or dark brown look of living specimens is produced by the thick covering of black or dark brown scales (Fig. 7A).
Appendicular scales distribution.Dorsally on Ant.I, femur on middle and hind legs and ventral face of furcula.Scales absent from Ant. II-IV, fore legs, ventral tube and dorsal face of manubrium.
Legs.Trochanteral organ with up to 41 setae.Tenent hair spatulate on all legs.Unguis with 3-4 inner teeth, distal unpaired tooth sometimes absent; all teeth small; proximal unpaired tooth well separated from basal paired teeth.Fore and middle unguiculi relatively short, strongly truncate, with well marked inner tooth and weakly serrate posterior edge; hind unguiculus usually lanceolate or weakly truncate, rarely strongly truncate, inner tooth absent or weakly delineated; hind unguiculus always clearly longer (surpassing inner proximal unpaired ungual tooth) than fore and middle unguiculi (barely reaching inner proximal unpaired tooth).
Ventral tube.All faces covered by many finely ciliate setae; posterior face with 1+1 smooth setae on distal margin in addition to ciliate setae (Fig. 11C) Furcula.Manubrium and dens without smooth setae.Basal tubercle of dens apically rounded, somewhat asymmetrical.Mucro with apical tooth slightly longer than basal tooth.Mucronal spine with minute basal denticles.
Remarks.This species is characterized by the enlarged, rounded mesothoracic hood, absence of dorsal head Mc posterior to A2, smooth labial setae, absence of seta m3e on Abd.II, four inner Mc on Abd.IV, heteromorphic unguiculi (truncate on fore and middle legs, lanceolate or weakly truncate on hind leg), and a rounded but somewhat asymmetric tubercle on the dens.
As pointed out by Bernard et al. (2015) the large bodied members of the L. nigrosetosus species group (L.nigrosetosus, L. immaculatus Folsom, 1932, L. leleupi Jacquemart, 1976and L. geayides Denis, 1931) are very similar and difficult to distinguish.Bernard et al. (2015) suggested that L. leleupi, originally described from the island of Santa Cruz in the Galápagos, was likely to be a junior synonym of L. nigrosetosus.Our collections of Lepidocyrtus from Santa Cruz fit the color pattern description of L. leleupi and at first the specimens were identified as that species, but evaluation of other morphological characters showed the specimens to fit the range of variation reported for L. nigrosetosus (Mari Mutt 1986).To confirm these observations we studied the type series of L. leleupi deposited in the Royal Belgian Institute of Natural Sciences in Brussels, Belgium.
The type series of L. leleupi comprises the holotype, 11 paratypes and 4 additional specimens mounted on slides.The slide labeled holotype holds two individuals.Jacquemart (1976) did not specify which of the two individuals was the holotype, hence here we designate an individual as the holotype (arrow in Figure 12).The holotype and other members of the type series L. leleupi lack mc m3e on Abd.II (Fig. 9B) and show the same range of morphological variation seen in specimens of L. nigrosetosus from Puerto Rico.For this reason we consider L. leleupi a junior synonym of L. nigrosetosus.This makes L. nigrosetosus the most geographically widespread member of the genus in the Neotropical region (Soto-Adames and Anderson in press).Remarks.Pseudosinella intermixta, originally described by Folsom (1924) from material collected on Baltra Island (as South Seymore Island), is the only member of the genus with 3+3 eyes, an apical Ant.IV bulb, with head Mc M2, 1 Mc on Th.II, and without Mc on Th.III.Folsom's species is very similar to P. stewartpecki sp.n. described below, but can be separated by the presence in P. inermixta of head Mc M2 and an enlarged unpaired ungual tooth that is larger than both inner paired teeth.Folsom's (1924) description of P. intermixta is relatively incomplete, lacking details for many important characters, thus it was necessary to revisit the type material to determine if our specimens (described below as P. stewartpecki sp.n.) differed from P. intermixta.The syntypes, stored in ethanol with labels (Fig. 13C, D), were in extremely poor condition.Despite their poor condition, after slide mounting the most complete specimen (Fig. 13A) we were able to supplement Folsom's (1924) description with the few additional characters provided above.Folsom (1924) described and illustrated P. intermixta with only 2 subequal inner teeth on the unguis.However, the mounted syntype clearly has 3 inner teeth: 2 minute paired teeth and 1 large unpaired distal tooth (Fig. 13B).Due to old age and poor preservation, dorsal chaetotaxy is mostly obscured on the mounted type specimen, but the head clearly carries Mc M2.
Appendicular scales distribution.Scales present on head, body and ventral face of furcula.Antennae, legs, ventral tube and dorsal face of furcula without scales.
Remarks.Pseudosinella stewartpecki sp.n. is the only member of the genus with 3+3 eyes, an apical antennal bulb, head series M and S without Mc, with 1 Mc on Th.II, and without Mc on Th.III.This new species is most similar to P. intermixta, also described from the Galápagos Islands (Folsom 1924).However, P. stewartpecki sp.n. lacks head Mc in rows M and S, and the unpaired inner tooth of the unguis is smaller than the largest of the two inner paired teeth (Fig. 17B), whereas in P. intermixta, head Mc M2 is present and the unpaired inner tooth is substantially larger than both paired teeth (Fig. 13B).Pseudosinella intermixta was collected on Baltra Island, which can be characterized as dry, lowland (maximum elevation of 100m) habitat.Pseudosinella stewartpecki sp.n. was collected in relatively moist, upland (251 m) habitat on Santa Cruz Island.These differences, in morphology and habitat, are sufficient for the separation P. intermixta and P. stewartpecki sp.n.

Pseudosinella vulcana
Appendicular scales distribution.Scales limited to head, body and ventral face of furcula.Antennae, legs, ventral tube and dorsal face of furcula without scales.
Remarks.Pseudosinella vulcana sp.n. is the only member of the genus with 1+1 eyes and with a wing tooth on both the unguis and unguiculus.This new species is most similar to P. biunguiculata Ellis, 1967, sensu Mari Mutt (1986), but differs by having 1+1 eyes, Abd.IV supplemental seta s present (Fig. 20B), and only 1+1 paired setae on the posterior face of the collophore (Fig. 20D), where as in P. biunguiculata eyes and supplemental seta s are absent and the posterior face of the collophore has 2+2 paired setae and 1 unpaired medial seta.Pseudosinella vulcana sp.n. has 2 thickened, apically blunt metatibiotarsal setae, which was originally thought to differ from P. biunguiculata since Mari Mutt (1986) described only 1 blunt metatibiotarsal seta.However, we observed 2 metatibiotarsal setae on P. biunguiculata from Puerto Rico (Guajataca Commonwealth Forest, at end of trail #10, leaf litter, 19.v.2009, F. Soto, coll.).Pseudosinella caoi Chen, Wang & Christiansen, 2002and Pseudosinella fujiokai Yosii, 1964, sensu Christiansen and Bellinger (1992), also have blunt metatibiotarsal setae, ungual wing teeth, Head Mc A3 (R2), and lack Mc on Th.II and Th.III, but can be differentiated from Pseudosinella vulcana sp.n. by characters outlined in Table 2.
Mari Mutt (1986) described 3 morphologically distinct forms among and within populations of P. biunguiculata in Puerto Rico that differ in dorsal chaetotaxy (presence/absence of head Mc A3 (R2)) and tenent hair morphology (clavate/acuminate).He also noted differences between the Puerto Rican forms and Ellis' (1967) type specimens; primarily the absence of head Mc A2 (R1), A3 (R2), and Pa5 (Po) on the holotype.Furthermore, the original description of P. biunguiculata does not show the presence of m4i on Abd.II, which is distinctly present in the Puerto Rico populations.The high levels of morphological variation exhibited among these forms suggest that P. biunguiculata represents a species complex: small and seemingly insignificant differences in morphology have been shown to correlate with large genetic distances among populations, indicating the presence of species complexes (Porco et al. 2012, Cicconardi et al. 2013, Katz et al. 2015a).In fact, Soto-Adames (2002b) observed large genetic differences between sympatric individuals of Puerto Rican P. biunguiculata, but these differences could not be correlated with differences in morphology due to destructive DNA extraction methods.Additional investigations utilizing morphological and molecular data may clarify species-level relationships among populations of P. biunguiculata, a species with a widespread neotropical distribution.See Table 2 for a list of diagnostic characters separating the different forms and descriptions of P. biunguiculata.
Pseudosinella vulcana sp.n. was collected from entrance and from the surface of a drip pool (Fig. 2A) within the dark zone, in generally cool, moist, low light conditions (Table 3), suggesting the species may be a troglophile.
Coecobrya sp.A Fig. 21A Remarks.A single juvenile male with 1+1 eyes was collected on the surface (Table 3) at Cueva Gallardo.Four species of Coecobrya with 1+1 eyes have been described (Xu and Zhang 2015): C. boneti (Denis, 1948), C. indonesiensis (Chen & Deharveng, 1997), C. sanmingensis Xu &Zhang, 2015, andC. tukmeas Zhang, Deharveng &Chen, 2009.Our specimen is very similar to the four species listed above, but the combination of characters on the labial triangle, Abd.III, and unguiculus suggest our specimen represents an undescribed species.Unfortunately, a single juvenile is insufficient material on    Etymology.A patronym honoring the naturalist Charles R. Darwin (1809Darwin ( -1882) ) for his work on the Galápagos Islands, which helped inspire his contributions to evolutionary theory.We believe Charles Darwin would have overcome his views on Collembola had he seen the color patterns of this new species: "They [Collembola] are wingless, dull-coloured, minute insects with ugly, almost misshapen head and bodies" (Darwin 1871).Description.Body shape and color pattern.Body slightly dorso-ventrally flattened.Length up to 1.57mm (♀) and 1.23mm (♂).Males and females with no obvious difference in color pattern.Color pattern with slight variation (Fig. 22): light orange/ tan background with black or dark purple pigment forming two irregular lateral tri-angles or sometimes broken angled bands on the posterior margin of Abd.III; an additional pair of lateral angled broken bands on Abd.II; dark pigment present along lateral margins of Th.II through Abd.IV, sometimes broken along lateral margin of Abd.III; dark irregular/broken transverse band along posterior margin of Abd.III -Abd.V. Abd.IV usually with an irregular U-or "11"-shaped pattern connecting basally with band along posterior margin.Antennae usually with uniform purple pigment.Legs white, with purple patches on apical end of femora and tibia.
Remarks.Entomobrya darwini sp.n. is the only member in this genus with the combination of color pattern and chaetotaxy presented in the description above.In addition, E. darwini sp.n. has some unique diagnostic characters that, to our knowledge, have not been previously documented: the conspicuously long lateral sensilla on Th.II and Th.III (Fig. 23D, E) and the spine-like setae on the trochanteral organ are uncharacteristically enlarged, stout, and slightly truncated (Fig. 24D) This species shares a similar color pattern with Entomobrya litigiosa fasciata Denis, 1931 described from Costa Rica, but E. darwini sp.n. has two dark broken/irregular triangles or angled bands along the lateral margins of Abd.III that are always absent in E. litigiosa fasciata, in addition, the unguiculus is lanceolate in E. darwini sp.n., whereas in E. litigiosa fasciata, it is truncate.The color forms of the Nearctic species Entomobrya decemfasciata (Packard, 1873), sensu Katz et al. (2015b), also include angled bands on Abd.III that superficially resemble those exhibited by E. darwini sp.n.However, these species can easily be separated by chaetotaxy: E. decemfasciata is characterized by having an extreme abundance of dorsal Mc, where as in E. darwini sp.n., the dorsal Mc are generally reduced (Figs 23B, D-G; 24A, B).Entomobrya nicoleti (Lubbock, 1871) also shares a similar color pattern with E. darwini sp.n. that includes forms with angled lateral bands on Abd.III, but differences in dorsal chaetotaxy (e.g., E. nicoleti has a1 on Abd.III that is absent on E. darwini sp.n.) easily separates these two species.
Entomobrya darwini sp.n. was collected from both surface and entrance habitats (Table 3) at three caves.This is the first record of the genus Entomobrya from the Galápagos Islands.

Family Paronellidae
Cyphoderus cf.agnotus Börner, 1906 Fig. 21B Remarks.A single individual was collected in the entrance area of a lava tube.Our specimen has a bidentate mucro and the unguis lacks unpaired ungual distal teeth, distinguishing it from Cyphoderus galapagoensis Jacquemart 1976 while matching the description of Cyphoderus agnotus, sensu Cassagnau (1963), albeit Cassagnau's (1963) species description is relatively vague and lacks details about many important characters.
Unfortunately, the material available is insufficient for a more complete redescription.Distribution.Santa Cruz Island, Galápagos, Ecuador; widespread throughout Neotropics.

Salina sp. A Soto-Adames, 2010b
Remarks.A single female, likely a juvenile, was collected.Our specimen keys out to Salina thibaudi Soto-Adames, 2010b according to the preliminary key to American Salina (Soto-Adames 2010b).However, evaluation of additional characters listed in Soto-Adames (2010b) indicates that this is a new species, similar or identical to Salina sp.A from Panama reported in Table 1 in Soto-Adames (2010b).The material available is insufficient to make a complete description and provide an unambiguous diagnosis.Nevertheless, this is the first record of the genus Salina from the Galápagos Islands.
This species was collected from a surface habitat adjacent to the entrance of a lava tube (Table 3).

Updated checklist of Superfamily Entomobryoidea (Collembola) of the Galápagos
The following checklist includes all valid entomobryoid species previously reported from the Galápagos Islands, as well as new species records, new genus records, and the three newly described species presented in this paper.
The island records cited above from Peck (2001) were gathered from unknown sources (pers. comm. Peck December 10th, 2015) and need additional confirmation.The following species are listed in Peck and Jacquemart (2013), but are excluded from the checklist above: Cyphoderus innominatus Mills, 1938 andLepidocyrtus cf. lanuginosus (Gmelin, 1788) do not have formally published records of Galápagos distributions; five additional species names in the genera Cyphoderus, Entomobrya, Pseudosinella, and Seira listed in the Charles Darwin Foundation online checklist (Peck and Jacquemart 2013) have not been formally published.

Discussion
Seven species were identified from the collections, including three new species, raising the total number of Galápagos entomobryoids from five to twelve species.Four genera are reported from the Galápagos for the first time: Coecobrya, Entomobrya, Heteromurus, and Salina.While none of the entomobryoid species other than P. vulcana sp.n. seem to show a particular association with lava tubes, the relatively cool moist conditions found in the entrance and twilight zones of these caves (Fig. 3) provide both a refuge from harsh surface conditions for these and many other organisms, and a window into subsurface volcanic habitats which would otherwise be under-sampled by biologists.
Remarkably, the specimens studied for this paper represent the first new collections of Galápagos Collembola in almost 40 years to be identified and subsequently reported and described.In spite of numerous scientific studies of invertebrates in the Galápagos Island, the new records and new species descriptions presented in this paper and the long gap between new collections being represented in the Galápagos springtail literature suggest that our understanding of the species richness of springtails and other microarthropods in the Galápagos Islands is still incomplete.Additionally, springtail collections from the Galápagos Islands made by Heinrich Schatz (University of Innsbruck) contain dozens of species not presently recorded from the islands (Palacios-Vargas, Pers.Comm.January 2016).Thus, further biological sampling in Galápagos lava tubes and lava tube entrances, sampling of microarthropods from other microhabitats, as well as further taxonomic studies of other material from our 2014 collections, should yield more additions to the fauna of this unique archipelago.
: Row An with 7-12 Mc; anterior Mc A0 and A2 present, relatively small but with differentiated sockets, and inserted among a group of enlarged fusiform elements field normally including only Mc A0 (Fig. 7B; fig. 2 in Jacquemart 1976), all other anterior Mc absent; posterior Mc absent, element Ps5 enlarged and fusiform; postocular bothiotrix displaced laterally and inserted behind eyes E and F. Prelabral and all labral setae smooth.

Figure 13 .
Figure 13.Pseudosinella intermixta syntypes.A syntype specimen chosen for slide mount (INHS Acc.567,409) B hind claw complex C detail of preservation condition of syntypes in ethanol D original labels in vial.

Figure 23 .
Figure 23.Entomobrya darwini sp.n.A apical sense organ of Ant.III B dorsal chaetotaxy of head C labral papillae D-G dorsal chaetotaxy: D Th. II E Th. III F Abd.I G Abd. II.

Table 1 .
Hernández et al. (1992) in the Galápagos Islands (Ecuador) that were sampled for invertebrates in March 2014.Elevations based on 3m DEM data.Lengths and cave maps are from cited sources.Cueva Aguirre is not the same as Cueva de Raul Aguirre ofHernández et al. (1992).
ing of invertebrate biodiversity in the Galápagos Islands and throughout the Western Hemisphere.Type material.Holotype, ♂ on slide, Ecuador, Galápagos, Santa Cruz Island: La Llegada, leaf litter from entrance,12.iii.2014(C.Plowman, D. Butler and G.

Table 3 .
Environmental conditions associated with March 2014 collections of Entomobryoidea from lava tube sites on Santa Cruz and Isabela islands, Galápagos Islands, Ecuador.