INTRODUCTION

As recently as 1995 the only species of bee known from the rich amber deposits of the Dominican Republic was the common stingless bee, Proplebeia dominicana (Wille and Chandler) (Wille and Chandler, 1964; Michener, 1982). During the intervening years, however, the total diversity and available material has risen significantly, with numerous species of xeromelissine, halictine, panurgine, megachiline, euglossine, and further meliponine bees documented (table 1).⁵ Although species have continued to accumulate, not since 1999 have any new tribes been recorded from these deposits. It is therefore of great significance that a new species of Apinae representing the tribe Exomalopsini was recently recognized among new material of Dominican amber.

Exomalopsines are restricted to the Western Hemisphere, ranging from the United States to Argentina, and are particularly abundant and diverse in arid habitats. They are small to moderate-sized anthophoriform bees, densely setose and commonly with pale metasomal bands of setae, and lacking maculation except in some species for the clypeus, labrum, and mandibles. The tribe in its broadest sense (Michener, 2007) comprises five genera (Anthophorula Cockerell, Chilimalopsis Toro, Eremapis Ogloblin, Exomalopsis Spinola, and Teratognatha Ogloblin) and approximately 150 modern species (table 2). Silveira (1995) developed a phylogenetic hypothesis for the tribe and excluded at that time Teratognatha and Chilimalopsis in their own tribe. However, given the weakness of some of the characters used by Silveira (1995) to separate Teratognathini, Michener (2007) considered this decision premature and regarded them as a subtribe. Although Exomalopsini are widely distributed in the Americas, all five exomalopsine genera occur in South America, but only a few lineages of the genera Anthophorula and Exomalopsis have reached and diversified in Central and North America. The two major genera are Exomalopsis and Anthophorula. The former ranges from the southernmost United States, the Bahamas, and the Antilles south throughout the Neotropical region to Argentina, whereas the latter, whose species have until recently been included in Exomalopsis, occurs from the central western United States (Oregon, Nebraska) south to southernmost Mexico, and seems to be especially abundant in xeric or mesic areas. Timberlake (1980), RoigAlsina (1992), Almeida and Silveira (1999), González-Vaquero and Roig-Alsina (2005), and Silveira and Almeida (2008) have revised most of the groups.

Herein we provide a brief description of the new fossil material as well as a brief discussion on exomalopsines and the Dominican amber melittofauna in general.

MATERIAL AND METHODS

The age and origin of Dominican amber are discussed by Grimaldi and Engel (2005), while the bee fauna was most recently summarized by Engel (2009), Greco et al. (2011), and Michez et al. (2012). For the descriptions morphological terminology generally follows that of Engel (2001) and Michener (2007). The description and all measurements provided are based on the holotype (figs. 1–4), with supplemental details taken from a second, fragmentary female preserved in the same amber piece. Measurements were prepared using an ocular micrometer on an Olympus SZX-12 stereomicroscope while photomicrographs were made using a Canon EOS 7D digital camera attached to an Infinity K-2 long-distance microscope lens. The amber piece is conserved in the amber fossil collection of the Division of Invertebrate Zoology, American Museum of Natural History, New York.

TABLE 1.

Hierarchical outline of the known Dominican amber bees (Apoidea: Anthophila), updated from Engel (2009). Dates and citations provided only for genus- and species-group taxa. Those for suprageneric taxa provided by Engel (2005).

continued

TABLE 2.

Summary of generic and subgeneric classification and modern diversity of Exomalopsini. NA = North America; CA = Central America; SA = South America; * = includes Caribbean Region. Dates and citations for family- and genus-group taxa provided by Engel (2005) and Michener (2007).

SYSTEMATIC PALEONTOLOGY

FIGURES 1–3.

Photomicrographs of holotype female of Anthophorula (Anthophorula) persephone Engel, new species (AMNH DR-KL1). 1. Dorsal oblique view of holotype female as preserved. 2. Facial view. 3. Detail of metabasitibial plate and branched scopal setae.

FIGURE 4.

Apical forewing venation of fragmentary female of Anthophorula (Anthophorula) persephone Engel, new species (AMNH DR-KL1); note that fracturing in wing as preserved distorts some proportions (holotype is preserved completely [fig. 1], but detailed images of the venation appear dark and without contrast against the darkened metasoma immediately beneath it).

DISCUSSION

It is exciting to record for the first time in many years a tribe new to the Dominican amber fauna. It is greatly hoped that additional tribes and genera, particularly one of the cleptoparasitic lineages that are present today, will be forthcoming from these deposits in time. The new fossil species can be recognized as an exomalopsine by the well-developed and plumose scopa (fig. 1), the clypeus rather flat in profile, the vertex not convex in facial view (fig. 2), the second abscissa of M + Cu in the hind wing over one-half as long as M, the lack of a paraocular carina, the presence of a line of fine setae along the inner orbit of the compound eye, and the forewing marginal cell obliquely truncate and bent away from the costal margin, its apex somewhat pointed and the total length of the cell greater than the distance from its apex to the wing tip (observable in the holotype and right forewing of the partial female [left forewing depicted in fig. 4]). Among exomalopsines the fossil most closely resembles the modern species of Anthophorula subgenus Anthophorula as evidenced by the combination of the second medial cell of the forewing shorter than the second cubital cell, the labrum and clypeus dark, the simple mandible (apparently so; difficult to discern in holotype), the metabasitarsus with an apical process and penicillus, the metabasitibial plate small and with a planar surface, the first metasomal tergum with a distinct transverse carina, a dense scope consisting of strongly plumose setae, and a pterostigma that is just about three times longer than the prestigma. However, it should be noted that A. persephone does differ from other species of the subgenus by the broader pterostigma (its inner breadth greater than the breadth of its marginal veins), in this regard resembling the subgenus Isomalopsis Michener and Moure (known only from xeric western Argentina) and many Exomalopsis Spinola. Indeed, the pterostigma, while shorter than in species of Isomalopsis (where it is four times the length of the prestigma), is at the upper end of the range of pterostigmal lengths for other Anthophorula s.l. Overall the broader pterostigma likely represents a plesiomorphy for the subgenus or genus as a whole. While a new subgenus could be proposed for this isolated species, to do so would serve no purpose at this time, particularly given that the sole, prominent difference from Anthophorula s.str. is such a minor trait, and one for which the polarity, as noted, likely indicates it as simply basal within the subgenus.

It is likely that Exomalopsini arose in semidesertic habitats in South America given that all five genera occur there while only a few lineages have reached and diversified in Central and North America; they are also particularly abundant and diverse in those types of habitats (Michener and Moure, 1957; Silveira, 1995). Although a northward migration was surely facilitated when the Isthmus of Panama was formed during the Pliocene, the presence of A. persephone in Dominican amber indicates that such a migration may have occurred much earlier throughout the Antilles, as documented for other taxa (Iturralde-Vinent and MacPhee, 1999; Engel, 1999a, 1999b, 2000). It also indicates that Anthophorula was also present in the West Indies and then became extinct with subsequent climatic and geographical changes, a phenomenon recorded in other bee taxa (Engel, 2004; Ohl and Engel, 2007). It is interesting to note that, except for the two South American species placed in the subgenus Isomalopsis (known only from xeric western Argentina), all Anthophorula are primarily found in North America. Such a distribution of the genus and the distinctive pterostigma of A. persephone described above supports the idea of the basal position of this extinct taxon within the subgenus. While it may seem somewhat peculiar for fossils of a group of largely xeric or mesic species to be discovered alongside or as close relatives of fossils from an otherwise tropical paleofauna, such relationships and patterns are known to exist (e.g., Bennett and Engel, 2006). Furthermore, species of Diomalopsis Michener and Moure and several of Exomalopsis s.str. are apparently restricted to moist tropical environments (Silveira, 1995), indicating ecological variation among species and the potential to colonize and occupy other environments.

The modern exomalopsine fauna of the West Indies is relatively poor. About six species are known, all of them in Exomalopsis s.str. (table 3). It is not known whether such colonization may have been from Central or South America. A species-level phylogenetic analysis, including A. persephone, could help explain the relationships as well as to clarify the historical biogeographical patterns of Exomalopsini. However, higher-level analyses are often preferred over species-level studies perhaps because of their apparent greater impact, but the latter are more informative and explanatory for addressing such biological questions. Understanding such patterns, particularly in the present case, requires testing of species circumscriptions (fortunately well established in Exomalopsini) and cladistic analyses at the specific level, and for which such biological and biogeographic transitions are taking place (e.g., Franz and Engel, 2010; Engel, 2011).

It is noteworthy that after more than 15 years of research, we have at this time a moderately robust perspective regarding the bee fauna of ancient Hispaniola. The modern fauna comprises around 75 species, 6.3% of which are introduced (Genaro, 2007), and by comparison the 21 species already recorded from the extinct fauna would appear to be a moderately decent sampling of the presumed total potential paleofauna. What is of interest is that while several elements of these faunas are quite similar, such as the shared presence of many tribes and genera (table 4), there are some just as stark differences. For example, the native corbiculate bee species in ancient Hispaniola, specifically the stingless bees and orchid bees, which today are not only absent in Hispaniola, but are either entirely or largely absent from the West Indies as a whole. During the complex tectonic history of the region over the last 19 million years it is no wonder that there was considerable faunal turnover. Clearly, lineages such as Euglossini, Meliponini, and the other bee tribes in Dominican amber had invaded the West Indies by the Early Miocene, most likely entering via South America. Most were extirpated as islands subsided and were again uplifted later (Donnelly, 1988; Iturralde-Vinent and MacPhee, 1999), and then the same genera or close relatives reinvaded the Caribbean, but this time from a myriad of sources, many coming again via South America, but with others clearly entering from Central and North America (e.g., Eickwort, 1988). Naturally, some may have persisted on one or more landmasses throughout the Miocene, Pliocene, and Quaternary, but evidence of this is lacking. If none managed to persist by moving from island to island as the West Indies adjusted in composition over the intervening 19 million years, then is interesting to note the similarities, at least at the generic/subgeneric level, between the extinct and modern faunas of Hispaniola (table 4). In order to more fully elucidate the biogeography of the region and the implications of Dominican amber taxa, more cladistic analyses are needed of species within particular genera, including the fossils. Only such treatments will refine hypotheses as to the sources of the particular Dominican amber elements. In addition, it shall be exciting to discover the degree to which similarities between the faunas continue to accumulate as we undoubtedly recover more species from the mines of the Dominican Republic.

TABLE 3.

List of species of Exomalopsis known from the West Indies. All species belong to the nominate subgenus. Distributions based on Genaro (2007, 2008), Genaro and Franz (2008), and Silveira (2007).

TABLE 4.

Comparison of the living and extinct faunas of bees from Hispaniola (only native genera and subgenera included: see Genaro, 2007). With about 75 species living on the island today (and at least 6.3% of these are introduced, and only 45.6% of the total are endemic), the 21 species from Dominican amber represents a relatively robust sampling of the potential fossil fauna.

continued

continued