Introduction

Apoid wasps of the subtribe Crabronina, or “moustache wasps”, are rarely discovered as amber inclusions, despite the worldwide occurrence of numerous wood- or twig-nesting species. Most species occur in more xeric environments, areas outside of the tropical or subtropical forests producing amber, and many nest in soil. As such, the rarity of moustache wasps in amber can be at least partially explained by habitat preferences. Nonetheless, a significant variety of forest-loving groups that nest in various plant materials in tropical and temperate areas exists, and thus the problem of fully understanding why these wasps are rare in amber remains. In the search for an explanation, note that ground-nesting is the primitive condition for the subtribe (Bohart and Menke, 1976) and that to a large extent wood- and twig-nesting habits are found among the most diverse genera, such as Ectemnius, Rhopalum, and Crossocerus. If these are derived groups representing recent radiations into niches formerly unused by moustache wasps, then the paucity of these wasps in fossiliferous resins could be explained by habitat preferences in combination with fundamental nesting differences between modern and prehistoric faunas. Whether such speculation can be developed into a supported hypothesis awaits a formal cladistic analysis of the Crabronina (under way by D.J.B.) and a more precise understanding of the evolution of nesting behavior in the subtribe.

Prior to this report, two species of the genus Tracheliodes described by Cockerell (1909) from Baltic amber were the only representatives of the subtribe known from fossiliferous resin (table 1). However, given that Cockerell did not describe the relative position of the eyes in these specimens (note that an important character of Tracheliodes is widely separated eyes ventrally), his generic assignments must be treated with caution (W.J. Pulawski, personal commun.). Compression fossils of Crabronina are equally rare, with the only species being another Tracheliodes and an Ectemnius from Colorado's Eocene-Oligocene Florissant Shale (Cockerell, 1906, 1910) and a putative species of Ectemnius from the Early Miocene of Germany (Meunier, 1911). Interestingly, putative species of Tracheliodes should represent half of the Crabronina known as fossils, as this group is likely basal within the subtribe. Today the genus is composed of about a dozen species occurring in the western United States, Central Europe, and Mediterranean region. The fossil record and modern distribution of Tracheliodes hint at a formerly more diverse and widespread distribution. We herein report on the occurrence of two specimens of the closely related crabronine genus Lindenius recently discovered in Dominican amber (figs. 1–3). This is the first record of a crabronine in Dominican amber and the first fossil of Lindenius.

Table 1

Figure 1–3

Photomicrographs of holotype female of Lindenius paleomystax, new species (AMNH DR-14-1091). 1. Oblique lateral habitus. 2. Facial aspect. 3. Dorsal aspect.

Lindenius is a moderately large genus presently comprised of 60 extant species distributed throughout the Holarctic region (Pulawski, 2006). Most of these are Palearctic, with the highest diversity occurring in Mediterranean countries (Bohart and Menke, 1976). The handful of species studied construct relatively shallow nests in soil and predominantly prey on small flies or Heteroptera, though prey records also include various Hymenoptera, including Chalcidoidea, Braconidae, and ants (Hamm and Richards, 1926; Leclercq, 1954; Court, 1961; Evans, 1970; Bohart and Menke, 1976). North American species were treated in an unpublished Master's thesis (Court, 1961); Palearctic taxa were treated by de Beaumont (1956) and later by Leclercq (1989).

The age of Dominican amber has been unnecessarily confused in recent years, with age estimates ranging from the Miocene through the Eocene. In fact, critical analysis of multiple lines of evidence indicates a Neogene age, and all valid evidence supports an Early Miocene (Burdigalian) origin (e.g., Iturralde-Vinent and MacPhee, 1996). Furthermore, the inclusions themselves attest to a relatively recent age for Dominican amber in comparison with definitively more ancient resins (e.g., refer to comparisons of Dominican and Baltic amber faunas of flies and bees by Grimaldi, 1994 and Engel, 2001, respectively). Morphological terminology generally follows that of Bohart and Menke (1976).

Systematic Paleontology

Figure 4–5

Line drawings of Lindenius paleomystax, new species, as preserved. 4. Face depicting clypeal margin. 5. Apical two-thirds of forewing.

Discussion

Like their more intensely studied relatives, the bees (Anthophila), apoid wasps are uncommon as fossils. Both have relatively comparable numbers of fossil species documented as both compressions and amber inclusions (the bees have slightly more species, but this is a bias resulting from the fact that their geological history has been the focus of work by one of us [M.S.E.] during the last decade). Unlike the bees, however, the apoid wasps have a far better representation in the Cretaceous, thereby providing greater insight into the earliest history of the Apoidea. The earliest apoid wasps are a series of plesiomorphic species classified in the family Angarosphecidae. Unfortunately, angarosphecids are defined only by plesiomorphies and thereby likely represent a grade leading to all other apoid families. This also strongly suggests that the assertion that this artificial family persisted into the Paleogene (based on Eosphecium from the Eocene of Quilchena, British Columbia; Pulawski et al., 2000) is erroneous since it is based solely on a forewing compression that preserves no derived features of any apoid family. As such, Eosphecium could just as likely represent a plesiomorphic species of Sphecidae (highly probable) or Crabronidae (less likely). This grade of plesiomorphic wasps is recorded from as long ago as the Early Cretaceous of Brazil, Europe, and Central Asia (Grimaldi and Engel, 2005). By the early epochs of the mid-Cretaceous, the families Ampulicidae and Crabronidae are definitively recorded from the Mesozoic, as evidenced by the exquisitely preserved Burmese amber fauna (Antropov, 2000b; Engel, personal obs.). Perhaps most importantly, the presence of crabronids, the most derived family of apoid wasps, immediately indicates that prior cladogenetic events in the superfamily (i.e., diversification of the families of apoid wasps) had taken place earlier in the Cretaceous. Furthermore, the occurrence of crabronids indicates that the bees, their sister group (Lomholdt, 1982; Alexander, 1992; Prentice, 1998; Melo, 1999; Grimaldi and Engel, 2005), had also already diverged by this time (Engel, 2001, 2004; Grimaldi and Engel, 2005).

Within the Crabronidae the nominate subfamily is presently documented only from the Tertiary, with most species in amber and most being of the tribe Trypoxylini (e.g., table 1). Nel (2005) enigmatically asserted that the Trypoxylini described by him from Parisian amber (Eocene) were the earliest records of Sphecidae. Aside from the attribution to the wrong family (prior to 1993 all apoid wasps were classified as one family “Sphecidae”), these are not older than fossils of Sphecidae (in the modern strict sense) or other fossils of Crabronidae, the family to which these actually belong (see table 1).

As mentioned previously, the moustache wasps (subtribe Crabronina) are known, like Trypoxylini, only from the Tertiary, ranging from the middle Eocene (i.e., Baltic amber species) through the Miocene (species documented herein). Prior associations of Lindenius with genera such as Entomognathus, Encopognathus, and Oxybelus (taxa outside of the Crabronina) indicate that Lindenius is likely basal or relatively basal within the subtribe, as is Tracheliodes. It is therefore interesting that species from these groups should make up most of the fossil record for the subtribe. The records of Ectemnius from Colorado and Germany are more intriguing, as the genus is apparently derived among the moustache wasps. However, characters defining Ectemnius were not preserved in these compression fossils, and attribution to the genus is exceedingly speculative, based almost entirely on general habitus (Cockerell, 1906, 1910; Meunier, 1911). As such, these records must be considered tentative and the types should be restudied before further conclusions based on these records are proposed. Overall, the sparse available records and their apparent phylogenetic position suggest an Early Tertiary (perhaps Paleocene?) origin of the Crabronina.

With regard to Lindenius in particular, extant species of the genus are nearly restricted to the Holarctic region; one species, Lindenius montezuma (Cameron), reaches Neotropical portions of southern Mexico (Amarante, 2002). It is notable that an extinct and perhaps basal member of the genus should be found outside of its modern range and amid exceptional habitat, which allows speculation that the modern diversity in the Mediterranean region may not be attributable to an origin there. Like its relative Tracheliodes, L. paleomystax suggests a relict aspect to the distribution and habitat preference of modern Lindenius. Clearly, the natural history of Lindenius is more complicated than that revealed by Recent taxa alone. Once cladistic work on the Crabronina has been completed, more insight on questions of moustache wasp origins and historical biogeography, particularly that of Lindenius, can be gained.