The neuropterid fauna of early Miocene Dominican and Oligocene-Miocene Mexican amber is treated. The fauna consists entirely of Megaloptera and Neuroptera while the snakeflies (Raphidioptera) are not presently known in Tertiary ambers from the New World. Fifteen species are recognized, all in living genera and subgenera: Sialidae, Sialis (Protosialis) casca, new species (Dominican); Ascalaphidae, Ululodes paleonesia, new species (Dominican), Amoea electrodominica, new species (Dominican); Chrysopidae, Chrysopa glaesaria, new species (Dominican), C. vetula, new species (Dominican), Leucochrysa (Nodita) prisca, new species (Dominican); Coniopterygidae, Coniopteryx antiquua, new species (Dominican), Spiloconis glaesaria Meinander (Dominican), S. oediloma, new species (Dominican); Hemerobiidae, Notiobiella thaumasta Oswald (Dominican), Sympherobius sp. (Dominican); Mantispidae, Dicromantispa electromexicana, new species (Mexican), D. moronei, new species (Dominican); Feroseta prisca Poinar, nomen emendatum (Dominican); Myrmeleontidae, Porrerus dominicanus Poinar and Stange (Dominican). Immature stages of Ascalaphidae (two species of Ululodes; Dominican), Chrysopidae (genus indet.; Dominican), and Myrmeleontidae (P. dominicanus?; Dominican) are reported. The neuropterid fossil record is summarized and the Dominican and Mexican amber faunas are compared to other neuropterid amber faunas. The biogeographic and paleoecological implications of the Miocene fossils are discussed. Abundance and diversity of Neuroptera in ambers appear to be related to the abundance of Sternorrhyncha, on which many neuropterans feed.
Introduction
The superorder Neuropterida comprises three distinctive orders of holometabolous insects: the Megaloptera, the Neuroptera, and the Raphidioptera. These three orders form a monophylum defined by the fusion of the third valvulae in the ovipositor, the medially divided metapostnotum, the unpaired diverticulum of the proventriculus, and the caudally bifid mediolongitudinal suture on the first abdominal tergum (Mickoleit, 1973; Achtelig, 1975; Kristensen, 1991). Although such characters are obscure, each of the major monophyletic groups within the Neuropterida has a distinctive wing venation. Venation has allowed interpretation of fossils, and in fact the abundant record of wing impressions in rocks gives the Neuropterida among the most extensive fossil record of all Holometabola. Together with the Coleoptera, their closest living relatives among the Holometabola (Kristensen, 1991, 1995; Hörnschemeyer, 1998; Whiting et al., 1997; Carpenter and Wheeler, 1999; Wheeler et al., 2001; Grimaldi and Engel, 2005), the neuropterids are an ancient offshoot of endopterygote insects with geological histories extending back to the Early Permian (Kukalová-Peck, 1991; Carpenter, 1992; Grimaldi and Engel, 2005). Unlike the beetles, however, the neuropterid orders are not very diverse and some, like the families of Raphidioptera, are perhaps in decline globally (Aspöck, 1998; Engel, 2002a). In fact, a remarkable diversity of extinct neuropterid lineages is known from the Mesozoic and, to a lesser degree, earlier deposits (e.g., Schlüter, 1986; Carpenter, 1992; Grimaldi and Engel, 2005). Owing to the combination of their ancient age, highly apomorphic relict taxa, and numerous extinct lineages, the Neuropterida is one of those groups whereby inclusion of fossil taxa into cladistic studies is critical (Gauthier et al., 1989). It will be interesting to explore the effects of such fossil taxa on our understanding of neuropterid phylogeny at all hierarchical levels. Most neuropterid fossils, however, are compression fossils that preserve only wings or very limited, structural details. Preservation in amber, however, is unsurpassed, preserving not only the external morphology in essentially lifelike detail but often internal tissues (Grimaldi et al., 1994; Grimaldi and Engel, 2005). Amber inclusions offer a greater opportunity to make meaningful comparisons between Recent and fossil taxa as well as simultaneous cladistic analyses of paleontological and neontological data (e.g., Grimaldi and Cumming, 1999; Engel, 2001).
The most diverse neuropterid fauna in amber is from the Baltic region, which has classically received the greatest attention from systematists. There are presently 33 described species in Baltic amber (table 1). This amber is middle Eocene in age (dating reviewed in Engel, 2001; Weitschat and Wichard, 2002). Baltic amber is the largest deposit of amber in the world. It has been studied for centuries and is the most intensively studied of amber Lagerstätte; however, very few modern studies have investigated inclusions of Neuropterida, and comprehensive revisionary work is needed. Even basic descriptive work of new species in Baltic amber is ongoing (e.g., Wichard, 1997; Engel, 1999a; Dobosz and Krsemiński, 2000; Wichard and Engel, 2006). Among Cretaceous ambers, the fauna of the Turonian New Jersey amber is only slightly less diverse than that of Baltic amber (table 2), particularly when accounting for the much smaller size of the deposits of the former. The only other Cretaceous amber with a significant neuropterid fauna is the Late Albian amber from Myanmar (Grimaldi et al., 2002; Engel, 2002a, 2004a, 2004b, in prep.). Other amber deposits have revealed a few other Neuropterida (table 3), principally of families well documented in the major deposits. Lastly, the Cenozoic amber deposits of North America have similarly revealed only sparse material of a few neuropterids. Previously only three species in the families Myrmeleontidae, Coniopterygidae, and Hemerobiidae have been described from Dominican amber while none has been described from Mexican amber. Herein we report on new neuropterids discovered in Dominican and Mexican amber; these include the first New World fossil of an alderfly (Sialidae) and the first green lacewings (Chrysopidae) and the first adult owlfly (Ascalaphidae) in amber. In total the fauna has now increased to 13 species distributed among seven families (table 4). The age of Dominican amber has been unfortunately confused, but it is certainly no older than Early Miocene in origin (Iturralde-Vinent and MacPhee, 1996, 1999). Mexican amber is of a similar age, originating from the Late Oligocene–Early Miocene (Langenheim, 1966).
Table 1
Named Neuropterida in Baltic Ambera
Table 2
Named Neuropterida in New Jersey Amber(from Grimaldi, 2000; Engel, 2002b: all are placed in extinct genera)
Table 3
Named Neuropterida in Lebanese, Burmese, French, Siberian, Hat Creek, “British”,a and Parisian Ambersb(all are in extinct genera)
Table 4
Neuropterida in Dominican and Mexicana Amber
With such a long geological duration of neuropteridan history (Permian to present), Tertiary fossils are of little phylogenetic consequence regarding higher taxa. The significance of Dominican and Mexican amber fossils lies in (1) the preservation of amber, allowing close species comparisons; (2) information for West Indian biogeography, particularly assessing any geographical extinctions; and (3) paleoecology, or taphonomic inferences provided by the taxa that are preserved. The Dominican and Mexican amber faunas are relatively similar to tropical neuropterid faunas in these regions today. The fauna consists entirely of Megaloptera and Neuroptera while the snakeflies (Raphidioptera) are presently unknown in Tertiary ambers of the New World. Tertiary compression fossils of Raphidioptera are known from North America and Europe (e.g., Aspöck et al., 1991; Nel, 1993; Engel, 2002a, 2003) and snakeflies are present in mid-Eocene ambers of Europe (Carpenter, 1956; Engel, 1995, 2002a; Weitschat and Wichard, 1998, 2002; Aspöck and Aspöck, 2004). However, on the whole the Cenozoic record of this order is insignificant and it is in the Mesozoic that a remarkable diversity of Raphidioptera from throughout the world is known (e.g., Aspöck, 1998; Grimaldi, 2000; Engel, 2002a, unpubl. data; Engel et al., 2006; Perrichot and Engel, in press).
Materials and Methods
All measurements were made using an ocular micrometer and should be considered approximate since the optimal angle for any given metric was not always achievable owing to the uneven surface of the amber. The acronyms AMNH and MACT are used, respectively, for the American Museum of Natural History, New York, and the Morone Amber Collection, Turin, Italy.
Systematic Paleontology
Order Megaloptera Latreille Family Sialidae Leach
The Sialidae, or alderflies, are one of two living families of Megaloptera. Sialid larvae, like those of all megalopterans, are aquatic predators, living in anything from small streams, rivers, ponds, lakes, or even phytotelmata and under stones or where dead plant material accumulates. Larvae emerge onto land to pupate in vegetation or debris. Adults are short-lived, occur along watersides, and apparently may feed on pollen and nectar, if at all. Generally little is known of adult megalopteran biology, particularly so for Sialidae.
Today the family is represented by approximately 66 living species (most in the nominate genus Sialis) that are distributed worldwide but are principally Holarctic with a few species in the New World tropics (Sialis s.l.: includes Protosialis and Nipponosialis after Whiting [1994] and New and Theischinger [1993], respectively), South Africa (Leptosialis), Australia (Austrosialis and Stenosialis), and Southeast Asia (Indosialis). In addition, 10 extinct species have been previously proposed (table 5), almost all in Tertiary deposits. The primitive Dobbertinia reticulata (Handlirsch) from the Liassic of Germany is the oldest record of the family. This species is classified in its own basal subfamily, the Dobbertiniinae.
Table 5
Named Fossil Sialidae
Herein we describe a new species of Sialis (Protosialis) in amber from the Dominican Republic, a group which is today unknown from the West Indies except for one species on the island of Cuba, that is, Sialis (Protosialis) bifasciata Hagen (Penny, 1977; Contreras-Ramos, 1999).
Diagnosis
The orange coloration of the head and pronotum and narrowed costal area with a reduced number of costal crossveins indicate this to be a species of the subgenus Protosialis. For Protosialis species the presence of only three costal crossveins is unique while the black markings on the head are similar to that seen in Sialis (Protosialis) flammata (Penny) but are rounded in the fossil (not flamelike as in S. flammata).
Description
Forewing length (preserved) 7.7 mm, (estimated: from bent forewing) 8.4 mm; total body length 7.6 mm (fig. 1). Head and pronotum orange with black markings; antennae black; head with black markings along margins of eye and extending posteriorly near to posterior border of head, black extended medially but not reaching midline, borders of markings rounded (not flamelike). Apical segments of maxillary palpi with medioapical tooth. Ocelli absent. Pronotum apparently twice as wide as long, remainder of thorax dark brown. Protibia, mesofemur, mesotibia, and metatibia black, remaining leg segments whitish; fourth tarsomere bilobed. Wing membranes fuscous; veins dark brown to black; forewing with costal area distinctly narrowed just before middle of wing; three costal crossveins, distal two veins distinctly oblique (i.e., not perpendicular to C or Sc); distal r-rs crossvein meeting R2+3 slightly distad to R2+3/R4+5 fork; posterior branch of R4/R5 separating near wing apex (i.e., distad R2/R3 branching); outer series of gradate crossveins confluent (figs. 2, 3). Abdomen slightly distended and bent ventrally near midpoint; apparently light brown; apex of abdomen curled under, with genitalia pointed anteriad.
Order Neuroptera Linnaeus Family Mantispidae Leach
The mantispids are perhaps one of the most instantly recognizable of neuropteran families and consist of approximately 400 species worldwide. Their convergent form with mantises (Mantodea) of an elongate prothorax and raptorial forelegs is remarkable. Adults tend to be diurnal and are often found on flowers, although a few lineages are nocturnal or crepuscular. Some species are brightly colored and appear to mimic social, aculeate Hymenoptera. The most basal mantispids, the Symphrasinae, have sedentary larvae that are generalist predators found in the nests of social Hymenoptera, particularly vespids, or burrows of scarab beetle larvae. Similarly, the Calomantispinae appear to be generalist predators although the larvae of at least one studied species are ambulatory. The subfamily Drepanicinae is entirely unknown biologically. The largest subfamily, the Mantispinae and the subfamily to which the fossils described below belong, are specialized, obligate parasites of spiders, and the first-instar larva either boards a female spider and enters the egg sac as it is constructed, to feed on the eggs, or it searches directly for an egg sac. Later larval instars are relatively immobile. Pupation occurs within the host egg sac. Adults are actively predaceous on a variety of other insect groups. Redborg (1998) has reviewed the available information on the remarkably specialized biology of Mantispinae. Although Willmann (1990) included the Rhachiberothinae in Mantispidae, we follow Aspöck and Mansell (1994) and Aspöck et al. (2001) by considering this group as allied to the Berothidae.
The geological history of the Mantispidae has only recently begun to come together. Previously there were nine fossil species described from deposits ranging from the Early Jurassic to the Miocene (table 6: Wedmann and Makarkin, 2007), although the exact systematic position of some of these taxa is not entirely certain. Only two mantispid species had been discovered in amber until this time, but one is likely not a member of this family. Walfera venatrix (Whalley) was discovered in amber that had washed up on the eastern shore of Britain and presumably was derived from the Baltic amber deposits of the blaue Erde. The identity of Whalfera, however, has been questioned by many, and the genus is very likely a rhachiberothid (Aspöck and Mansell, 1994; Engel, 2004b; Grimaldi and Engel, 2005; Wedmann and Makarkin, 2007). The only other amber mantispid is Feroseta priscus Poinar in Miocene Dominican amber (Poinar, 2006). Herein we report two further, definitive mantispids in Cenozoic amber from the Western Hemisphere. Spiders are common in Tertiary ambers of North America (e.g., Wunderlich, 1988; Penney, 1999), and it is therefore perhaps not surprising that their parasites would also eventually be discovered.
Table 6
Named Fossil Mantispidaea
Figs. 6–8.
Dicromantispa electromexicana, new species (CH-MP5). 6. Detail of raptorial foreleg depicting profemoral spination. 7. Dorsal aspect of abdominal apex. 8. Ventral aspect of abdominal apex.
Mantispa sp.: Engel, 2004c: 184.
Mantispa sp.: Grimaldi and Engel, 2005: 354, fig. 9.35.
Diagnosis
This species is superficially similar to Zeugomantispa minuta (Fabricius) but differs by the more elongate pronotum that apparently lacks markings or bumps and has exceedingly sparse, minute pubescence, and by the structure of the ectoprocts. It can be separated from D. moronei in Dominican amber (see below) by the more elongate and unexpanded pronotum, absence of pronotal markings, major profemoral spine longer than protarsus, fewer c-sc crossveins, and smaller body size. The species can be separated from all modern Dicromantispa by the absence of the ventromedian lobe of the ectoproct. Such a feature would appear to exclude the species from the genus. However, this feature is perhaps plesiomorphic, and in the absence of further genitalic details it seems unwarranted to establish a new genus for the species. In all other features the species best approximates Dicromantispa among New World mantispids.
Description
Forewing length 7.97 mm; total body length 10.4 mm (figs. 4, 5); pronotum length 3.35 mm, width (median) 0.27 mm. Occipital margin of head absent. Scape, pedicel, and basal five flagellomeres piceous, remainder fuscous; 28 flagellomeres. Labial and maxillary palp segments elongate and tapering at apices. Pronotum elongate (length approx. 10 times width), not greatly expanded anteriorly; pubescence minute and extremely sparse; pronotum without markings; integument imbricate. Profemur with long, sharply pointed basal spine along inner margin, followed on outer margin by numerous short, pointed teeth intermixed with at least three intermediate length spines; basal, inner spine longest, longer than protarsus (fig. 6); integument of outer surface verrucose, otherwise imbricate. Pretarsal claws with three inner teeth except propretarsus with single claw and no arolium. Forewing longer than abdomen; six subcostal crossveins (c-sc); pterostigma elongate, expanded apically, without markings. Hindwing with single gradate series; CuP bending sharply toward 1A; CuP separated from 1A at bend by extremely short crossvein (cup-a). Wing membrane hyaline.
Abdomen without markings; integument imbricate. Ninth sternum broadly rounded along apical margin. Ectoproct elongate and rounded with patch of dense, stiff, black setae on inner margin, without ventromedial lobe (figs. 7, 8).
Etymology
The specific epithet is a reference to the occurrence of this species in Mexican amber.
Fig. 10.
Holotype of Dicromantispa moronei, new species (MACT-432), with details of foreleg enlarged.
Mantispid sp.; Grimaldi, 1996: 93.
Mantispa sp.; Poinar and Poinar, 1999: 130, fig. 129.
Mantispa sp.; Poinar, 2006: 416.
Description
Forewing length 14.5 mm; total body length 17.7 mm (fig. 9); pronotum length 7.7 mm, width (median) 0.6 mm, width (anterior) 1.3 mm. Occipital margin of head absent. Antenna light brown; 29 flagellomeres; F2–28 each wider than long; F1 longer than wide and slightly longer than combined lengths of F2–3; F29 conical. Labial and maxiallary palp segments elongate and tapering at apices. Pronotum elongate (length approx. seven times maximum width), greatly expanded anteriorly; pubescence minute and extremely sparse; pronotum with light brown stripes along lateral posterior thirds; integument imbricate. Profemur with long, sharply pointed basal spine along inner margin, followed on outer margin by numerous short, pointed teeth intermixed with at least five intermediate length spines; basal, inner spine longest, slightly shorter than protarsus (fig. 10); integument of outer surface verrucose, otherwise imbricate. Pretarsal claws with three inner teeth except propretarsus with single claw and no arolium. Forewing longer than abdomen; seven subcostal crossveins (c-sc); pterostigma elongate, expanded apically, without markings. Hindwing with single gradate series; CuP bending sharply toward 1A; CuP separated from 1A at bend by extremely short crossvein (cup-a). Wing membrane hyaline.
Abdomen distended, apparently without markings.
Etymology
The specific epithet is a patronymic honoring Dott. Ettore Morone, Turin, Italy.
Feroseta prisca Poinar, nomen emendatum
Feroseta priscus Poinar, 2006: 412.
Comments
This species was only recently described from Miocene amber of the Dominican Republic. We have not had the privilege of examining the holotype, as it resides in a private collection, but the description and published sketches of the specimen demonstrate that it is not conspecific with the taxa we have established herein. The species was placed in a new genus, the validity of which seems somewhat dubious. We have not been able to further investigate the placement of this species, and it is hoped that future students of the Mantispidae will be able provide greater clarity on the affinity of this fossil. The construction of the generic name, Feroseta, indicates a feminine gender and the specific epithet, therefore, must be emended to prisca.
Family Coniopterygidae Burmeister
The dustywings, family Coniopterygidae, perhaps occupy a relatively basal position among planipennian families owing to the retention of a well-developed and projecting labrum in the larva (Withycombe, 1924), although it has also been suggested that the family is more derived and allied to the Sisyridae and the Berothidae + Mantispidae + Dilaridae + Rhachiberothidae clade (e.g., Aspöck, 1992; Aspöck et al., 2001). There are approximately 460 living species presently recognized in the family (e.g., Meinander, 1972, 1990, 1995, 1998b; Sziraki, 1994, 1997: and various other supplements), which is distributed worldwide. Dustywings are relatively plant-specific and occur on conifers and deciduous trees or bushes. Adults and larvae are generalist predators, feeding on aphids, coccids, and mites, among other minute organisms.
At present, 16 fossil and two subfossil species of Coniopterygidae have been described (table 7), with the oldest being two species from the Late Jurassic of Eurasia. The fossil species were reviewed by Meinander (1975) with additions by Whalley (1980), Meinander (1998a), Grimaldi (2000), and Engel (2002b, 2004a). Until now only a single described species occurred in Dominican amber (Meinander, 1998a). Here we describe two new species: one of the genus Coniopteryx and the other of Spiloconis. Today Coniopteryx is distributed worldwide, and at least two species occur in the West Indies (Dominican Republic). The two species of apparently the genus Spiloconis would represent a remarkable extinction for the West Indian fauna, owing to the restricted presence of this genus to southeast Asia, Australia, and some Oceanic Pacific Islands. However, the monophyly of Spiloconis is uncertain and may be an artifact of unresolved specific relationships.
Table 7
Named Fossil Coniopterygidaea
Coniopteryx antiquua, new species
Figs. 11–15.
Coniopteryx antiquua, new species. 11. Holotype (DR-14-25). 12. Paratype female (DR-014-268). 13. Paratype male (DR-14-853). 14. Paratype (DR-10-14). 15. Paratype (DR-10-15).
Diagnosis
Body lacking typical scalelike setae, basal flagellomeres wider than long, and wing membranes hyaline.
Description
Total body length 1.05 (1.05–1.77) mm; forewing length 1.48 (1.41–2.01) mm. Maxillary palpus normal, i.e., third segment slightly longer than broad. Frons slightly swollen between and above antennae, without hook. Antennae with whorls of setae, some as long as flagellomere width; basal flagellar articles wider than long, progressively becoming as wide as long. Integument pale brown, antennae concolorous. Body with sparse, minute setae, without scalelike setae; legs with scattered short setae, mostly on tibiae. Wing membrane hyaline; veins pale brown.
Paratypes
Female; AMNH DR-10-14 (fig. 14), Miocene amber of the Dominican Republic. AMNH DR-14-268 (fig. 12), Miocene amber of the Dominican Republic. AMNH DR-10-15 (fig. 15), Miocene amber of the Dominican Republic. AMNH DR-14-853 (fig. 13), Miocene amber of the Dominican Republic. MACT-2966 (fig. 16), Miocene amber of the Dominican Republic. MACT-2760, Miocene amber of the Dominican Republic.
Etymology
The specific epithet is taken from Latin antiquus and is a reference to the antiquity of the species.
Spiloconis glaesaria Meinander
Figs. 17, 18.
Spiloconis glaesaria Meinander. 17. Holotype (DR-10-59). 18. Newly discovered female (DR-14-1094).
Spiloconis glaesaria Meinander, 1998a: 33.
Diagnosis
A larger species with forewing length 2.33–2.35 mm and body size 2.00–2.05 mm (latter metrics from holotype). Basalmost flagellomere similar to following segments, not distinctly swollen by comparison. Crossvein r-rs meeting stem of Rs slightly before R2+3/R4+5 fork.
Comments
Since this species was described a second female has been identified in the collection of the AMNH (refer to Meinander [1998a] for a description of the species). The second specimen reported herein agrees with the holotype except in the position of the distalmost seta on M being apicad of the junction of the basal rs-m and cua-m crossveins.
Spiloconis oediloma, new species
Diagnosis
This species differs from S. glaesaria by the smaller forewing length and body size (fig. 19), the swollen basal flagellomere that is distinctly larger than the following segments, and the r-rs crossvein distad to the fork of Rs (thereby meeting R2+3) (fig. 20).
Description
Forewing length 1.85 mm; total body length 1.38 mm. Compound eyes not reduced, more than one-half of head height; gena only slightly narrower than compound eye in lateral view (the gena is considerably more narrow than the compound eye in S. glaesaria); head brown. Antenna 23-segmented, uniformly dark brown; scape twice as long as broad; basal flagellomere distinctly larger than following flagellomeres; segments with two whorls of setae. Thorax dark brown; legs light brown. Forewing with dark markings faint but as described for S. glaesaria; distal setae on M near junction of basal rs-m and cua-m crossveins; basal rs-m crossvein shortly beyond R/Rs fork; distal rs-m on R4+5; sc-r crossvein confluent with r-rs crossvein, r-rs distad fork or Rs and meeting R2+3 (fig. 20).
Family Hemerobiidae Latreille
The family Hemerobiidae, the brown lacewings, consists of approximately 560 species distributed throughout the world. Species occur on vegetation as both adults and larvae, presumably owing to the occurrence of their principle prey (e.g., Sternorrhyncha) in these habitats. Larvae are predaceous and while adults are presumed to be as well, little information is available on their feeding habits. Some adults have been discovered eating honeydew.
Brown lacewings are not common in the fossil record (table 8), and a confident identity for those described in Baltic amber has yet to be achieved. The oldest hemerobiid is presently a single species from the Jurassic of Kazakhstan and shares at least a few synapomorphies typical of Tertiary and recent hemerobiids (Oswald, 1993). Presently there are two species known in Dominican amber, one of which (Notiobiella thaumasta Oswald) appears to be more common than other Dominican neuropterid fossils.
Table 8
Named Fossil Hemerobiidaea(an unnamed species of Sympherobius is reported herein in Dominican amber)
The first amber hemerobiid discovered was of the Notiobiellinae and the nominate genus, in particular. The genus Notiobiella is widely distributed with species occurring in South and Central America, Africa, southeast Asia, Australia, and scattered Pacific islands. Of the approximately 36 described species, two occur today in the West Indies, and the fossil N. thaumasta is apparently most closely related to N. israeli from Cuba and Hispaniola (Oswald, 1999). The second species is a new species of the Sympherobiinae. The species described below falls readily into the Sympherobiinae and Sympherobius except for one character (presence/absence of 4m-cu: see Comments below under the species). Sympherobius is widely distributed in tropical regions of North and South America and temperate regions of the former two continents as well as Africa, Asia, and Europe. Like Notiobiella, only two living species, Sympherobius insulanus Banks and S. zelenyi Alayo, are distributed in the West Indies, both on the island of Cuba (Alayo, 1968; Oswald, 1988).
Morphological terminology follows that of Oswald (1993).
Notiobiella thaumasta Oswald
Figs. 21–25.
Photomicrographs of Notiobiella thaumata Oswald. 21. MACT-3397. 22. MACT-2972. 23. MACT-3240. 24. MACT-2965. 25. Holotype (DR-14-1124).
Notiobiella thaumasta Oswald, 1999 [2000]: 298.
Paratypes
Female; AMNH DR-14-1124, Miocene amber of the Dominican Republic. Female; AMNH DR-14-1129, Miocene amber of the Dominican Republic. Female; AMNH DR-SH-18, Miocene amber of the Dominican Republic.
New Material
MACT-3397 (fig. 21), Miocene amber of the Dominican Republic. MACT-2972 (fig. 22), Miocene amber of the Dominican Republic. MACT-3240 (fig. 23), Miocene amber of the Dominican Republic. MACT-3241, Miocene amber of the Dominican Republic. MACT-2965 (fig. 24), Miocene amber of the Dominican Republic. MACT-1092, Miocene amber of the Dominican Republic.
Comments
This species was recently described on the basis of a single male and three female specimens. The new material does not alter the concept of the species, and therefore the description of Oswald (1999) is not repeated or emended here.
Sympherobius sp.
Diagnosis
The species is unique within the genus for the plesiomorphic retention of crossvein 4m-cu (absent in all Recent Sympherobius), the presence of the distal radial crossvein (r2+3–r4+5), and the unpatterened wings.
Description
Total body length 2.52 mm; forewing length 2.53 mm. Integument light brown; flagellum concolorous; 42 flagellomeres (observed for left antenna; right antenna damaged slightly beyond midpoint). Wing membrane hyaline; longitudinal veins and crossveins light brown. Forewing with proximal humeral trace strongly recurrent, trichosores present on humeral margin, humeral veinlets simple; costal space broad; Sc widely separated from R along its entire length; 2sc-r absent; two prestigmal radial sectors, posterior radial sector with a single branch proximad midpoint; four interradial crossveins comprising outer gradate series, outer gradates alternating in position (not stepped in a single direction); proximal radial crossvein (r1+2+3–r4+5) absent; distal radial crossvein (r2+3–r4+5) present; two r-m crossveins (i.e., 2r-m and 3r-m), 4r-m absent; 3im present, 4im absent; 4m-cu present; CuA branching distad 2m-cu, next branch confluent with 3m-cu; CuP simple, not forked proximad 2cua-cup. Setae generally pale brown, minute, and sparse except on wings setae more elongate and numerous and at abdominal apex setae elongate. Terminalia not visible.
Comments
Owing to the uncertain final repository of the only known specimen (which resides in the collection of a dealer who intends to sell it), we have avoided describing this species as new until such time as new material is discovered or the specimen is deposited in a more permanent collection, preferably that of a museum.
The fossil can be easily assigned to the Sympherobiinae based on the following combination of characters in the forewing: proximal humeral trace recurrent (strongly so in the fossil), trichosores present on humeral margin, broad costal space, 2sc-r absent, two prestigmal radial sectors, posterior radial sector branching proximad its midpoint, 4r-m absent, and CuP simple, not forked proximad 2cua-cup. Within the sympherobiines, the fossil is apparently sister to living Sympherobius and is placed basal within this genus. Although the fossil lacks one of the four synapomorphies of the genus (i.e., apomorphic loss of forewing 4m-cu), we have placed the species here rather than erect a monotypic genus sister to Sympherobius based upon a single character (moreover, a single plesiomorphic character!). Both Nomerobius and Neosympherobius plesiomorphically retain 4m-cu (as well as 4im in the forewing, which is absent in the fossil). The fossil can be immediately excluded from both of these genera as follows: absence of 4im in forewing (present in Nomerobius and Neosympherobius), presence of outer gradates (absent in Neosympherobius), four outer gradates alternating in position (stepped in a single direction in Nomerobius). Until a cladistic study of the world Sympherobius fauna is completed, it is not possible to definitively determine whether the fossil is indeed sister to the remainder of the genus, but the loss of 4m-cu would apparently be a good synapomorphy of Recent Sympherobius, as it was originally hypothesized by Oswald (1993). However, note that the fossil is not without some similarity to the principally eastern United States species S. amiculus (Fitch) and S. umbratus (Banks), all of which are unique among Sympherobius for the presence of a distal radial crossvein, this being either a plesiomorphy or the secondary reacquisition of this feature (as hypothesized by Oswald [1988] for the two living species). It is certainly plausible that this character is a plesiomorphy for the fossil and a synapomorphy for the two living species rather than a character uniting all three taxa (i.e., if the fossil is indeed basal while S. amiculus and S. umbratus are more derived, as other characters would suggest). Such determination will await a cladistic analysis for all Sympherobius.
Family Chrysopidae Schneider
Larvae are active, generalist predators. Some species rapidly cover themselves with debris (held in place by long, curved setae), which may include remains of their prey. Most species are arboreal. Adults are commonly predators although some lineages are specialized to feed on honeydew and harbor symbiotic yeasts, presumably to facilitate digestion.
Although a veritable plethora of chrysopid fossils have been previously described or reported on (e.g., Adams, 1967; Carpenter, 1935; Cockerell, 1909, 1914; Handlirsch, 1908; Handschin, 1937; Makarkin, 1991, 1994; Martins-Neto and Vulcano, 1989c; Martynov, 1927; Nel and Henrotay, 1994; Nel and Séméria, 1986; Panfilov, 1980; Peñalver et al., 1995; Ren and Guo, 1996; Schlüter, 1982; Scudder, 1890; Séméria and Nel, 1990; Statz, 1936; Willmann and Brooks, 1991), none has been discovered in amber until now. Fossils recognizable as Chrysopidae (albeit of an extinct, plesiomorphic subfamily) are known as far back as the Late Jurassic. The wing terminology of Brooks and Barnard (1990) is used for the three chrysopine fossils described herein to facilitate comparison with their monograph on the world fauna of living Chrysopidae.
The genera Chrysopa and Chrysoperla are separated mostly by behavior and minutiae of the terminalia. However, there are some minor differences of wing venation that distinguish these two genera. Both of the fossils considered below fall within the Chrysopa class of wing venation and are therefore considered to be of this genus.
Chrysopa? glaesaria, new species
Chrysopa sp.: Grimaldi and Engel, 2005: 351, fig. 9.30.
Diagnosis
This species differs most notably from C. vetula, also in Dominican amber, by the submedian longitudinal stripes on the pronotum and larger size. Other differences are presented in the following description.
Description
Forewing length 16 mm; total body length 12.1 mm; pronotum length 1.0 mm, width 1.25 mm. Integument generally golden with faint green highlights except when indicated below; integument faintly imbricate. Palpi tapering at apices; labrum emarginate; lateral quarters of clypeus black; scapes close together with black on inner surfaces; flagellum about as long as forewing.
Pronotum with two narrow, longitudinal stripes of dark brown on either side of midline, lateral margins not marked, slightly wider than long, anterior margin broadly rounded; meso- and metathoracic pleura dark brown, mesothorax with narrow, longitudinal stripes of dark brown on either side of midline, laterally unmarked.
Forewing oval and broad, longer than abdomen, with sparse setae except along margins; 22 costal crossveins (c-sc); costal area narrows progressively toward apex; Sc and R deviate slightly near apex, Sc joins C at apex; R straight, not sinuous; single sc-r crossvein toward wing base (vein bsc), four sc-r crossveins below pterostigma; pterostigma elongate and unmarked; cell m1 much smaller than m2 (as typical for Chrysopa, while in Plesiochrysa these cells are closer in size); cell im ovate, broad, first rs-m crossvein meets im subapically; no radial crossvein (r-m) basal to origin of Rs, radial crossveins straight; cell c1 shorter than c2; cell dcc narrowly open along posterior wing margin; six closed cells between PsM and PsC; two gradate series, inner gradate series incomplete, anterior gradates absent, gradates parallel; 1A forked (fig. 27). Hindwing with two gradate series. Wing membrane hyaline; veins light brown.
Abdomen apparently with basal bands of dark brown on terga; eighth and ninth terga not fused; sternum two without stridulatory apparatus; callus cerci rounded.
Etymology
The specific epithet is derived from the Latin word glaesarius, meaning “of amber”.
Chrysopa ? vetula, new species
Description
Forewing length 9.7 mm; total body length 7.6 mm; pronotum length 0.7 mm, width 0.75 mm. Integument generally golden with faint green highlights except when indicated below; integument strongly imbricate. Palpi tapering at apices; labrum emarginate; lateralmost borders of clypeus black; scapes close together with black on inner and outer surfaces; flagellum about as long as forewing.
Pronotum with two narrow, longitudinal stripes of dark brown on lateral borders, slightly wider than long, anterior margin broadly rounded; meso- and metathoracic pleura apparently light brown, mesothorax with narrow, longitudinal stripes of dark brown on anterolateral margins.
Forewing oval and broad, longer than abdomen, with sparse setae; 18 costal crossveins (c-sc); costal area narrows progressively toward apex; Sc and R deviate slightly near apex, Sc joins C at apex; R straight, not sinuous; single sc-r crossvein toward wing base (vein bsc), three sc-r crossveins below pterostigma; pterostigma elongate and unmarked; cell m1 much smaller than m2; cell im ovate, relatively narrow, first rs-m crossvein meets im subapically; no radial crossvein (r-m) basal to origin of Rs, radial crossveins straight; cell c1 shorter than c2; cell dcc narrowly open along posterior wing margin; six closed cells between PsM and PsC; two gradate series, inner gradate series incomplete, anterior gradates absent, gradates parallel; 1A forked. Hindwing with two gradate series. Wing membrane hyaline; veins light brown.
Abdomen apparently light brown, somewhat distended; eighth and ninth terga apparently not fused; sternum two without stridulatory apparatus.
Etymology
The specific epithet is derived from the Latin word vetus (meaning “old”).
Leucochrysa (Nodita) prisca, new species
Diagnosis
A relatively small species of Leucochrysa (at a forewing length of 10.5 mm) lacking the red or brown markings of the head.
Description
Forewing length 10.5 mm; total body length 6.6 mm; pronotum length 0.63 mm, width 0.73 mm. Head apparently light brown, antennae uniformly light brown, face darker brown below level of antennae, markings apparently absent; integument apparently imbricate. Palpi tapering at apices; labrum emarginate; flagellum about as long as forewing.
Pronotum apparently unmarked, slightly longer than wide, anterior margin broadly rounded; remainder of body apparently light brown, perhaps with metallic highlights.
Forewing oval and narrow, longer than abdomen, with sparse setae except along margins; 16 costal crossveins (c-sc); costal area narrows progressively toward apex; Sc and R deviate slightly near apex, Sc joins C at apex; R straight, not sinuous; single sc-r crossvein toward wing base (vein bsc), three sc-r crossveins below pterostigma; pterostigma elongate and marked with faint clouds of smoky brown; cell m1 much smaller than m2; cell im ovate, broad, first rs-m crossvein meets im subapically; no radial crossvein (r-m) basal to origin of Rs, radial crossveins straight; cell c1 as long as c2; cell dcc narrowly open along posterior wing margin, longitudinally divided anteriorly to form small cell c3; six closed cells between PsM and PsC; two gradate series, inner gradate series complete, gradates parallel; 1A forked (fig. 33). Hindwing with two gradate series. Wing membranes hyaline; major longitudinal veins light brown, although darkened at points where crossveins form; crossveins distinctly darker than longitudinal veins, darker where two or more veins meet each other.
Abdomen uniformly colored; sternum two without stridulatory apparatus.
Etymology
The specific epithet is taken from the Latin word priscus (meaning “old”).
Chrysopidae larva
Chrysopidae larva: Grimaldi and Engel, 2005: 352, fig. 9.31.
Comments
Integument reddish-brown throughout, without apparent markings and largely lacking microtrichiae. Head distinctly flattened and roughly quadrate, slightly longer than wide; antennae longer than mandibles and slightly longer than head capsule; mandibles nearly as long as head capsule; mandibles simple, without dentition. Lateral tubercles on thorax and abdominal segments I–VII well developed (fig. 34), tubercles spherical and nodulate at apices at setal origins; prothoracic segment with secondary, smaller, lateral tubercles posterior to main lateral tubercles, secondary tubercles with short, simple, erect setae; main lateral tubercles of each segment with 4–6 simple, greatly elongate setae forming a dorsal space or corbicula for the accumulation of debris on the dorsal surface of the thorax and abdomen. Thoracic and abdominal dorsa relatively flattened except slightly upcurved and flared dorsolaterally to form dorsal corbicula; dorsum without setae except metathoracic and first four abdominal segments with short, erect, setae with spatulate apices medially (fig. 34). Abdomen tapering in width to relatively narrow apical segment; apical segment tapered to rough point at apex.
Comments
Like modern chrysopid larvae, the fossil exhibits the flattened dorsum and elongate, curved, lateral setae designed to form a dorsal corbicula. The larva fills the corbicula with debris so as to disguise itself not only from predators but from prey (the so-called “sheep in wolf's clothing”). Given the distribution of this behavior and such morphological adaptations across the diversity of chrysopids, it is not surprising to find a larva exhibiting this suite of features as along ago as the Miocene. Indeed, such design and behavior may stretch back into the Cretaceous.
Family Myrmeleontidae Latreille
The antlions are notable predators, familiar for the conical pits constructed by larvae of some species. Larvae dwell in the ground or on trees and it seems that eggs are laid singly, sometimes scattered in dry soil or sand. Most larvae are subterranean and some construct conical pits that are used as traps for prey. The larva lies buried at the bottom of the pit with only the open jaws protruding. In other groups, the larvae burrow freely in sand, and move under the sand or on the surface at night. Still others appear to live on trees and closely resemble bark. Larvae can often be found under overhangs where they are protected from rain. Adults are typically predatory although some are known to feed on pollen.
Several fossils of the family are known as compressions from the Early Cretaceous of Brazil and China (e.g., Martins-Neto, 1990, 1992; Martins-Neto and Vulcano, 1989a, 1989b). Many of these species have been assigned to a variety of new, extinct subfamilies but the validity of these requires careful reevaluation (e.g., the Araripeneurinae is a synonym of Palparinae). The only other Cretaceous records of the antlions are the fossils Samsonileon fragmentus Ponomarenko from Israel (Lower Cretaceous; Dobruskina et al., 1997) and Palaeoleon ferrogeneticus Rice from Labrador (Upper Cretaceous; Rice, 1969), both sometimes placed in a separate family, Palaeoleontidae (e.g., Dobruskina et al., 1997; Martins-Neto, 2000). Additionally, Whalley (1980) reported an incomplete, unnamed specimen of a putative myrmeleontid in Lebanese amber but we have not been able to confirm its identity. Aside from the fossils discussed below, Dendroleon septemmontanus Statz is the only compression of a myrmeleontine from the Tertiary (Statz, 1936). Only a single, definitive myrmeleontid species is known in amber from any deposit, this being the recently described Porrerus dominicanus from Dominican amber (Poinar and Stange, 1996). Below we make minor corrections to the original description, present new figures of the holotype, and report on additional specimens.
Porrerus dominicanus Poinar and Stange
Figs. 39–41.
Wing venation and leg spination of Porrerus dominicanus Stange and Poinar (MACT-3496). 39. Forewing. 40. Hind wing. 41. Leg spination.
Porrerus dominicanus Poinar and Stange, 1996: 384.
Holotype
Female; MACT-1220 (figs. 35, 36), Miocene amber of the Dominican Republic, La Toca mine (formerly in the Work Collection, Oregon).
New Material
Female; MACT-1170 (figs. 37, 38), Miocene amber of the Dominican Republic. Female; MACT-1414, Miocene amber of the Dominican Republic. Female; MACT-3496 (figs. 39–41), Miocene amber of the Dominican Republic.
Comments
This species was recently described on the basis of a single female (fig. 35, 36). The original description requires some alteration and is emended here. The number of presectoral crossveins reported by Poinar and Stange (1996) is in error not only by comparison with the two new specimens but also with the holotype. The basalmost crossvein has become almost entirely cleared as a product of preservation and was apparently overlooked by these authors. Thus, the number of presectoral crossveins must be increased by one (to five). The other two specimens reported below agree with the holotype. Some minor variations exist between the specimens merely in the number of crossveins (by one or two). The complete fore- and hind wing venation of P. dominicanus is depited in figures 39 and 40, while figure 41 depicts the hindleg spination of the species.
Porrerus dominicanus? [Larva]
Porrerus larva: Grimaldi and Engel, 2005: 346, fig. 9.20.
Description
Head length 1.2 mm; anterior width 0.8 mm; posterior width 0.46 mm; total body length 4.5 mm. Head, pronotum, and legs more strongly sclerotized than remainder of body. Head elongate and trapezoidal, anterior width approximately 1.75 times posterior width; without dolichasters; with numerous, scattered, elongate, black setae; dorsum of head very gently convex along posterior two-thirds, anterioly between mandibles strongly concave; venter of head relatively flat. Mandibles slightly upturned, longer than head, widely separated (separated by approximately 3.75 times basal mandibular width); three teeth, first tooth situated about at midpoint of mandible, teeth progressively longer toward mandible apex, parallel; setae on outer margin of mandible longer than greatest width of mandible; fringe of appressed setae along outer basal margin of mandible. Labial palpus about as long as basal mandibular width. Ocular tubercle present, short (i.e., not prominent), cylindrical, without dolichasters. Antennal tubercle positioned close to ocular tubercle, separated by less than width, without dolichasters; apparently 18 antennal segments. Pronotum quadrangular, apparently about as wide as long. Mesothoracic spiracle sessile, situated between tubercles. Claws elongate and relatively straight, simple; legs with elongate, black setae, longer than leg width. Meso- and metathoracic and abdominal segments with transverse rows of elongate, black setae, such setae not borne on tubercles; more dense, shorter setae situated as a patch medially between meso- and metathoracic segments, such setae not arranged in rows. Body without scoli, with paired lateral tubercles each bearing numerous elongate setae. Eighth sternum with pair of small, submedian teeth near posterior margin; ninth sternum with numerous stout setae, without bladelike digging setae, with some shorter, stout setae borne on a common base near posterior margin and slightly upturned.
Material
Larva; MACT-1282 (figs. 42, 43), Miocene amber of the Dominican Republic. The specimen is preserved along with a scelionid wasp, a mite, a beetle larva, and a collembolan.
Comments
The specimen is clearly an immature of the tribe Myrmeleontini as is evidenced by the combination of elongate mandibles, elongate mandibular setae, sessile mesothoracic spiracle, presence of submedian teeth on S8, and absence of bladelike digging setae on S9. Immatures of myrmeleontines are relatively homogeneous and characters separating the genera are presently not well defined. For this reason, confident assignment of this specimen to any genus of Myrmeleontini is tenuous but the presence of several adults of Porrerus (see above) is suggestive that this may be an immature of the same species. We have therefore tentatively assigned this immature to P. dominicanus.
Family Ascalaphidae Rambur
The owlflies, family Ascalaphidae, consist of approximately 430 species. The family is most diverse in both xeric and mountainous regions of the subtropics or tropics. Owlflies species are frequently large and some groups appear to represent a conglomeration of features from other orders of insects, overall sharing a superficial habitus with dragonflies. Female ascalaphids lay eggs in clusters on twigs or grass stems. The first-instar larvae aggregate at twig apices into a defensive ring. The larvae eventually disperse to live solitarily in the litter or on trees where they are generalist predators.
The geological record of Ascalaphidae is exceedingly sparse with four definitive ascalaphids previously proposed from Tertiary compression fossils in Europe and the Russian Far East, one species known only from the larva in Baltic amber, as well as one species dubiously assigned Lower Cretaceous fossil (table 9). Herein we describe the first adult ascalaphids in amber as well as larvae of two species of the genus Ululodes. Interestingly, the fossil species are representative of two of the three known subfamilies: Ascalaphinae (Ululodes paleonesia, new species) and Haplogleniinae (Amoea electrodominica, new species).
Table 9
Named Fossil Ascalaphidaea
Amoea electrodominicana, new species
Ululodes adult: Grimaldi and Engel, 2005: 347, fig. 9.23.
Diagnosis
Like modern Amoea, this species exhibits the absence of wing pigmentation (excluding the pterostigma) otherwise found in all other Haplogleniinae. The fossil species has more elongate antennae (longer than the forewing), lacks integumental markings, and has CuA relatively straight to the wing margin (fig. 44) (more strongly arched posteriorly in modern species).
Description
Total body length 21 mm; forewing length 23 mm. Integument light brown, without apparent maculations. Head longer than wide; compound eyes not divided; antenna arising near lower tangent of compound eyes; antenna elongate, longer than body or wings; scape and pedicel with elongate setae, setae nearly as long as scape; antennal articles each distinctly longer than wide except those of antennal club wider than long, with a few setae near apex of each article; antennal club composed of nine articles. Fore- and hind wings of equal length and approximately equivalent shape; pterostigma of fore- and hind wing formed of dark pigmented spot near wing apex where Sc meets R (fig. 44) and composed of several closed cells, remainder of wing membranes hyaline; veins black; forewing M and CuA relatively straight to wing margin, weakly arching posteriorly at extreme apex; forewing axillary angle obtuse, nearly orthogonal, not produced; hind wing CuP not arched posteriorly, instead running parallel to 1A; 2A absent; hind wing without anal lobe. Body relatively densely pubescent, setae simple and dark brown; those setae of femora and tibiae elongate and more stout (fig. 44); pretarsal claws long, thin, and simple; abdominal apex with dense, stiff, short, black setae.
Etymology
The specific epithet is a reference to the occurrence of this species in Dominican amber.
Ululodes paleonesia, new species
Diagnosis
The new species is noteworthy for the combination of the elongate antennae (longer than the forewing), pterostigma of forewing darkly pigmented and enlarged, and hyaline wing membranes. The species closely resembles the modern U. macleayana (Guilding) but differs by the more elongate antennae and the distinctly narrower area of the wing posterior to CuA (fig. 46) (maximal width of posterior area distinctly less than width of area anterior to CuA at same point while in U. macleayana the posterior area is broader than the anterior area).
Description
Length as preserved approximately 13 mm (only preserved from head through portion of abdominal segment IV: figs. 45, 46); forewing length 22 mm. Integument light brown, without apparent maculations. Head about as long as wide; compound eyes distinctly divided horizontally near midpoint; antenna arising near midpoint of head length; antenna elongate, longer than forewing; scape, pedicel, flagellar articles with simple, erect setae near apices; flagellar articles each distinctly longer than wide (approximately twice as long as wide) except those of antennal club wider than long. Pterostigma of forewing formed of darkly pigmented spot near wing apex where Sc meets R and composed of several closed cells (fig. 46), remainder of wing membrane hyaline; membrane of preserved portion of hind wings hyaline (figs. 45, 46); veins of both wings black; maximal width of posterior area of wing posterior to CuA less than width of area anterior to CuA at same point in forewing; hind wing CuP distinctly arched posteriorly toward its apex (fig. 46); 2A absent; hind wing without anal lobe. Body relatively densely pubescent, setae simple and dark brown; those setae of femora and tibiae of medium length, long but shorter than those of thorax; pretarsal claws long, thin, and simple; abdominal apex with dense, stiff, short, black setae.
Etymology
The specific epithet is a combination of the Greek words palaios (meaning “ancient”) and nesos (meaning “island”) and is a reference to the Miocene, island habitat of this species.
Ululodes sp. 1 [Larvae]
Description
Third instar?: Body length (exclusive of mandibles) 5.9 mm; head capsule length 1.5 mm; head capsule width 1.8 mm. Head, pronotum, and mandibles dark brown; remainder of body brown. Head quadrate, slightly wider than long, dorsoventrally flattened. Posterolateral margins weakly cordate. Labral margin narrow and strongly bilobed with a single wide notch at midline; labral lobes bulbous. Dorsum of head weakly convex medially and laterally; ventral surface convex with excavated anterolateral margins to permit retraction of jaws beneath ocular tubercle. Surface integument granulose, covered with scattered clumps of particulate debris. Ocular tubercles prominent although not particularly enlarged; slightly cylindrical, not tapered distally or flattened; all five stemmata similar in size. Antennal tubercles prominent, cylindrical, nearly one-half length of ocular tubercle, well separated from latter by approximately length of antennal tubercle. Jaws slightly falcate and with slight upward tilt, tapering at apices, much longer than head capsule. First tooth of mandible situated approximately at midpoint; central tooth at about midpoint between first tooth and apex; central tooth longer than other two teeth and slightly curved; distal tooth smaller than preceding two teeth and situated at midpoint between apex and central tooth.
Body ovoid in outline. Pronotum small, sclerotized, slightly convex, trapezoidal, about twice as wide as long, without scoli or tubercles but with short, stiff setae pointed anteriad. Remainder of body lightly sclerotized. Lateral margins of meso- and metathoracic and abdominal segments except ninth and tenth with prolonged, unflattened, fingerlike, setose scoli, 10 pairs in total and decreased in length posteriad. Each mesothoracic scolus as long as head capsule or nearly so, inclined forward basally and reflexed posteriorly at about its midpoint. Metathoracic scoli two-thirds as long as mesothoracic scoli; not relfexed. Remaining eight pairs of abdominal scoli straight, shorter than thoracic scoli. Pairs of vestigial scoli present, positioned behind and slightly ventrad elongate scoli on meso- and metathorax.
Dense, double fringe of long serrate setae present on lateral margins of head and peripherally on all scoli. Single row of shorter, distally flared, serrate dolichasters closely set along entire anterior labral margin. Ocular tubercle with still shorter, tightly packed dolichasters, with two elongate, subequal setae projecting posteriad from posterior margin. Antennal tubercle bearing a group of three long dolichasters. Two pairs of stout, toothlike digging setae present on posterior border of ninth abdominal segment; dolichasters present along margin of apical abdominal segment.
Material
Larva; MACT-1200 (figs. 47, 48), Miocene amber of the Dominican Republic. Larval head; AMNH DR-10-2010 (fig. 49), Miocene amber of the Dominican Republic.
Comments
This larva can be confidently assigned to the genus Ululodes based on the excellent descriptive work of living ascalaphid immatures by Henry (1976). The absence of pronotal tubercles, presence of 10 scoli, and reflexed mesothoracic scoli, among many other characters (Henry, 1976), are all indicative of the genus Ululodes. Although tempted to do so, we have not assigned this fossil immature to the species U. paleonesia despite the presence of an adult of this species in the same deposit. Given that two diagnosable species of larval Ululodes can be recognized (see Ululodes sp. 2 below) it is entirely unknown to which, if any, the adult might belong. The behavior of ascalaphid larvae mentioned above perhaps accounts for the entrapment of this immature in resin and preservation in amber.
Ululodes sp. 2 [Larva]
Ululodes larva: Grimaldi and Engel, 2005: 348, fig. 9.24.
Description
Third instar?: As described for Ululodes sp. 1 except as follows: Body length (exclusive of mandibles) ca. 4 mm (difficult to ascertain as body is bent: fig. 50); head capsule length 0.9 mm; head capsule width 1.2 mm. Labral lobes bulbous but broader and less prominent than in Ululodes sp. 1. Head and pronotum more densely covered by setae; those elongate setae on lateral margins of head distinctly more numerous and dense (fig. 51). First tooth of mandible situated well beyond midpoint; first tooth slightly shorter than immediately following, central tooth, these each longer than distal tooth (fig. 51). Dolichasters of mandible proximal to first tooth more numerous than in Ululodes sp. 1 (cf. figures 48, 49, and 51).
Discussion
Paleoecology
Many of the neuropterans in Dominican and Mexican amber were predacious as larvae and/or as adults on other arthropods that probably frequented trunks of the amber-producing Hymenaea trees. Adult Mantispidae are generalized predators, much like their dictyopteran analogs, the mantises (Mantodea). Larvae of some mantispids (i.e., Plega and other symphrasines) are terrestrial predators of larval social aculeates and of scarab larvae. Other larvae, including those of Dicromantispa, are parasitoids of spiders, and there is an abundance and considerable diversity of spiders preserved in both Dominican and Mexican amber (Wunderlich, 1988; Penney, 1999). Larval Myrmeleontoidea are morphologically highly specialized predators that are quite generalized in their types of prey. Although the pit-constructing “antlions” are known best, this behavior occurs mostly in Myrmeleontinae. Larvae of some myrmeleontids are roaming predators on tree trunks, which no doubt was the habit of MACT-1282. Larvae of many ascalaphids are ambush predators on tree trunks, to which they are superbly adapted with their very flat bodies that they camouflage with bits of lichen and other debris. Adult ascalaphids often perch on tree trunks, and some are even known to roost each day on hanging branches and vines.
Adult and larval Hemerobiidae, Chrysopidae, and Coniopterygidae tend to be most prevalent where there are dense aggregations of sedentary or sessile Sternorrhyncha, on which they extensively feed. Hemerobiidae are more prevalent in forested areas, in contrast to Chrysopidae, which prefer more open areas. This may account for the slightly greater ease of locating hemerobiids in Dominican amber, although the larger size of chrysopids may bias against their fossilization in amber. Chrysopidae and Coniopterygidae are particularly well adapted to feeding amongst aggregations of aphids, psyllids, whiteflies, and scale insects (Sternorrhyncha). Larval and adult coniopterygids have a covering of flocculent wax over the body, which no doubt disguises them among their prey (many of which also produce wax in abundance). Larval chrysopids camouflage themselves by placing debris, even the husks of their victims, amongst their long hairs (fig. 34) (Eisner et al., 1978; Eisner and Silberglied, 1988).
Oddly, sternorrhynchans of all groups are rare in Dominican amber, comprising less than 1% of all inclusions (Grimaldi, unpubl. data), which probably reflects the distinctly broad-leafed tropical forests in which Dominican and Mexican ambers were produced. In contrast, sternorrhynchans are among the most abundant inclusions in middle Eocene Baltic amber (Aphidoidea: 23% of all inclusions; Larsson, 1978), Campanian-aged Canadian amber (Aphidoidea: 22% of all inclusions; McAlpine and Martin, 1969), and Turonian-aged amber from New Jersey (Coccoidea: 10% of all inclusions; Grimaldi et al., 2000; Koteja, 2000). All of these amber deposits were produced in warm temperate or subtropical coniferous forests. The abundance of Coccoidea in New Jersey amber probably explains the abundance and diversity of Neuropterida in that amber (table 2), particularly regarding the fact that the amount of New Jersey amber excavated is a tiny fraction of that mined in the Dominican Republic or the Baltic region. The New Jersey amber fauna includes several bizarre mantispids, a larval and adult psychopsid, larval and adult mesoraphidiids, and a significant diversity and abundance of berothids and coniopterygids (with the latter two groups being significant coccoid predators). The Baltic amber fauna (table 1) is comprised thus far of 19 species in 10 families of Neuroptera—the most diverse of any amber deposit. While this may reflect the enormous quantities of Baltic amber collected over centuries, it is also probably a reflection of the abundant sternorrhynchan fauna in that amber. Neuropterida are rare in all ambers but appear to be less so (particularly for Neuroptera) for those deposits with a significant abundance of sternorrhynchan prey.
Age of Fauna and Biogeography
Without question, the neuropterid fauna of Dominican and Mexican amber is very similar to the modern fauna of Mesoamerican tropical forests. While the age of Dominican amber has been unnecessarily confused as being Miocene to even Eocene in age (see critique in Grimaldi, 1995a), all critical evidence indicates that it is early Miocene in age (Iturralde-Vinent and MacPhee, 1996, 1999). This conclusion is congruent with the mostly modern character of the biological inclusions in Dominican amber (e.g., Grimaldi, 1995b, for flies; Engel, 1999b, 2001, for bees).
There is an interesting mixture of biogeographic elements among the Neuropterida in Dominican amber. Taxa that have a predominantly Holarctic distribution are Sialis and Chrysopa. Neotropical taxa are represented by Leucochyrsa (also Nearctic), Porrerus, Ululodes (also southern Nearctic), and Sympherobius (also Nearctic). The most interesting biogeographic connection is the presence of two species of Spiloconis, a group today found only in Madagascar, Southeast Asia, and Australia. Old World extinctions in Dominican amber are well documented: Valeseguya woodgnats (Diptera: Anisopodidae; Grimaldi, 1991) and Mastotermes termites (Isoptera: Mastotermitidae; Krishna and Emerson, 1983; Krishna and Grimaldi, 1991), both presently in Australia; certain species of Ogcodes (Diptera: Acroceridae; Grimaldi, 1995b) in Africa and Asia; and Leptomyrmex from the Indo-Australian region (e.g., Baroni Urbani, 1980) as well as some other kinds of ants principally restricted to the Eastern Hemisphere (e.g., Brandão et al., 1999). The interpretation and concomitant implications of these extinctions, however, depend on monophyletic taxa. Unrecognized paraphyly of particular groups can mislead reconstructions by implying a significant biogeographic extinction when, in fact, modern counterparts, living in the same general region as the fossils under question, are classified in a separate supraspecific group that merely represents a monophyletic derivative of the same lineage. The extinction of Spiloconis in the New World seems, for the time being, well founded. The reduced, essentially internalized, male terminalia typical of Spiloconis species is apomorphic for the Aleuropteryginae, thereby supporting the monophyly of the genus. Although male terminalia are unknown for the Spiloconis fossils discussed herein, the observable apomorphic traits in the fossils are known only in species of Spiloconis among aleuropterygines and support the conclusion that the fossils belong to this lineage. Factors that might have led to the localized decline and eventual loss of this genus in the Western Hemisphere remain elusive.
Acknowledgments
We are indebted to Dott. Ettore Morone for graciously allowing us to study his spectacular collection of Dominican amber inclusions and for his patience during the long gestation of this manuscript. Additional thanks are owed to Antonio Arillo and Roy J. Beckemeyer for reading earlier versions of the text and providing valuable criticisms; to Ismael A. Hinojosa-Díaz for translating our abstract into Spanish; and to Sonja Wedmann for an advance copy of her work on fossil mantispids. We are further grateful to Robert G. Goelet, Chairman Emeritus of the AMNH Board of Trustees, who provided funding for the acquisition of the AMNH specimens reported herein and supported M.S.E. during the initial phases of this project at the AMNH. All photographs and illustrations were prepared by the authors (all line illustrations were by D.A.G.) except for the photos presented as figures 16 and 21–24, which were taken by Roy Larimer. Partial support in the latter phase of the project was provided by NSF EF-0341724 and a Guggenheim Fellowship from the John Simon Guggenheim Memorial Foundation (to M.S.E.). This is contribution No. 3457 of the Division of Entomology, Natural History Museum and Biodiversity Research Center, University of Kansas.
