Synopsis of the cyclocephaline scarab beetles (Coleoptera, Scarabaeidae, Dynastinae)

Abstract The cyclocephaline scarabs (Scarabaeidae: Dynastinae: Cyclocephalini) are a speciose tribe of beetles that include species that are ecologically and economically important as pollinators and pests of agriculture and turf. We provide an overview and synopsis of the 14 genera of Cyclocephalini that includes information on: 1) the taxonomic and nomenclatural history of the group; 2) diagnosis and identification of immature life-stages; 3) economic importance in agroecosystems; 4) natural enemies of these beetles; 5) use as food by humans; 6) the importance of adults as pollination mutualists; 7) fossil cyclocephalines and the evolution of the group; 8) generic-level identification of adults. We provide an expanded identification key to genera of world Cyclocephalini and diagnoses for each genus. Character illustrations and generic-level distribution maps are provided along with discussions on the relationships of the tribe’s genera.


Introduction
The cyclocephaline scarabs (Coleoptera: Scarabaeidae: Dynastinae) are remarkable among rhinoceros beetles for the group's immense species richness and ecological importance. Cyclocephalini is a pan-tropical tribe with several genera considered to be keystone pollinators in New and Old World tropical ecosystems. By one estimate, pollination mutualisms between cyclocephalines and early-diverging angiosperms suggest that nearly 900 species of Neotropical plants rely upon these scarab beetles for sexual reproduction (Schatz 1990). Beyond tropical forests, cyclocephaline scarab beetle species are important to human industry as pests in tropical and temperate agroecosystems and turfgrass in North America. Due to these factors, the group has received considerable alphataxonomic attention as species identity (and identification) is crucial for understanding the fascinating biology of these scarabs. However, almost nothing is known about the evolution of the group into their incredible ecological roles.
This paper synthesizes all available information on cyclocephaline scarab beetles into these broad categories: 1) taxonomic and nomenclatural history of the group organized by major worker, including an exegesis of Endrődi's German-language revision of the tribe; 2) state of knowledge surrounding diagnosis and identification of immature life-stages; 3) economic importance in agroecosystems; 4) natural enemies of these beetles; 5) use as food by humans; 6) importance of adults as pollination mutualists; 7) knowledge of the fossil record and evolution; and 8) an overview of each genus, including expanded diagnoses and a key to world genera of Cyclocephalini.

Carl Linnaeus and his students
The taxonomic and nomenclatural history of Cyclocephalini traces to the works of Carl Linnaeus and several of his students. The 12 th edition of Systema Naturae included the description of Scarabaeus amazonus Linnaeus, 1767, which was later designated as the type species of Cyclocephala Dejean (Linnaeus 1767, Casey 1915, Endrődi 1966. This was the only cyclocephaline species described by Linnaeus. The short Latin description of S. amazonus indicated that this beetle was from "Suriname," was smaller than many dung beetles (with a relatively shorter pronotum), and was testaceous with longitudinal, black stripes (Linnaeus 1767). Unfortunately, the type specimen of S. amazonus is apparently lost. A serious effort to find this Linnaean type was undertaken by Sebő Endrődi and fellow Coleopterist Bengt-Oloft Landin.
Landin, an expert in Linnaean scarabaeoid types (e.g., see Landin 1956), was in correspondence with Endrődi during the early phases of the latter's revisionary works (Endrődi 1966). They determined that the type specimen of S. amazonus was not present in any of the museums that housed parts of the Linnaeus beetle collection: The De Geer collection at Naturhistoriska Riksmuseet (Stockholm, Sweden), Uppsala University, Museum of Evolution, Zoology Section (Uppsala, Sweden), and The Natural History Museum (London, United Kingdom). In a personal correspondence with Endrődi, Landin speculated that the specimen that became the type of S. amazonus was passed from Daniel Rolander (an apostle of Linnaeus sent to Suriname), then to Baron Charles De Geer, and eventually to Linnaeus (Endrődi 1966). Dejean (1821) authored the genus Cyclocephala in the first edition of the catalog of his collection. There was longstanding confusion in the literature surrounding the proper authorship of the genus Cyclocephala, with most historical workers crediting the genus to Latreille (1829) (e.g., Arrow [1937b], Blackwelder [1944], and Endrődi [1966Endrődi [ , 1985a). This confusion stemmed from Dejean's practice of proposing new genera without describing them in the catalogs of his collection (Bousquet and Bouchard 2013a, b). Dejean (1821) also attributed authorship to other workers who had applied names to species in their own collections, but before the names were formally described in the literature. Thus, subsequent authors treated Dejean's new genera and species as invalid nomina nuda. However, because Dejean (1821) included one or more available species-group names in Cyclocephala, the genus-group name became available from that work (ICZN Article 12.2.5; see Bousquet and Bouchard 2013a for further discussion).
Cyclocephala was first described and illustrated by Latreille (1829Latreille ( , 1837. Latreille's (1829) short description of Cyclocephala utilized characters of the protarsal claws (unequal in size and cleft at the apex), labrum (visible anteriorly), body shape (ovoid with the head uncovered), elytra (weakly edged without significant lateral dilation), and mandibles (narrow, not strongly produced beyond clypeus, without a lateral sinus, and variably toothed). The genus was also considered variable enough to warrant subgeneric division into Chalepus and Cyclocephala (Latreille 1829). Figure plates illustrated a dorsal habitus of Cyclocephala frontalis Chevrolat, 1844 and the anatomy of the head, labrum, maxilla, and protarsus of Cyclocephala geminata (Fabricius) (=Dyscinetus dubius [Olivier]) (Latreille 1837). These illustrations are some of the earliest scientific depictions of the group. Laporte (1840) was the first author to propose a tribal-level taxon for the cyclocephaline scarab beetles. This group, Cyclocephalites, was included along with Dynastites and Rutélites in the family Xylophiles (Laporte 1840). Cyclocephalites was not originally proposed in a Latinized form (see Smith 2006, Bouchard et al. 2011). However, because the name was subsequently Latinized by several authors (e.g., Cyclocephalidae by Burmeister [1847] and Imhoff [1856], and Cyclocephalinae by Bates [1888]) and was generally accepted, the family-group name is available from this work per ICZN Article 11.7.2. Cyclocephalites sensu Laporte (1840) was diagnosed by having the mandibles mostly covered by the clypeus and the labrum not extending anteriorly beyond the apex of the clypeus. Laporte included two divisions in Cyclocephalites. The first division, diagnosed by arched and hooked mandibles, included only Cyclocephala geminata (Fabricius, 1801) (=Dyscinetus dubius [Olivier, 1789]). The second division of Cyclocephala was diagnosed by having straight, truncate, or obtuse mandibular apices (Laporte 1840). This second division contained six species, and these are still classified in Cyclocephala.

Hermann Burmeister
The German naturalist and entomologist Karl Hermann Konrad Burmeister made major contributions to dynastine scarab research in the mid-19 th century (Berg 1894). Burmeister's (1844Burmeister's ( , 1847Burmeister's ( , 1855 Handbuch der Entomologie volumes systematically organized a large portion of Scarabaeoidea. Burmeister (1847) was one of the first authors to unite members of the subfamily Dynastinae, nearly as currently circumscribed, into a single family and recognizable tribes in the modern sense. This family, Xylophila, was subdivided into Cyclocephalidae, Phileuridae, Dynastidae, Agaocephalidae, Strategidae, Oryctidae, and Xylophila amphibola (=Scarabaeidae: Cetoniinae: Trichiini, in part) (Burmeister 1847). Seven of the genera included in Burmeister's Cyclocephalidae are still part of Cyclocephalini (Table 1). Additionally, Burmeister described five new genera and 71 species-group taxa (56 of which are valid species or subspecies) that are still included in Cyclocephalini.

Henry Walter Bates
Famous English naturalist Henry Walter Bates treated cyclocephalines in his contributions to the scientific opus Biologia Centrali-Americana and Edward Whymper's Travels Amongst the Great Andes of the Equator (Bates 1888(Bates , 1891. Between these two works, Bates covered over 50 cyclocephaline species-level taxa, described nearly 30 new species (20 of which are still accepted as valid), and contributed to the generic-level classification of the group. For example, he recognized the distinctiveness of Ancognatha Erichson 1847 and revalidated the genus, which had been synonymized with Cyclocephala (Erichson 1847, Lacordaire 1856, Bates 1888. He described two new cyclocephaline genera: Aspidolea Bates 1888 and the eventual junior synonym Barotheus Bates, 1891 (=Ancognatha Erichson).
Following Lacordaire's (1856) system, Bates classified the cyclocephaline scarab beetles as a subfamily (Cyclocephalinae) within Dynastidae. He only provided diagnoses for two higher groups (what he called "subtribes" within Lamellicornia) based upon labial morphology. Thus, Bates did not propose a character-based circumscription of the cyclocephaline scarabs or dynastines more broadly. However, some of the earliest detailed discussion and comparison of generic-level diagnostic characters among cyclocephalines can be found in Biologia Centrali-Americana (Bates 1888). For example, the toothless (or nearly toothless) maxillary galeae of Aspidolea and Ancognatha were recognized as providing partial justification for accepting these genera as being distinct from Cyclocephala (Bates 1888). Bates (1888) divided Cyclocephala into a series of informal species-groups. For example, group I, which contained C. signata Fabricius (=C. amazona) was diagnosed by: 1) an elongated or protracted clypeus; 2) the clypeal apex sometimes bent at the margin; and, 3) the apex of the ligula deeply divided and widely splayed (Bates 1888). Similar diagnoses that relied upon a combination of clypeal and labial morphology were provided for five major Cyclocephala species-groups. Sexual dimorphism of the antennal club (elongated in males) was used to further subdivide one of these species-groups (Bates 1888). Bates also covered the cyclocephaline genera Dyscinetus and Stenocrates. With less available material, he was unable to make many meaningful character comparisons for these genera. However, he did mention that the dorsoventrally flattened tibiae of Stenocrates serve to diagnose that genus (Bates 1888).

Thomas Lincoln Casey, Jr.
Lieutenant Colonel Thomas Casey's major contribution to scarabaeology was the sixth volume of Memoirs on the Coleoptera (Casey 1915). This volume covered Cetoniinae, Rutelinae, and Dynastinae of Central and North America. It provided keys to tribes, genera, and species, reported distributional data, and served as an outlet for the description of many new taxa. Casey (1909Casey ( , 1915 treated Cyclocephalini as a tribe of Dynastinae, and he was the first Coleopterist to propose extensive generic-level reorganization of the tribe and the genus Cyclocephala. Most of Casey's new taxa (genera, species, and subspecies) in Cyclocephalini were not accepted as valid by subsequent workers. For example, Casey described over 60 new species and subspecies of cyclocephaline scarabs. Only seven of these taxa are currently accepted as valid. Casey (1915) proposed 16 new genera and subgenera in Cyclocephalini, among which only Chalepides Casey is currently in use (Table 2). Casey (1915) was the first author to definitively place Anoplocephalus Schaeffer, 1906(=Coscincocephalus Prell, 1936 in Cyclocephalini.

Gilbert John Arrow
English entomologist Gilbert Arrow was notable among early 20 th century workers for his global knowledge of Dynastinae and Rutelinae. Arrow's work in The Natural History Museum allowed him to meaningfully compare characters between diverse New and Old World taxa. For example, the genus Peltonotus (considered by most authors to be a cyclocephaline since Burmeister) was transferred into Rutelinae based on the form of the labrum (chitinized apically and projected anteriorly beyond the apex of the clypeus), which it shares with several Asian, parastasiine-like genera (Arrow 1908(Arrow , 1910. Arrow (1908) described the Afrotropical cyclocephaline genus Ruteloryctes, which he compared to the New World genus Dyscinetus.
Cyclocephalines, as currently circumscribed, were covered in 11 of Arrow's publications (Arrow 1900(Arrow , 1902(Arrow , 1903(Arrow , 1908(Arrow , 1910(Arrow , 1911(Arrow , 1913(Arrow , 1914(Arrow , 1931(Arrow , 1937a. Arrow described over 40 new species or subspecies of cyclocephalines, and most of these were in the genus Cyclocephala. An early critic of Casey's (1915) genus and species concepts, Arrow (1937a) argued that many of Casey's new dynastine taxa created unnecessary "disorder" in Cyclocephalini and the subfamily more broadly. Arrow attributed this upheaval to Casey's ignorance of species that invalidated his generic diagnoses. For example, Arrow criticized Casey's overreliance on geographic separation of taxa and his intolerance for intraspecific variation, specimen wear, and recognition of teratological forms as distinct taxa. Arrow (1937b) published the first comprehensive catalog of Dynastinae since Gemminger and Harold's Catalogus Coleopterorum (see Harold 1869b). By Arrow's admission, incorporating Casey's cyclocephaline taxa into this catalog was challenging. Arrow struggled to place most species within Casey's (1915) generic and subgeneric framework or assign synonymy to many species. He generally listed Casey's higher taxa as subgeneric-level synonyms within Cyclocephala (Arrow 1937a, b). Mimeoma was accepted by Arrow (1937b), and he included a second species in the genus. Chalepides was also accepted as valid, and he elevated the subgenus to genus status (Arrow 1937a, b). Arrow expanded the composition of Cyclocephalini (Table 3) to include several Australasian genera that were later transferred to Oryctoderini (Scarabaeidae: Dynastinae) (Endrődi 1966(Endrődi , 1971a. Some of these Australasian genera had been placed into Cyclocephalini at the time of their description (e.g., Chalcocrates Heller, 1903).

Lawrence Saylor
American entomologist Lawrence Saylor authored five publications (Saylor 1936(Saylor , 1937(Saylor , 1945(Saylor , 1946(Saylor , 1948) that included cyclocephaline scarab beetles, especially focusing on North American species. Saylor's publications were very important for the time because they offered high-quality diagnoses, keys, and illustrations for species of Ancognatha, Cyclocephala, Dyscinetus, and Erioscelis. Saylor's approach and implied species concept arguably influenced Endrődi's revision of the tribe (see Ratcliffe 2016 for further discussion). Saylor's role was not as a describer of new species in the group, but rather as a primary reviser of many North American dynastine taxa that had been neglected since the works of John Lawrence LeConte (1854,1861,1862,1863,1866) and George Henry Horn (1871Horn ( , 1875Horn ( , 1894 and further obfuscated by Casey (1915). The problem of Casey's numerous cyclocephaline synonyms also fell firmly on Saylor. Saylor (1937 synonymized over 30 of Casey's taxa in Cyclocephala and Dyscinetus, which created more reliable and precise diagnoses of North American species in these genera.

Sebő Endrődi
The Hungarian Sebő Endrődi, a lawyer by formal training, was the most prolific and important dynastine worker of the 20 th century (Kaszab and Papp 1986). Endrődi, a scarabaeoid beetle specialist, was the principal author of over 200 scientific articles and books on beetle systematics (Kaszab and Papp 1986). In the post-World War II period, Endrődi vigorously undertook a world revision of the subfamily Dynastinae. These revisionary studies, the "Monographie der Dynastinae", were published from 1966 through 1978 as a 22-part series. The series was later translated into English, synthesized, and published as a single volume, The Dynastinae of the World (Endrődi 1985a). Endrődi's revisions (both the more detailed German-language series and the English-language book) are the basis of modern dynastine systematics research and identification. Endrődi authored or coauthored 27 works that covered cyclocephaline scarabs from 1960 to 1985 (Endrődi 1960, 1963, 1967a-c, 1969a, b, 1970, 1971b, 1973a, b, 1975a-c, 1977a, b, 1979, 1985a, b, Howden and Endrődi 1966, Endrődi and Dechambre 1976, Dechambre and Endrődi 1983. In total, Endrődi named over 110 species and subspecies (>90 of these taxa are still valid) in cyclocephaline genera. The majority (~50% of valid taxa) of these new taxa were described in the speciose genus Cyclocephala. Generally, Endrődi did not describe new genera in this group (the junior synonym Surutoides Endrődi 1981 is the lone exception) and instead favored lumping species into relatively large genera (e.g., Cyclocephala sensu Endrődi 1966 included over 180 taxa). The tribe Cyclocephalini was covered in the first installment of the "Monographie der Dynastinae" series (Endrődi 1966). One of the earliest modern discussions on the phylogenetic position of Cyclocephalini, and Dynastinae more broadly, was included in this first installment (Endrődi 1966). Many of the most detailed portions in the German-language monograph of Cyclocephalini (Endrődi 1966) were not included in The Dynastinae of the World and these details warrant further discussion.
Cyclocephalini was considered by Endrődi to be the most primitive tribe of Dynastinae, with many species sharing characters with Rutelinae (Endrődi 1960(Endrődi , 1966Fig. 1). Endrődi's (1966) methodology for assessing the relationships of dynastine tribes defies precise categorization within modern approaches. He attempted, with poor justification, to polarize a suite of characters into primitive and derived states within Dynastinae. Nine characters were scored as three states, which ranged from 1 (most derived) to 3 (ancestral). Character states scored as "2" indicated that both derived and ancestral states, or "partially differentiated" states, were present in each tribe (Endrődi 1966). Tribes with the highest total numerical value (numbers were summed across the matrix) were considered the most primitive overall.
This analysis suggests that Endrődi was attempting a very rudimentary cladistic approach to understanding dynastine tribal relationships. However, he did not define clear synapomorphic characters nor did he discuss homoplasy. This rudimentary approach was used only to hypothesize how "evolved" each of the eight dynastine tribes were compared to the outgroup Rutelinae. His results indicated the Cyclocephalini (score of 25) was the earliest diverging dynastine tribe, while Dynastini (score of 16) was the most derived tribe. Endrődi (1966) utilized the following characters in this analysis: 1) "body form" differentiated from Rutelinae or not; 2) presence or absence of "striking" sexual dimorphism; 3) relative length of the legs; 4) relative thickness of the protarsomeres and protarsal claws in males; 5) form of the anterior margin of the meso-and metatibia ("Hinterschienenspitze"); 6) presence or absence of stridulatory structures on the abdomen; 7) relative degree of expansion of the female elytral epipleuron; 8) presence or absence of hindwings; and 9) global distribution.
Endrődi also considered relationships among genera. A similar character polarization method was applied to cyclocephaline genera that were considered by Endrődi as valid (Endrődi 1966). Nine characters were used in this analysis: 1) clypeus short and simple to strongly differentiated; 2) lamellate club of the antennae elongated in males or not; 3) male protarsomeres thickened or not; 4) body shape vaulted, oval, or dif-  Endrődi (1966). Hypothetical relationships among the tribes of Dynastinae. "Primitive form" is translated from the German "Urform." ferentiated; 5) male parameres simple or differentiated; 6) prosternal peg short or elongated; 7) clypeus with or without bumps ("Höckern"); 8) elytral punctation disorganized, unistriate, or in paired striae; and 9) female elytral epipleuron strongly thickened or not. Augoderia, Arriguttia, and Ruteloryctes were thought to be the most "primitive" cyclocephaline genera, though Endrődi's analysis provided only weak justification. By Endrődi's (1966) own admission, this exercise did not yield clear results ("Aus diesen Wertzahlen ist deutlich zu erkennen, daß schon bei den Gattungen die Auswertung der primitiven und fortgeschritten Formen nur schwer vorgenommen werden kann"). Endrődi's (1966) diagnosis of Cyclocephalini is the most detailed published for the group, and it offers further discussion on the distribution of some character states among the tribe's genera. Members of Cyclocephalini were diagnosed as being smallto medium-sized, primitive dynastines that share the oval and convex body shape of Rutelinae. The body shapes of the genera Arriguttia (anteroposteriorly compressed) and Surutu (dorsoventrally flattened) were considered exceptional in the tribe. Cyclocephaline mandibles were considered small (varying in width or broadness) for the subfamily and lacking teeth on the lateral, outer margin. Cephalic morphology in the tribe was notable for its lack of horns, tubercles, carinae, or sulci. The slightly raised frontoclypeal suture present in some Ancognatha species was a possible exception to this lack of armature on the head. These "tubercles", however, were not considered homologous with tubercles of the head present in other dynastines (Endrődi 1966).
Cyclocephaline antennae are comprised of 8-10 antennomeres with the lamellate club always three-segmented and occasionally elongated in males (Endrődi 1966). The pronotum is convex and only dorsoventrally flattened in Surutu, while the scutellum is triangular. The elytra are usually 1.5 times longer than wide and are rarely shorter (e.g., Arriguttia). Elytral punctation is regularly spaced and paired when punctures form striae (except for Augoderia and Surutu). The females of many species have pronounced expansions of the elytral epipleural margin with or without produced lateral flanges (Endrődi 1966).
The propygidium of cyclocephalines lacks a stridulatory apparatus (Endrődi 1966). Pygidial morphology varies between the group's genera. The pygidium is reduced in Chalepides, while it is a large segment in all other cyclocephaline genera. The prosternal process is relatively long and generally rounded at the apex, but has a variably present or absent button-like folding of the cuticle (Endrődi 1966). Protibial morphology in the tribe is also highly variable. The outer lateral margins of the protibia in males have 1-3 produced teeth, while most genera have no teeth on the inner lateral margin of the protibia. Harposceles is the lone exception for the tribe, having a small tooth on the inner margin of the protibia (Endrődi 1966).
Three genera included in Cyclocephalini sensu Endrődi (1966) lack thickened, foreshortened protarsomeres and enlarged (and sometimes cleft) protarsal claws in males: Erioscelis, Stenocrates, and Coscinocephalus. The meso-and metatarsomeres are not thickened and foreshortened in any cyclocephaline genera (though metatarsomeres are reduced in females of some Cyclocephala species). The apical margins of the meso-and metatibia are simple in cyclocephalines, lacking crenulated extensions ("Hinterschienenspitze fast immer gefingert") (Endrődi 1966). Cyclocephalini sensu Endrődi (1966Endrődi ( , 1985a included 14 genera and was a strictly New World tribe, except for the Afrotropical genus Ruteloryctes (Table 4). Endrődi (1966) considered the criteria for defining genera like those used to define families. Within this concept, genera were phylogenetic units that needed to show several characteristics in a "constant state" to be valid (Endrődi 1966). This line of argumentation was extended into a criticism of several genera and subgenera proposed within Cyclocephalini. Casey's generic-level hypotheses in Cyclocephalini were especially in violation of this guiding principle. Endrődi considered most of Casey's genera as based upon only a single character and were thus invalid within his paradigm. It was also argued that Casey's subgenera were based upon species-level characters and not applicable to higher-level classification schemes (Endrődi 1966). This led to the synonymy of nearly all of Casey's higher-level cyclocephaline groups, some of which were tentatively adopted by other authors in the intervening period (e.g., Arrow 1937a, b, Saylor 1937, and Buchanan 1927 (Casey 1915, Endrődi 1966. The subgenus Cyclocephala (Paraclinidia) was ambiguously synonymized within Cyclocephala, and Endrődi (1966) commented that the group could "at most be considered a subgenus." An explanation of some aspects of Endrődi's (1966) morphological approach to his revision of Cyclocephalini was provided in a section entitled "Morphologie der Tribus." Three types of coloration schemes are found in the tribe: 1) species that are all black or dark brown, except in teneral specimens (e.g., Surutu, Harposceles, Coscinocephalus, Erioscelis, Ruteloryctes, Stenocrates, Dyscinetus, Chalepides, and occasionally other genera); 2) species that are monotoned and light in color, sometimes with darkened legs and head, and lacking dorsal maculae (e.g., Cyclocephala and Aspidolea); and 3) species with red or black dorsal maculae (e.g., Augoderia, Ancognatha, and Cyclocephala) (Endrődi 1966). Among species with dorsal maculae, Endrődi considered these characteristics to be highly variable within a "system" of patterning that displayed some species-level specificity. Some species vary from having elaborate dorsal maculae to being nearly free of patterning, and these species were the most challenging for precise identification (Endrődi 1966).  Endrődi (1966Endrődi ( , 1985a.

Genera
Biogeographic Realm Current Tribal Classification Ancognatha Erichson, 1847 Neotropical and Nearctic Dynastinae: Cyclocephalini Arriguttia Martínez, 1960  Short or long setae on the head and thorax were useful characters for diagnosing species. Endrődi thought that setae on the frons and anterolateral margins of the pronotum were particularly easy to observe (even when eroded) because they were erect and in obvious punctures. The shape of the clypeus, important since Burmeister (1847), was considered diagnostic in Mimeoma, Ancognatha, Stenocrates, and Aspidolea (Endrődi 1966). However, clypeal shape was considered too variable among species for diagnosing groups in other genera such as Cyclocephala (Endrődi 1966). Sculpturing and rugosity of the frons, interocular distance, and shape of the frontoclypeal suture were considered stable characters within species (Endrődi 1966). He noted that there is significant variation of the mouthparts (labrum, ligula, maxillae, and mandibles) among cyclocephalines and observed this variation mostly from dissected Burmeister type specimens. Due to the number of species and specimens he needed to examine, Endrődi eschewed characters that required dissection (except for male genitalia) to observe. Thus, he generally did not use mouthpart or hindwing characters in his diagnoses for genera or species. The usefulness of mouthpart and hindwing characters for circumscribing groups remains largely unevaluated in Cyclocephalini and Dynastinae.

Late 20 th and early 21 st century French workers: Roger-Paul Dechambre, Fabien Dupuis, and Fortuné Chalumeau
Dynastine scarab enthusiast Roger-Paul Dechambre, a former curator of Coleoptera at Museum National d'Histoire Naturelle in Paris, published 21 papers or book chapters on Cyclocephalini (Dechambre 1979a-c, 1980, 1991a, b, 1992, 1995, 2006a, b, Dechambre and Duranton 2005, Dechambre and Endrődi 1983, Dechambre and Hardy 2004, Dupuis and Dechambre 1995, Ponchel and Dechambre 2003. Dechambre was a prolific describer of cyclocephaline taxa, having authored or coauthored over 80 species and subspecies in the group (only five of which are currently junior synonyms). Most of these taxa were described in Cyclocephala (65 species and subspecies) and Stenocrates Burmeister (15 species). Beyond his Cyclocephala expertise, he described the second species of the African genus Ruteloryctes (Dechambre 2006b), a species of Chalepides (Ponchel and Dechambre 2003), a species of Ancognatha (Dechambre 2000), and three species of Aspidolea (Dechambre 1992). Nearly all of Dechambre's new cyclocephaline taxa are South American, which highlights the need for continued work on that fauna.
Dechambre's treatment of cyclocephaline genera was conservative. Dechambre did not describe any new cyclocephaline genera, and he synonymized Surutoides with Cyclocephala (Dechambre 1991a). Dechambre seems to have favored treating "species groups" in lieu of upsetting the classification of Cyclocephala. For example, Dechambre (1997) revised the "Cyclocephala cribrata species group" which included the relatively large, black species of Cyclocephala previously included in Mononidia and Surutoides.
Late 20 th and early 21 st century North, Central, and South American workers: Brett Ratcliffe, Ronald Cave, Luis Joly, and Mary Liz Jameson Brett Ratcliffe, Curator of Entomology at the University of Nebraska State Museum, greatly expanded upon Endrődi's dynastine research in the Nearctic and Neotropical realms. Ratcliffe has authored or coauthored 39 publications that cover cyclocephaline scarabs, and many of these are monographic in scope (Ratcliffe 1977, 1978, 1992a-d, 2002a, b, 2003, Ratcliffe and Cave 2002, Ratcliffe et al. 2013, Ratcliffe and Delgado-Castillo 1990, Ratcliffe and Hoffman 2011, Ratcliffe and Morón 1997, Ratcliffe and Paulsen 2008, Figueroa and Ratcliffe 2016, Gasca-Álvarez et al. 2014, Jameson et al. 2002, Maes and Ratcliffe 1996, Maes et al. 1997, Neita-Moreno et al. 2006, Saltin and Ratcliffe 2012. This body of research includes the description of over 60 new cyclocephaline species, only eight of which are in synonymy. These publications are mostly focused on Central or Mesoamerican taxa, but they also enhance knowledge of the poorly known South American genera Surutu and Harposceles. Ratcliffe, with collaborators Ronald Cave and Enio Cano, have systematically treated Dynastinae north of Panama, including the West Indies (Ratcliffe 2003, Ratcliffe and Cave 2006, Ratcliffe et al. 2013). These monumental works provide the most comprehensive, authoritative taxonomic treatment (synonymy and consistent species concept), identification tools, distribution data, and synthesized biological information ever produced for the subfamily in the New World. Venezuelan scarabaeologist Luis Joly, along with collaborator Hermes Escalona, advanced understanding of the group in South America, having revised Chalepides and the Dyscinetus of Venezuela Escalona 2002, 2010). Joly has also described several new species of Cyclocephala from across South America and the West Indies.
Recent publications have generally been conservative regarding the generic composition of Cyclocephalini. Morón and Ratcliffe (1996) transferred the genus Coscinocephalus from Cyclocephalini to Pentodontini based on characters of the head, mouthparts, and parameres shared with Orizabus Fairmaire, 1878. The work of Mary Liz Jameson, while focused mainly on the subfamily Rutelinae, has altered the concept of Cyclocephalini (Jameson 1998, Jameson et al. 2002, Jameson and Wada 2004, Jameson and Jákl 2010, Jameson and Drumont 2013. Two genera, Acrobolbia Ohaus, 1912 and Peltonotus, previously classified in Rutelinae were transferred into Cyclocephalini based on morphological phylogenetic analyses (Jameson 1998, Jameson et al. 2002, Jameson and Wada 2004.

Immature stages: diagnosis and identification
Research interest in cyclocephaline immature stages has recently increased, with approximately 80% of larval and pupal descriptions published after 1990 (Morelli 1991, Morelli and Alzugaray 1994, Vincini et al. 2000, Ramírez-Salinas et al. 2004, Vallejo and Morón 2008, Neita-Moreno and Morón 2008, Bran et al. 2006, Neita-Moreno et al. 2007, Vallejo and Morón 2008, Neita-Moreno and Morón 2008, Lugo-García et al. 2009, Stechauner-Rohringer and Pardo-Locarno 2010, Neita-Moreno and Yepes 2011, Albuquerque et al. 2014, Souza et al. 2014a, b, Morón et al. 2014). It is not yet possible to characterize cyclocephaline larvae or pupae at the tribal level as only 4 of 14 genera have described immatures (Table 5 and Table 6). Neita-Moreno et al. (2007) offered the most detailed tribal-level diagnosis of third-instar larvae and noted all the species known to them shared the following characters: 1) dorsal surface of last antennal segment with two sensory spots and 2) each tarsal claw with two setae. Characters  Martínez Ritcher 1944, Gordon and Anderson 1981, King 1984, Morelli 1989, Morelli and Alzugaray 1994, Bran et al. 2006 (Fabricius) D. rugifrons (Burmeister) of the haptomeral process (epipharynx), plegmatia (epipharynx), ocelli (head), and raster palidia (abdomen) were consistent in many, but not all, known species at the time (Neita-Moreno et al. 2007, Morón et al. 2014. Eleven additional species of Ancognatha, Cyclocephala, and Dyscinetus had their larvae described since Neita-Morena et al. (2007), and these authors' diagnosis for the tribe should be reevaluated with the data presented in Table 7. The presence of two dorsal sensory spots on the terminal antennal segment is a consistent character for the tribe, except for C. barrerai (Morón et al. 2014) (Table 7). Cyclocephala barrerai has a variably present or absent third dorsal sensory spot on the terminal antennomere (Morón et al. 2014). The tarsal claws of known cyclocephaline larvae have two setae (one basal seta and one prebasal seta). Cyclocephala celata is the exception in the tribe, and this species has an additional prebasal seta (Souza et al. 2014b). The haptomerum of the epipharynx has a raised bilobed or entire ridge in the subfamily Dynastinae (Ritcher 1966). Among the known Cyclocephala and Aspidolea larvae (the genera with the most similar adult morphology that are comparable), the haptomerum is a tooth-like process that is divided into two lobes (or "teeth") ( Table 7). This character may prove useful for diagnosing larvae of Cyclocephalalike genera in the tribe if they are described in the future (e.g., Arriguttia, Augoderia, former Mimeoma species, and additional Cyclocephala species). Ancognatha manca has an entire haptomeral process, making it unique for the known larvae in the genus.
Several identification keys incorporating these species have been developed. For example, Lugo-García et al. (2009 proposed an identification key for all species of phytophagous scarab larvae (including Cyclocephala) associated with agave and maize cultivation in Jalisco and Sinaloa, Mexico. Country specific keys for Cyclocephala larvae were developed for Uruguay and Colombia (Morelli and Alzugaray 1994, Bran et al. 2006, Stechauner-Rohringer and Pardo-Locarno 2010. Neita-Moreno et al. (2007) proposed a generic-level key to the tribe that included Ancognatha, Aspidolea, Cyclocephala, and Dyscinetus. Neita-Moreno and Yepes (2011) provided a key to the larvae of Dyscinetus and several authors have proposed keys to the known larvae of Cyclocephala (Souza et al. 2014a, b, Albuquerque et al. 2014). The four new larval descriptions   (2008) provided a key to the known larvae of Ancognatha.

Economic importance of larvae and adults
The habits of cyclocephaline larvae are poorly known, especially for species that are restricted to tropical forests. Species commonly encountered in temperate zones or agricultural areas are the source of the most detailed larval life history data. Cyclocephaline larvae go through three instars and pupate in soil (Ritcher 1966, Santos and Ávila 2007, Stechauner-Rohringer and Pardo-Locarno 2010, Rodrigues et al. 2010, Souza et al. 2015. Economic data from turfgrass researchers suggested that the larvae of tem-perate Cyclocephala species are rhizophagous (e.g., see Blanco-Montero and Ward 1995 and Crutchfield et al. 1995). Data from Central and South American agroecosystems indicated that Cyclocephala larvae are at least facultatively saprophagous, feeding on decaying plant matter and leaf litter. Information about immature stages in tropical forests is sparse, but the larvae and pupae of Harposceles paradoxus were found in the organic litter accumulated between leaf sheaths of the palm Astrocaryum carnosum F. Kahn & B. Millán (Arecaceae) (Couturier and Kahn 1992). Cyclocephala cribrata Burmeister larvae reportedly eat the roots of bromeliads in Brazil (Luederwaldt 1926). Cyclocephala atricapilla Mannerheim adults and larvae were found beneath litter near their Annona host plants, and the larvae were observed feeding on decaying material (Costa et al. 2017). The economic importance of Cyclocephala larvae in agroecosystems is difficult to generalize as beneficial, negative, or neutral. The widespread species C. lunulata has been laboratory reared on decaying sugarcane and humus, indicating some saprophagous habits (Stechauner-Rohringer and Pardo-Locarno 2010). In agroecosystems, C.  Morón et al. 2014). However, this species is not thought to be a major damaging pest in crop systems (Aragón et al. 2001).
In contrast, C. parallela larvae are considered a pest in Florida "sand-muck" sugarcane production (Gordon and Anderson 1981). Sugarcane production may produce favorable soil conditions for cyclocephaline scarab beetle larvae as Cyclocephala and Dyscinetus species have been reported to be numerous in fields in Cuba, Puerto Rico, Nicaragua, Colombia, and Guyana (Box 1925, Stahl and Scaramuzza 1929, Squire 1932, Maes and Tellez 1988, Posada Ochoa 1989. Cyclocephala testacea can reach densities of 160 larvae/m 2 of soil in Uruguayan pastures (Morelli and Alzugaray 1994). At these densities, the larvae form noticeable mounds, denude soil, and possibly contribute to weediness of fields (Morelli and Alzugaray 1994).
Adult cyclocephaline scarab beetles are relatively less important as pests of agroecosystems. However, some species have been recorded to chew on the foliage, consume pollen, seeds, and fruits. The conditions in which adults of these species become pests in these systems is not clear and well documented examples are rare. Colombian Cyclocephala ruficollis Burmeister were observed to chew on the foliage of sesame (Sesamum indicum L.; Pedaliaceae), cotton (Gossypium spp.; Malvaceae), maize, banana shoots (Musa spp.; Musaceae), and sunflowers (Helianthus annuus L; Asteraceae) (Posada Ochoa 1989). Cyclocephala ovulum Burmeister has also been reported to attack seeds of sunflower in Argentina (Hayward 1946). The foliage of common beans (Phaseolus vulgaris L.; Fabaceae) and African oil palm (Elaeis guineensis Jacq.; Arecaceae) are chewed by C. amazona (reported as C. signata) in Colombia (Posada Ochoa 1989). An unidentified Cyclocephala chews foliage of cassava (Manihot esculenta Crantz; Euphorbiaceae) (Posada Ochoa 1989). In addition to sunflowers, C. ruficollis and C. amazona reportedly feed on the flowers of Citrus (Rutaceae), various ornamental plants, maize, and C. ruficollis will feed on the pollen of sorghum (Sorghum sp.; Poaceae) in Colombia (Posada Ochoa 1989). Similar flower feeding on Citrus has also been reported for C. melanocephala in Brazil (Remillet 1988). At least two Cyclocephala species will eat fruit of cultivated rose apples (Syzygium jambos (L.) Alston; Myrtaceae), custard apples (Annona spp.; Annonaceae), and guava (Psidium guajava L.; Myrtaceae) (Posada Ochoa 1989). A Stenocrates sp. may also attack foliage of maize in Colombia and sugarcane in Brazil (Lima 1953, Posada Ochoa 1989. The role of Dyscinetus species in agroecosystems is not clear. It is possible that some reports of damage to crops by Dyscinetus are complicated by misidentifications of the similar looking genus Euetheola Bates (Scarabaeidae: Dynastinae: Pentodontini) (Phillips and Fox 1924). In some cases, Dyscinetus species have been reported in association with crop systems but are considered non-damaging saprophages. For example, the larvae of Dyscinetus sp. in Puerto Rico can be found in association with rotting stumps of sugarcane but they apparently do not attack the roots of living plants (Smyth 1916). In contrast, D. rugifrons is considered a pest of cultivated sugar cane in Argentina where the larvae burrow into internodes and buds (Costilla 1991). Adult D. rugifrons attack the shoots, but this is rare (Costilla 1991). In another case of conflicting information, Phillips and Fox (1924) reported that D. morator would not attack maize in their experiments. However, adults of this species will attack young maize shoots in North Carolina in fields with wet, high organic matter soil (Anonymous 1980). Dyscinetus gagates Burmeister can be a silvicultural pest in Argentina during years when populations of the beetles are high. Dyscinetus rugifrons adults attack the stems and roots of young cultivated Populus hybrids (Salicaceae) (Moore 1958) and Eucalyptus (Myrtaceae) (Bosq 1945), killing the plants. In Florida, D. morator adults attack carrots (Apiaceae), radishes (Brassicaceae) (Foster et al. 1986), and the bulbs, buds, and petioles of cultivated Caladium (Araceae) (Anonymous 1971, Price andKring 1991). Larvae of this species also damage Pangola-grass pastures in Florida when at high den-sities (Anonymous 1956). In Maryland, D. morator larvae can damage the roots of azaleas (Rhododendron spp.; Ericaceae) (Staines 1990). Dyscinetus morator larvae can damage the fine root tips of cranberry (Vaccinium sp.; Ericaceae) in bog cropping systems, though they are considered minor pests (Scammell 1917).

Invertebrate predators and parasitoids
Cyclocephaline scarab beetle larvae are subject to parasitism by ecto-and endoparasitoid flies and wasps. The fly Mallophora ruficauda Wiedemann (Diptera: Asilidae) is a koinobiont parasitoid of C. signaticollis (Barrantes and Castelo 2014). Mallophora ruficauda can also attack C. putrida and C. modesta, but the fly does not complete its development on these hosts or the adult flies are stunted and deformed (Barrantes and Castelo 2014). Two other asilid flies, M. sylverii Macquart and Diogmites vulgaris Carrera, parasitize Dyscinetus rugifrons in Brazil (Dennis and Knutson 1988). Dyscinetus species are parasitized by Tiphia parallela Smith (Hymenoptera: Tiphiidae) in Guyana (Box 1925). Tiphia pygidialis Allen parasitizes C. borealis, C. lurida lurida, and C. pasadenae (Rogers and Potter 2004). Cyclocephala pasadenae was demonstrated to be toxic to spiders of several families when eaten, though the mechanism of this toxicity remains unexplained (Cokendolpher 1993). Ants can be significant egg and larval predators of C. lurida lurida in turfgrass (Zenger and Gibb 2001). The parasitoid larvae of Plega banski Rehn (Neuroptera: Mantispidae: Symphrasinae) attack Cyclocephala pupae in Arizona (Werner and Butler 1965).
Cyclocephalines, like many relatively large beetles, are hosts of phoretic mites. Acarid and macrochelid mites have been reported from Cyclocephala (Goldwasser 1987, Crocker et al. 1992. Phoretic macrochelid mites on Cyclocephala are common in aroid inflorescences visited by the beetles, and the mites appear to feed on floral exudates (Goldwasser 1987). The mesostigmatid Dyscinetonyssus hystricosus Moss and Funk is hypothesized to be a parasite of D. morator (Moss and Funk 1965). This conclusion was based on morphological features of the mites consistent with parasitic habits and the observation that all life-stages and sexes of the mites are present on D. morator (Moss and Funk 1965).

Entomopathogenic nematodes and worms
Entomopathogenic nematodes are remarkable for their ability to attack and kill numerous insect pests. Their flexibility of use, combinability with other chemical and biological controls, and safety has led to their use in IPM strategies for control of C. borealis, C. pasadenae, C. lurida lurida, and C. hirta grubs (Kaya et al. 1995, Koppenhöffer and Kaya 1997, Converse and Grewal 1998, Koppenhöffer and Fuzy 2003, Koppenhöffer et al. 1999, 2002. Many species and strains of Steinernema Travassos (Nematoda: Steinernematidae) and Heterorhabditis Poinar (Heterorhabditidae) infect these Cyclocephala species, though C. pasadenae appears to have the most natural resistance to nematode infection among examined North American Cyclocephala (Koppenhöffer and Kaya 1996, Koppenhöffer et al. 2004. Nematode infections of South American cyclocephalines have received some attention. The Argentinian pest grub C. signaticollis is naturally infected by two rhabditid and two thelastomatid nematodes (Reboredo and Camino 2000, Camino and Reboredo 2005, Camino and Achinelly 2012. Cyclocephala modesta hosts a thelastomatid parasitic nematode in its alimentary canal (Achinelly and Camino 2008). Ancognatha scarabaeoides, a major grub pest in Colombia, can be readily infected by Steinernema nematodes (Lucero Malfa et al. 2006 (Fincher et al. 1969). Beyond nematodes, information regarding the infection of cyclocephalines by other worms is lacking. The only known example is that of D. gagates adults, which are suitable intermediate hosts of the rat tapeworm (Hymenolepis diminuta [Rudolphi]; Cestoda: Hymenolepididae) under laboratory conditions (Bacigalupo 1939).

Entomopathogenic bacteria and fungi
Bacterial and fungal pathogens have proven useful for IPM of injurious scarab grubs, especially Japanese beetle (Popillia japonica Newman). Several of the most important pathogens for P. japonica control have been explored for use on Cyclocephala species. The fungal parasites Beauveria bassiana (Bals.-Criv.) Vuill and Metarhizium anisopliae (Metchnikoff) Sorokin (both Sordariomycetes: Hypocreales) have been evaluated for pathogenicity and virulence in C. signaticollis, C. borealis, and C. lurida lurida Diaz 2005, Redmond andPotter 2010). Experiments demonstrated that one Brazilian strain of B. bassiana caused significant mortality against C. signaticollis, while native strains of M. anisopliae were not pathogenic in this species (Berón and Diaz 2005). This relatively low mortality caused by B. bassiana and M. anisopliae was also observed in C. lurida lurida, but both fungal pathogens display synergism with entomopathogenic nematodes (Wu et al. 2014). Cyclocephala borealis and C. lurida lurida larvae surveyed from Kentucky golf courses also showed low infection rates by M. anisopliae (Redmond and Potter 2010). Cyclocephala parallela can also be naturally infected by M. anisopliae in sugarcane fields (Boucias et al. 1986). Metarhizium anisopliae -based control measures of A. scarabaeoides may have promise in Colombia, as at least one identified strain causes high mortality in this species (Marino et al. 2004).
Milky disease, caused by the bacterium Paenibacillus popilliae Dutky (Bacillales: Paenibacillaceae), is the only registered biological control specifically for P. japonica (Koppenhöfer et al. 2000). Infections of the disease are chronic in populations, but infection rates grow slowly (Klein 1992). Thus, milky disease is effective for inoculative, long-term treatments rather than as an emergency control measure (Klein 1992). Several Cyclocephala species can be infected by P. popilliae. Cyclocephala parallela larvae infected by P. popilliae show significantly higher mortality than healthy larvae (Boucias et al. 1986, Cherry and Boucias 1989, Cherry and Klein 1997. Bacillus thuringiensis Berliner (Bt) is the most important bacterial biological control agent of insects, but there is a lack of information about infectivity in cyclocephalines. What is known about Bt in Cyclocephala suggests that infections enhance other biological control methods. Like fungal infections, bacterial infections by B. t. subspecies japonensis Buiui and P. popilliae cause additive or synergistic mortality with entomopathogenic nematodes in C. hirta and C. pasadenae (Thurston et al. 1993, Koppenhöfer and Kaya 1997, Koppenhöfer et al. 1999. Bt isolated from C. signaticollis in Argentina caused 100% mortality in inoculated larvae (Consolo et al. 2010).

Human use as food
Beetles are the most commonly consumed insects by humans (van Huis et al. 2013). Many phytophagous scarab larvae reach large sizes by the 3 rd instar and can be found in abundance, making these beetles a valuable food resource. Data about the consumption of cyclocephaline scarab beetles is lacking, but there are a few well documented examples. The Lacandon people of Chiapas eat larval, pupal, and adult C. fasciolata (Ramos-Elorduy and Pino Moreno 2002). Additionally, C. capitata Höhne is eaten in southwestern Mexico and C. guttata Bates larvae and adults are eaten in Veracruz (Ramos-Elorduy and Pino Moreno 2004). Ecuadorians eat the larvae of Ancognatha castanea Erichson, A. jamesoni Murray, andA. vulgaris Arrow (Onore 1997, 2005). Similarly, the larvae of an unidentified Ancognatha species may be regularly fried and eaten in Cauca, Colombia (DeFoliart 2012). Among American Indians in the western US, the Mono Lake and Owens Valley Paiute would roast and eat adult Phyllophaga sp. (Scarabaeidae: Melolonthinae) (Sutton 1988). These groups may have also eaten common Cyclocephala spp., but this is unconfirmed (Sutton 1988). In Thailand, Karen-speaking people from the Tak province fry and eat adult Peltonotus nasutus Arrow that they collect from the inflorescences of Amorphophallus paeoniifolius (Araceae) (Danell 2010).

Scope of the Mutualism
Based on the most specific available data, about 97 cyclocephaline scarab beetle species have been reported from the flowers of at least 58 plant genera representing 17 families and 15 orders (Moore and Jameson 2013), though new data are being published often. The preponderance of data suggests that tropical cyclocephaline species are involved in a pollination mutualism with species in the early-diverging angiosperm families Nymphaeaceae, Annonaceae, Magnoliaceae, Araceae, Cyclanthaceae, and Arecaceae (Moore and Jameson 2013). More sporadic data suggests that cyclocephaline floral visitation of more derived angiosperm groups is opportunistic and not adequately explained. However, based on the observations of Prance (1976), Cyclocephala species may be unrecognized pollinators of some Neotropical genera of the Brazil nut family (Lecythidaceae).
The mutualism between cyclocephaline scarab beetles and these early-diverging angiosperms has resulted in a cantharophilous floral syndrome in these groups. This floral syndrome is the result of the convergent evolution of several floral traits that accommodate "mess-and-spoil" beetle pollination (Faegri and van der Pijl 1979). Among the families Nymphaeaceae, Annonaceae, Magnoliaceae, Araceae, Cyclanthaceae, and Arecaceae these convergent floral traits include: 1) bisexuality of flowers or inflorescences; 2) protogyny; 3) nocturnal flower activity; 4) relatively large flowers or inflorescences that provide a "pollination chamber" and are sturdy enough to withstand beetle damage; 5) thermogenesis during anthesis; 6) production of excess pollen, floral exudates, or sterile floral parts as a food reward; 7) coordination of timing between beetle behavior, thermogenesis, and floral sexual stages; 8) large pollen grains; 9) sticky floral exudates; 10) strong floral scents and; 11) pale colored flowers or inflorescences (Bawa and Beach 1981, Bernhardt 2000, Silberbauer-Gottsberger et al. 2001, Davis et al. 2008, Thien et al. 2009. Excellent observational and experimental evidence indicates that cyclocephaline scarab beetles are primary or secondary pollinators of these plant groups (Cramer et al. 1975, Beach 1982, Young 1986, 1988a, b, Gottsberger 1989, Dieringer et al. 1999, Maia et al. 2012. Cyclocephalines are offered rewards for their pollination of these families. These rewards include access to aggregation and mating sites, food, and metabolic boosts associated with floral thermogenicity. Facultative endothermy (sustained increase in thoracic muscle temperature) during rest, terrestrial activity, and preparation for flight has been documented in Coleoptera and Scarabaeidae more narrowly, including Cyclocephala species (Bartholomew and Casey 1977a, b). Among some examined dung beetles, changes in thermoregulation and behavior are associated with high levels of intra-and interspecific competition for rapidly depleting dung resources Bartholomew 1979, Ybarrondo andHeinrich 1996). Cyclocephala colasi Endrődi experience sporadic bouts of endothermy during the early evening when these beetles fly between inflorescences (Seymour et al. 2009). These bouts of endothermy are more intense at lower ambient temperatures and continue throughout the night, when they may be associated with feeding, mating, or escape behaviors (Seymour et al. 2009). The host plant, Philodendron solimoesense A.C.Sm. (Araceae), continues thermogenesis even after floral scent compounds have been volatilized (Seymour et al. 2003). This suggests that the increased temperature of the inflorescences serves as a thermal reward to the beetles, lowering the amount of energy spent achieving sporadic endothermy (Seymour et al. 2003(Seymour et al. , 2009. Thermal rewards of this nature are predicted to be more important in montane forest habitats with much lower average ambient temperatures than lowland rainforests (Seymour et al. 2009).

Attraction to flowers and inflorescences
Cyclocephaline attraction to their floral hosts is hypothesized to be driven by both longdistance chemical cues and short-distance visual stimuli. In the case of Philodendron bipinnatifidum Schott ex Endl. (Araceae), Erioscelis emarginata (Mannerheim) will not land on inflorescences covered in black cloth (obscuring visual stimuli associated with the scent releasing plant) (Gottsberger and Silberbauer-Gottsberger 1991). Furthermore, experiments demonstrated that these beetles were differentially attracted to P. bipinnatifidum spathes covered in yellow paper, indicating that contrasting colors play a role in close range attraction (Gottsberger and Silberbauer-Gottsberger 1991). Slight differences in spathe color and scent have also been hypothesized to influence the community of Cyclocephala spp. visiting Dieffenbachia spp. inflorescences in Costa Rica and Panama (Beath 1999). The white flowers of Victoria amazonica (Poepp.) J.C. Sowerby (Nymphaeaceae) have been hypothesized to aid in the attraction of cyclocephalines, along with their heavy floral scent (Prance and Arias 1975). Contrasting colors have also been suggested to play a role in the attraction of Cyclocephala species to Cyclanthus (Beach 1982).
The chemical composition of the floral scents attractive to cyclocephalines has received some research attention. These heavy scents are generally only volatile at elevated temperatures during floral thermogenesis. For example, protogynous P. bipinnatifidum inflorescences can reach an astonishing 46˚C during the female phase of anthesis (Gottsberger and Silberbauer-Gottsberger 1991). Research on these floral scents reveals that while they are complex chemical mixtures, a single dominant scent compound is sufficient for cyclocephaline attraction. In Brazil, the nitrogen and sulfur containing compound 4-methyl-5-vinylthiazole is the dominant floral scent constituent in four Annona spp. (Annonaceae) and Caladium bicolor (Aiton) Vent. (Araceae) pollinated by Cyclocephala species (Maia et al. 2012). Scent trap experiments confirmed that this compound alone was sufficient to attract these beetles (Maia et al. 2012). Dötterl et al. (2012) identified three main compounds present in the P. bipinnatifidum floral scent that are attractive to E. emarginata. The dominant compound alone, 4-vinylanisole (also called 4-methoxystyrene), was sufficient to attract E. emarginata and various mixtures of the three scents also served to attract the beetles . A mixture of dihydro-β-ionone and methyl jasmonate was synergistically attractive to E. emarginata, which pollinates Philodendron adamantium Mart. ex Schott (Araceae) (Pereira et al. 2014). Among Nymphaea spp. (Nymphaeaceae) pollinated by Cyclocephala, floral scents are dominated by aromatic ethers and aliphatic esters . 4-vinylanisole is also present in Nymphaea species pollinated by Cyclocephala, suggesting that some Nymphaea spp. and P. bipinnatifidum may have converged on a similar floral scent for attracting these beetles. The ester methyl-2-methylbutanoate is the dominant floral scent compound in Magnolia ovata (A.St.-Hil.) Spreng. (Magnoliaceae) and is sufficient to attract C. literata Burmeister (Gottsberger et al. 2012). (S)-2-hydroxy-5-methyl-3-hexanone is one of the dominant compounds in the floral scent of Taccarum ulei Engl. & K.Krause and is sufficient to attract its Cyclocephala pollinators (Maia et al. 2013).
The mechanisms of attraction of cyclocephalines to other flower groups is poorly understood. The phytelephantoid palms (Arecaceae) Phytelephas aequatorialis Spruce, P. macrocarpa Ruiz & Pav., P. seemannii O.F. Cook, and Aphandra natalia (Balslev & A.J. Hend.) Barfod, all visited by Cyclocephala, have floral scents that are dominated by 4-methylanisole and 2-methoxy-3-sec-butyl pyrazine (Ervik et al. 1999). The presence of anisoles in the floral scents of phytelephantoid palms, Nymphaeaceae, and Araceae suggests that this class of compounds may have convergently evolved in these groups for attraction of cyclocephalines. Cyclanthus bipartitus Poit., visited by several Cyclocephala species, has a floral scent dominated by a unique compound called (E)cyclanthone (Schultz et al. 1999). Heavy floral scents are likely to play a role in cyclocephaline attraction in every case. For example, C. melanocephala has been collected in the flowers of Datura and related genera (Solanaceae) from across its range (Moore and Jameson 2013). The dominant floral scent compounds found in these flowers are very different from those in early diverging angiosperms described above, and are comprised mostly of terpenes, terpenoids, and aromatic alcohols (Raguso et al. 2003).

Redundancy of pollinating cyclocephalines
Some authors have speculated that floral scent compounds are serving as surrogate sex pheromones for cyclocephalines (Schatz 1990, Dieringer et al. 1999. No specific Cyclocephala-derived sex pheromones have been chemically identified (Leal 1996), though some North American Cyclocephala species appear to use volatile pheromones. For example, C. lurida and C. borealis females use pheromones to attract males, and these pheromones are cross-attractive to males of both species (Potter 1980). Further experiments demonstrated that C. lurida larvae produce a similar male-attracting compound that elicits attempted mating (Haynes et al. 1992). These pheromones are present in all three instars and pupae (Haynes and Potter 1995). Cross-attractiveness of C. lurida pheromone extracts are limited to C. borealis, as C. pasadenae and C. longula are not attracted to these scents (Bauernfeind et al. 1999).
In cases of cross-attractive pheromones, it can be predicted that some other mechanism (temporal or behavioral) maintains species boundaries. For sympatric C. lurida and C. borealis in Kentucky, differences in peak flight time and mating periods throughout the night serve to temporally isolate these species (Potter 1980). If attractive floral scents are serving as sex pheromones for tropical cyclocephalines, then the mechanisms isolating species remain unexplained. Only one case of interspecific copulation has been documented for cyclocephalines. The South American species C. putrida was observed mating at light traps, and several male C. putrida copulated with females of a Tomarus sp. (Dynastinae: Pentodontini) (Bosq 1936). Because these tropical cyclocephalines often mate within their host inflorescences, it is unclear how sexual isolation is maintained when congenerics are present. Diagnostic secondary sexual characters of the elytral epipleuron in females and protarsal and paramere morphology in males may be involved in the sexual isolation of cyclocephaline species (Moore 2012).
Many different cyclocephaline species can be found associated with a floral host at a specific time or throughout a season. There is little evidence for monophagy in the group, and available data indicate that tropical cyclocephalines are predominantly oligophagous or polyphagous floral feeders (Moore and Jameson 2013). For example, C. bipartitus inflorescences can contain up to three Cyclocephala species at one time (Beach 1982). Parsing out how redundant cyclocephalines are in their pollinator functions has been assessed in a few cases. Detailed studies on Dieffenbachia Schott (Araceae) indicate that among a group of cyclocephaline floral visitors, some species are relatively more effective pollinators (Young 1988a). Seasonal abundance of cyclocephalines at a specific locality, along with floral phenology, may also determine which species are primary or secondary pollinators (Maia et al. 2010, Costa et al. 2017).

Fossil cyclocephalines
The only known cyclocephaline fossil is from the extant South American species C. signaticollis. A fossilized elytron and pronotum of an unsexed C. signaticollis individual were discovered in Buenos Aires Province, Argentina (Ramírez and Alonso 2016). The fossil is from the Late Pleistocene (Tarantian Stage) and the sediments containing the fossil dated between 12,100 ± 100 BP and 13,400 ± 200 BP (Ramírez and Alonso 2016). Neoichnological experiments demonstrated that C. borealis and C. lurida lurida larvae create diagnostic backfilled meniscate burrows and ellipsoidal chambers as they burrow through soil, while adults create poorly organized backfilled burrows (Counts and Hasiotis 2009). The diagnostic features of these burrows may allow for the future detection of cyclocephaline scarab beetle ichnofossils.

Cyclocephaline Phylogeny
Very little is known about the phylogeny of Dynastinae, and the monophyly of its tribes is in doubt. The lack of phylogenetic framework for the subfamily has limited the ability to hypothesize sister relationships among tribes and reconstruct the evolution of ecological (e.g., the floral feeding syndromes in Cyclocephalini) and morphological (e.g., such as thoracic and cephalic armature in Oryctini and Dynastini) traits. Indeed, the most meaningful comparison of characters for Dynastinae in the literature has centered around the subfamily's relationship to Rutelinae, especially among cyclocephalines (Jameson 1998, Jameson et al. 2002, Jameson and Wada 2004. Several studies have begun to address this gap in knowledge. The morphological phylogenetic analysis (128 characters) of Rutelina (Rutelinae: Rutelini) (Jameson 1998) was the first empirical study to suggest that the monobasic ruteline tribal-and subtribal-groups Peltonotini and Acrobolbiina were more closely related to Cyclocephalini than Rutelini. This analysis, however, did not include enough exemplar taxa from Dynastinae to conclude anything about tribal relationships in the subfamily. Schiestl and Dötterl (2012) used an analysis of 18S sequence data to examine the evolution of olfactory preferences in scarabaeoids. This analysis suggested a sister relationship between Dynastinae and Rutelinae, but it did not resolve intrasubfamilial relationships of the included genera nor did it report statistical support for recovered nodes (Schiestl and Dötterl 2012). A Cyclocephala exemplar species was included in this analysis, and this species fell within the dynastine clade (Schiestl and Dötterl 2012). Rowland and Miller (2012) performed a four-gene phylogenetic analysis of Dynastini (Dynastinae) that included one Cyclocephala exemplar. This analysis was useful for recovering subtribal relationships within Dynastini, but the relationship of Dynastini to Cyclocephalini (Cyclocephala) and Pentodontini (Orizabus) was unresolved (Rowland and Miller 2012, see also Jin et al. 2016).
The most informative molecular phylogenetic analyses of phytophagous scarabs to date were conducted by McKenna et al. (2014) and Gunter et al. (2016). Both studies represent huge leaps forward in our understanding of subfamilial relationships in Scarabaeidae due to their resolution, statistical support, and taxa sampling. Despite their strengths, these studies are difficult to compare because of differences in gene selection and small (but significant for interpretation) differences in taxa sampling. McKenna et al. (2014) utilized 28S and CAD to phylogenetically analyze staphyliniform beetle (Histeroidea, Hydrophiloidea, and Staphylinoidea) relationships while using Scarabaeiformia as an outgroup. The most derived group of Scarabaeidae recovered from this analysis was a clade that included Cetoniinae + (Dynastinae and Rutelinae) (McKenna et al. 2014) (Fig. 2). Rutelinae was recovered as polyphyletic (McKenna et al. 2014) (Fig. 2). Three orthochilous (labrum vertically produced from clypeus and fused to clypeus) and three homalochilous (labrum horizontally produced relative to the clypeus and separated from the clypeus by a suture) rutelines from four total tribes were included in the analysis (McKenna et al. 2014). The included orthochilous rutelines (Anoplognathini and Anatistini) were recovered in the same clade, but the group was not monophyletic (McKenna et al. 2014) (Fig. 2).
The homalochilous Rutelinae (Anomalini and Rutelini) were polyphyletic, with Oryctomorphus (Rutelini) falling into a clade including Anatistini and Anoplognathini  (Fig. 2). These results suggest that Cyclocephalini is correctly classified in Dynastinae, but that the tribe is polyphyletic if it includes Peltonotus. This phylogenetic analysis is more in line with the placement of Peltonotus near the Asian parastasiine rutelines by Arrow (1908Arrow ( , 1910 than the hypotheses of Jameson (1998). Gunter et al. (2016), building on the datasets of Ahrens et al. (2011Ahrens et al. ( , 2014, utilized 16S, 12S, CO1, and 28S to conduct a phylogenetic analysis of Scarabaeoidea that included over 400 taxa. A clade including Cetoniinae + (Dynastinae and Rutelinae) was recovered, but the node uniting these subfamilies was only weakly supported (0.89 posterior probability) (Gunter et al. 2016) (Fig. 3). These three analyses, built from similar datasets, together suggest that Rutelinae is a paraphyletic grade of tribes (Ahrens et al. 2011, Gunter et al. 2016). The subfamily Dynastinae in these analyses was consistently recovered as the most derived of all scarabaeoids (Ahrens et al. 2011, Gunter et al. 2016. Gunter et al. (2016) recovered a strongly supported node that suggests that the Asian orthochilous ruteline tribe Adoretini is sister to a monophyletic Dynastinae. This node had been similarly recovered by Ahrens et al. (2011). However, this relationship between Adoretini and Dynastini was weakly supported and interrupted by Pachydemini (Melolonthinae)  The analysis by Gunter et al. (2016) included 22 dynastine species from 18 genera in 5 tribes. Nodes were generally poorly supported within Dynastinae, making it difficult to assess relationships among tribes (Gunter et al. 2016) (Fig. 4). The study included one Cyclocephala species, which was recovered as sister to Onychionyx (Oryctoderini), but this relationship was weakly supported (0.83 posterior probability). These results suggest future analyses of Cyclocephalini should include oryctoderine genera (nearly all of which were at some point previously included in Cyclocephalini) to assess the boundaries of the two tribes. Additionally, these analyses do not support the monophyly of the tribes Oryctoderini, Pentodontini, and Phileurini (Gunter et al. 2016).
Taken together, these studies demonstrate that the position of Cyclocephalini in the broader phylogeny of Dynastinae and Rutelinae is not resolved. In addition, very little is known about the relationships among cyclocephaline genera and species. Breeschoten et al. (2013) presented a morphological phylogeny of cyclocephaline genera, but few details of the analysis were provided and the support for recovered relationships were not reported. Moore et al. (2015) suggested that Mimeoma species were nested among a clade of Cyclocephala that included the type species of the genus, C. amazona. These data also provided evidence of two major clades of Cyclocephala based on morphological and molecular evidence (Moore et al. 2015). However, the relationship of Cyclocephala to the other cyclocephaline genera is completely unevaluated.

Generic overviews
The section below summarizes information on the distribution, recognition, and hypothesized relationships of cyclocephaline scarab beetle genera. The provided diagnoses are roughly parallel to each other and, in many cases, discuss morphological characters that have not been adequately described for the group. Diagnoses also rely on the dissection of the mandibles, maxillae, and hindwings. These diagnoses should allow for enhanced identification when in doubt of generic-level affinities. The last identification key to genera for the tribe did not include Peltonotus (Jameson et al. 2002). The key to genera below builds on the work of Jameson et al. (2002) and is supplemental to that identification tool. This key requires dissection of the hindwings and mouthparts and will aid in precise identification of these groups, along with provided diagnoses.
Acrobolbia species can be recognized by the following combination of characters: 1) dorsal coloration varying from all black with variable reddish brown margins of the elytra and elytral suture, or with the elytra partially testaceous; 2) body not anteroposteriorly compressed or dorsoventrally flattened; 3) clypeal apex acuminate in dorsal view; 4) frontoclypeal suture distinct, but incomplete medially; 5) mandibles long, sickle-shaped, with pointed apex; 6) mandibular molar area with rows of circular micropunctures; 7) apical margin of mentum weakly emarginate to nearly straight; 8) galea of maxilla reduced to small, rectangular mound in dorsal view; 9) galea on inner surface with teeth greatly reduced to peg-like projections at the middle and apex; 10) galea on inner surface lacking teeth at base; 11) males with antennal club (segments 8-10) elongated, nearly twice as long as antennomeres 1-7; 12) pronotum with broadly incomplete beaded basal margin; 13) males and females with 3 protibial teeth, basal tooth reduced, removed from the apical 2 teeth, and oriented laterally; 14) protibial spur straight to weakly deflexed; 15) males with inner protarsal claw enlarged and narrowly cleft at apex; 16) mesocoxae touching, nearly contiguous; 17) meso-and metatibiae with distal, divided carinae; 18) metacoxae with lateral edge perpendicular to ventral surface; 19) anterior edge of hindwing distal to apical hinge lacking setae and with produced, membranous border; 20) vein RA with 2 rows of pegs extending distally nearly to margin of apical hinge.
The 22 species of Ancognatha are distributed from the southwestern United States south to Argentina (Fig. 52). The species diversity in the genus is concentrated in north and western South America and in Mexico, west of the Isthmus of Tehuantepec. Biological information on Ancognatha species is lacking, and almost nothing is known about the natural history of adults. In Meso-and Central America, Ancognatha species are associated with premontane, lower montane, and montane tropical forests with some species being collected at elevations from 2,000 to 3,500 m above sea level (Ratcliffe 2003, Ratcliffe and Cave 2006, Ratcliffe et al. 2013. This pattern also holds in South America. Several Ancognatha species have been recorded from elevations over 4,000 m in Peru and northern Chile (Mondaca 2016, Figueroa and Ratcliffe 2016). Some South American Ancognatha species can be very large for the tribe. For example, A. matilei Dechambre from Colombia is up to the 36 mm long (Dechambre 2000). Adults are attracted to lights at night.
Larvae are described for four Ancognatha species (Ritcher 1966, Ramírez-Salinas et al. 2004, Vallejo and Morón 2008, Neita-Moreno and Morón 2008. South American larval descriptions are largely based on material collected in agroecosystems, and thus the natural ecology of Ancognatha immatures is poorly known. Mondaca (2016) reported the larvae of A. aymara Mondaca feeding on grass roots high in the altiplano steppe of northern Chile.
Ancognatha species can be recognized by the following combination of characters: 1) dorsal coloration variable, from all or partially black or testaceous, to light brown with variable dark maculae; 2) body convex and not strongly anteroposteriorly or dorsoventrally compressed; 3) clypeal apex rounded to parabolic, never truncate or emarginate; 4) frontoclypeal suture incomplete medially; 5) males with anterolateral margin of the mandibles without teeth; 6) mandibular apices narrow and elongated, recurved dorsally; 7) mandibular molar area with rows of circular micropunctures; 8) apical margin of mentum narrowly and deeply emarginated; 9) galea of maxilla reduced to a roughly quadrate process; 10) galea of the maxilla on inner surface lacking welldeveloped teeth, teeth when present and visible greatly reduced into spine-like projec- tions; 11) males and females with 3 protibial teeth, basal tooth slightly removed from the more apical 2 teeth, and oriented laterally; 12) protibial spur straight to weakly deflexed; 13) males with inner protarsal claw enlarged and narrowly cleft at apex; 14) mesocoxae narrowly separated and touching; 15) meso-and metatibiae with distal, transverse carinae; 16) metacoxae with lateral edge perpendicular to ventral surface; 17) anterior edge of hindwing distal to apical hinge lacking setae and with produced, membranous border; 18) vein RA with single row of pegs extending distally nearly to margin of apical hinge; 19) elytral margin membranous.
The relationship of Ancognatha species to other cyclocephaline genera has not been evaluated. Acrobolbia may be related to Ancognatha based on characters of the clypeus, mentum, pronotum, prosternal process, protarsus, and mandibles (Jameson 1998, Jameson et al. 2002. Surutu also shares some intriguing characters with Ancognatha, which may be indicative of a close relationship between these two genera. For example, Ancognatha and Surutu species all have a rounded to parabolic clypeal apex and a narrowly, but deeply, emarginated apex of the mentum. Surutu species have a anteriorly projecting tooth at the apex of the labrum, and this is also shared in some Ancognatha species.
Arriguttia was compared to Surutu in the original description of the genus (Martínez 1960a). This is possibly confusing for identification purposes. Arriguttia shares many more characters with Cyclocephala and Augoderia than with Surutu. Arriguttia species can be recognized by the following combination of characters: 1) dorsal coloration varying from all black or with variable dark, reddish coloration on the elytra; 2) body convex and anteroposteriorly compressed, creating a relatively round gestalt; 3) clypeus quadrate in dorsal view, with sides nearly parallel, and the apex distinctly reflexed upwards (most obvious in lateral view); 4) frontoclypeal suture complete medially; 5) males with anterolateral margin of the mandibles weakly toothed; 6) mandibular molar area with rows of circular micropunctures; 7) apical margin of mentum weakly emarginated; 8) galea of the maxilla on inner surface with 3 fused basal teeth, a free median tooth, and 2 fused apical teeth (3-1-2 arrangement); 9) pronotum with broadly incomplete beaded basal margin; 10) males and females with 3 protibial teeth, basal tooth reduced, removed from the more apical 2 teeth, and oriented anteriorly; 11) protibial spur straight to weakly deflexed; 12) males with inner protarsal claw enlarged and narrowly cleft at apex; 13) mesocoxae widely separated; 14) meso-and metatibiae with distal, transverse carinae; 15) metacoxae with lateral edge perpendicular to ventral surface; 16) anterior edge of hindwing distal to apical hinge lacking setae and with produced, membranous border; 17) vein RA with 2 rows of pegs extending distally nearly to margin of apical hinge. The relationships of Arriguttia to other cyclocephaline genera have not been clearly discussed in the literature. Martínez (1968a) stated that Arriguttia should be "placed next to" Surutu, but he did not offer any character justifications for this hypothesis. Endrődi (1966) considered Arriguttia to be a "primitive" cyclocephaline based on his poorly justified character analysis. Arriguttia shares hindwing characters (two rows of pegs on vein RA and a membrane on the leading edge of the hindwing distal to the apical hinge) with Augoderia, Aspidolea, and Cyclocephala. The form of the maxilla (3-1-2 teeth arrangement), the mandibular form (males with anterolateral margin weakly toothed and the molar area with rows of circular micropunctures), the incomplete bead on the basal margin of the pronotum, and the shape and arrangement of the protibial teeth are shared among Arriguttia, Augoderia, and some Cyclocephala (especially species like C. sexpunctata Laporte and species formerly placed in Stigmalia Casey). Future analyses should focus on comparing characters in these Cyclocephala species-groups and genera to Arriguttia, rather than Surutu.
Aspidolea contains 26 species ranging from northern Mexico south through South America (Fig. 54) (Endrődi 1966, 1985a, Ratcliffe 2003, Ratcliffe and Cave 2006, Ratcliffe et al. 2013. The genus includes both widespread and narrowly distributed species. Most Aspidolea (22 of 26 species) are known only from a few South American localities. In contrast, A. fuliginea and A. singularis occur from Mexico south to Argentina and Ecuador, respectively. Bates (1888) described Aspidolea based upon the "elongate and robust" yet toothless maxillary galea found in the type species A. singularis. Bates (1888) noted a similar reduction in maxillary teeth in "Cyclocephala fuliginea Burmeister" and Ancognatha species. Aspidolea contained only A. singularis for over 30 years until Höhne (1922a, b, c) recircumscribed the genus and placed many new species into the group. Höhne (1922a) offered an expanded diagnosis of Aspidolea using characters of the clypeus (sides parallel at base with apical margin perpendicular to the sides), maxilla (toothless and with penicillate setae at the apex), and dorsum (yellow to brownish coloration and generally lacking maculae) to distinguish the genus. Cyclocephala clypeata Burmeister and C. laticeps Harold were transferred into Aspidolea along with ten new species described by Höhne (1922a). The new genus Paraspidolea was erected to contain species similar to Aspidolea, but with at least two small teeth present at the apex of the galea (Höhne 1922a). Six new species were included in Paraspidolea along with the Burmeister species C. fuliginea (Höhne 1922a, b). The subgenus Aspidolea (Aspidolites) was erected to contain the species A. atricollis Höhne (Höhne 1923c). The homonym Aspidolites Höhne was replaced with Aspidolella (Prell 1936). Aspidolea atricollis is conspecific with C. histrionica Burmeister (Endrődi 1966), and the subgenus Aspidolella is considered a synonym of Cyclocephala. Paraspidolea was also synonymized within Aspidolea (Endrődi 1966). The last major contribution to the knowledge of Aspidolea was provided by Dechambre (1992). Dechambre (1992) described three new Aspidolea species, which he included in the "Aspidolea helleri species-group" along with A. helleri (Höhne) and A. chalumeaui Endrődi. These species were placed into the "helleri species-group" based on the bidentate form of the protibial margin in males. This male protibial character is shared with species formerly included in Mimeoma and some Cyclocephala species (like C. amazona) (see Moore et al. 2015). The dorsal coloration of the "helleri species-group", especially the elongated, triangular maculae found along the elytral suture, is like that found in some former Mimeoma species (especially Cyclocephala acuta Arrow and C. englemani (Ratcliffe)). These characters suggest that Aspidolea may not be monophyletic as presently defined.
There is little available biological data for Aspidolea species. Aspidolea adults seem to be readily attracted to lights at night and can occasionally be collected in large numbers (Ratcliffe and Cave 2006, Touroult et al. 2010, Grossi et al. 2011. Floral association data for Aspidolea are mostly lacking. Aspidolea fuliginea were collected in male-and female-phase inflorescences of Oenocarpus bataua Mart. (Arecaceae) in Colombia, though they were only sporadically encountered (Núñez-Avellaneda and Rojas-Robles 2008). In French Guiana, A. quadrata Endrődi was collected from the inflorescence of Montrichardia arborescens (L.) Schott (Araceae) , Ponchel 2006. Neita-Moreno et al. (2007) described the larva and pupa of A. singularis. Larvae of A. singularis were collected from soil beneath cultivated cassava (Manihot esculenta Crantz; Euphorbiaceae) in Colombia (Neita-Moreno et al. 2007).
Aspidolea species can be recognized by the following combination of characters: 1) dorsal coloration highly variable, with or without black or brown maculae on the pronotum and elytra; 2) body not anteroposteriorly compressed or dorsoventrally flattened; 3) clypeus robust and broad, with sides more or less parallel at base, appearing quadrate in dorsal view; 4) frontoclypeal suture complete medially; 5) males with anterolateral margin of the mandibles weakly toothed (in A. fuliginea) or not; 6) mandibular molar area with rows of circular micropunctures; 7) apical margin of mentum broadly and deeply (nearly to level of labial palp insertion) emarginated; 8) galea of maxilla dorsoventrally flattened; 9) dentition of galea of maxilla variable, inner surface of galea lacking teeth or with reduced teeth (2 small, yet obvious teeth at the apex with 1 greatly reduced tooth at the base, presence or absence of medial teeth varies among species, teeth often obscured by dense setae); 10) apex of galea with dense brush of penicillate setae; 11) pronotum with broadly incomplete or complete beaded basal margin; 12) males with 2 or 3 protibial teeth, females with 3 protibial teeth, when 3 teeth are present, basal tooth reduced, removed from the more apical 2 teeth, and oriented laterally; 13) protibial spur straight to weakly deflexed or strongly deflexed; 14) males with inner protarsal claw enlarged and entire (not cleft with a small ramus) or narrowly cleft at apex; 15) mesocoxae widely separated; 16) meso-and metatibiae with distal, transverse carinae; 17) metacoxae with lateral edge acutely angled with respect to ventral surface; 18) anterior edge of hindwing distal to apical hinge lacking setae and with produced, membranous border; 19) vein RA with 2 rows of pegs extending distally nearly to margin of apical hinge.
The five species and subspecies of Augoderia are distributed in Argentina, Bolivia, Brazil, French Guiana, Peru, and Venezuela (Burmeister 1847, Harold 1869b, Arrow 1937b, Blackwelder 1944, Guimarães 1944, Martínez 1966, Gibbs et al. 1977, Endrődi 1966, 1967a, 1985a, Riehs 2005, Ronqui and Lopes 2006, Ponchel 2009, Grossi et al. 2011 (Fig. 55). Augoderia species are similar to some Cyclocephala in overall appearance, although three taxa (A. giuglarisi Ponchel, A. nitidula nitidula, and A. nitidula yungana Martínez) are notable for their metallic, mother-of-pearl luster that reflects circularly polarized light, a cuticular trait that is rare in Dynastinae (Endrődi 1967a, Ponchel 2009, Pye 2010. The biology of Augoderia species is completely unknown. Gibbs et al. (1977) reported A. nitidula as a floral visitor of Magnolia ovata, but this beetle was likely a misidentified Cyclocephala species (see Gottsberger et al. 2012, Moore and. The immature stages are undescribed. Adults are attracted to lights at night (Riehs 2005, Ronqui and Lopes 2006, Grossi et al. 2011. Augoderia, though maintained as a valid genus since Burmeister (1847), is poorly defined and diagnosed in the literature. The irregularly spaced punctures of the elytra and the mother-of-pearl sheen of some taxa are the only characters historically used to separate Augoderia from Cyclocephala. Thus, the genus has no clearly hypothesized synapomorphic characters. For example, many characters used to diagnose Augoderia in Endrődi's (1985a) Dynastinae of the World are all variably present in Cyclocephala, Arriguttia, and Aspidolea species: 1) body short, convex; 2) dorsal coloration yellow, with dark maculae, and with or without metallic reflections; 3) mandibles of males with small anterolateral tooth, lacking in females; 4) frontoclypeal suture complete; 5) 10-segmented antennae with a short club in both sexes; 6) large eyes; 7) males with thickened protarsi; and 8) protibia tridentate in both sexes.
The following combination of characters can be used to recognize Augoderia species: 1) dorsal coloration yellowish or light brown, with or without elytral maculae, with or without metallic, mother-of-pearl sheen; 2) body not anteroposteriorly compressed or dorsoventrally flattened; 3) clypeal apex evenly rounded in dorsal view; 4) frons mesad of eyes with long, erect setae; 5) frontoclypeal suture complete; 6) males with anterolateral margin of mandibles weakly toothed; 7) mandibular molar area with rows of circular micropunctures; 8) apical margin of mentum weakly emarginated; 9) galea of the maxilla on inner surface with 3 fused basal teeth, a free median tooth, and 2 fused apical teeth (3-1-2 arrangement); 10) pronotum at base with incomplete or complete marginal bead; 11) pronotum on anterolateral portions with long, erect setae; 12) males and females with 3 protibial teeth, basal tooth reduced, removed from the apical 2 teeth, and oriented anteriorly; 13) protibial spur straight to weakly deflexed; 14) males with inner protarsal claw enlarged and narrowly cleft at apex; 15) mesocoxae widely separated; 16) metatibiae without distal, transverse carinae; 17) metacoxae with lateral edge perpendicular to ventral surface; 18) anterior edge of hindwing distal to apical hinge lacking setae and with produced, membranous border; 19) vein RA with 2 rows of pegs extending distally nearly to margin of apical hinge.
The nomenclatural history of Chalepides was complicated by a case of homonymy. Chalepides was originally proposed as a subgenus of Parachalepus (Casey 1915). Parachalepus Casey, 1915 is a homonym of Parachalepus Baly, 1885 (Coleoptera: Chrysomelidae) (Prell 1936, Arrow 1937a. To rectify this problem, Chalepides was elevated to the status of genus and comprised the seven species originally included in Parachalepus (Casey 1915, Prell 1936, Arrow 1937a. Parachalepus was proposed based on abdominal characters. Parachalepus included Dyscinetus-like species with a rigid fusion of the propygidium and the pygidium (Casey 1915). The subgenus Parachalepus (Chalepides) was proposed for species with a dramatic reduction of the pygidium in addition to propygidial/pygidial fusion (Casey 1915). Chalepides has been recognized as a valid genus by subsequent authors and was recently revised (Arrow 1937a, b, Endrődi 1966, 1985a, Joly and Escalona 2002.
The 15 species of Chalepides are distributed across South America and the West Indies (Martínez 1978b, Endrődi 1966, 1973a, 1985a, Joly and Escalona 2002, Riehs 2005 (Fig. 56). Species of Chalepides described by Prokofiev (2012) require a special discussion. Chalepides euhirtus Prokofiev and C. unduavicus Prokofiev were described based on specimens from Peru and Bolivia (Prokofiev 2012), and the Peruvian data would represent a new country record for Chalepides. However, both species were placed into the wrong genus, based on the original descriptions and images of the holotypes. The holotype of C. euhirtus appears to be a female specimen of A. fuliginea (Prokofiev 2012). Chalepides unduavicus was later synonymized under A. scarabaeoides and was also considered an infrasubspecific ("ab.") entity (Prokofiev 2013(Prokofiev , 2014. The discussion below covering the biology and genus-level recognition of Chalepides will exclude information on the misclassified species C. euhirtus and C. unduavicus.
Relatively little is known about the biology and natural history of Chalepides species. It is unclear, based on available data, if Chalepides species are floral visitors. Mannerheim (1829) reported that C. dilatatus (Mannerheim) was collected in flowers without further detail. Valla and Cirino (1972) reported a single specimen of an unidentified Chalepides species from the inflorescence of a Victoria cruziana A.D. Orb. Chalepides barbatus adults and larvae are associated with sugar cane fields in Puerto Rico (Wolcott 1923(Wolcott , 1948. In Puerto Rico, adult C. barbatus are prey for the invasive cane toad R. marina (Wolcott 1937(Wolcott , 1948. Like Dyscinetus, Chalepides species may have some semi-aquatic habits. Chalepides luridus (Burmeister) and C. alliaceus (Burmeister) have been collected along the edges of river banks (Endrődi 1973a). Chalepides barbatus reportedly attacks the invasive, aquatic weed water hyacinth (Eichhornia crassipes [Mart.] Solms [Pontederiaceae]) in Uruguay (Silveira Guido 1965, Perkins 1974, Buckingham and Bennett 1989. Chalepides species are attracted to lights at night (Kusui 1992, Riehs 2005, Albuquerque et al. 2016. Chalepides species can be recognized by the following combination of characters: 1) dorsal coloration yellowish brown, dark brown, or almost black with greenish reflections in some species; 2) body convex, not strongly anteroposteriorly compressed or dorsoventrally flattened; 3) clypeus trapezoidal with apex truncate in dorsal view; 4) frontoclypeal suture complete or narrowly incomplete medially; 5) males with anterolateral margin of the mandibles lacking weak tooth; 6) mandibular molar area with rows of circular micropunctures; 7) mandibular molar area on proximal margin with 2 semicircular depressed pits; 8) galea of maxilla on inner surface with 2 fused basal teeth, 2 free medial teeth, and 2 fused apical teeth (2-2-2 arrangement); 9) pronotum with broadly incomplete beaded basal margin; 10) males and females with 3 protibial teeth on lateral margin, basal tooth not greatly reduced, only slightly removed from apical 2 teeth, and oriented laterally; 11) protibial spur straight to weakly deflexed; 12) males with inner protarsal claw enlarged and entire at apex, not cleft; 13) mesocoxae not widely separated, nearly touching; 14) metacoxae on lateral edge with transverse, depressed sulcus; 15) metacoxae with lateral edge perpendicular to ventral surface; 16) meso-and metatibiae with distal, transverse carinae; 17) anterior edge of hindwing distal to apical hinge with erect setae and lacking produced, membranous border; 18) vein RA with single row of pegs proximal to the apical hinge; 19) propygidium expanded, propygidium and pygidium fused, pygidium with long, dense setae.
The speciose genus Cyclocephala contains over 350 taxa distributed throughout the Nearctic and Neotropical realms (Fig. 57). Cyclocephala contains the only adventive species in Cyclocephalini, with C. pasadenae and C. signaticollis established in Hawaii and Australia, respectively (Carne 1956). The greatest number of Cyclocephala species is found in northern South America, but many endemic species occur in Meso-and Central America. Some Cyclocephala species are extremely geographically widespread. For example, C. lunulata occurs from the southwestern United States south to Argentina. In contrast, there are also cases of endemism in mainland species of the genus. The pollination mutualist C. jalapensis occurs only in a narrow band of habitat in eastern Mexico (Veracruz, Puebla, Oaxaca, Querétaro, and Hidalgo states) where its host plant Magnolia schiedeana Schltl. is found Delgado 1994, Dieringer andEspinosa 1994).
Cyclocephala is a difficult genus to diagnose due to its species richness, diversity of forms, and probable non-monophyly. Many of the character descriptions below are complicated by these factors. Cyclocephala species can be recognized by the following combination of characters: 1) dorsal coloration highly variable; unicolored black, green, or light brown, pronotum in some species cherry-red, light brown species often have complex maculae patterns of the pronotum and elytra; 2) body not anteroposteriorly compressed or dorsoventrally flattened; 3) clypeal apex variable; evenly rounded, parabolic, acute, emarginate, triemarginate, or nearly straight; 4) frons mesad of eyes with or without long, erect setae; 5) frontoclypeal suture complete or incomplete medially; 6) males with anterolateral margin of mandibles weakly toothed or not; 7) mandibular molar area with rows of circular micropunctures either present or absent; 8) apical margin of mentum weakly emarginated or broadly and deeply emarginated; 9) galea of the maxilla well-developed [with or without teeth] or reduced into a rounded process; 10) galea of the maxilla dorsoventrally flattened or not; 10) galea of maxilla on inner surface variable (not all character states are given here); with 3 fused basal teeth, a free median tooth, and 2 fused apical teeth (3-1-2 arrangement) (in C. amazona-like species and former Mimeoma, the galea are flattened and the basal tooth is compressed and rotated, giving the appearance of being bidentate with the third tooth shifted dorsally); with 2 fused basal tooth and 2 fused apical teeth (2-0-2 arrangement); with 2 fused basal teeth, 1 middle tooth, and 2 fused apical teeth (2-1-2 arrangement); 11) pronotum at base with incomplete or complete marginal bead; 12) pronotum on anterolateral portions with or without long, erect setae; 13) males with 2 or 3 protibial teeth, females always with 3; 14) protibial spur straight to weakly deflexed or strongly decurved; 15) males with inner protarsal claw enlarged and narrowly cleft at apex or entire at apex; 16) mesocoxae widely separated or nearly touching, contiguous; 17) metatibiae with or without distal, transverse carinae; 18) metacoxae with lateral edge perpendicular to ventral surface or with lateral edge angled underneath the ventral surface; 19) anterior edge of hindwing distal to apical hinge lacking setae and with produced, membranous border or lacking membranous border and with decumbent setae (C. cribrata species-group); 20) vein RA with 2 rows of pegs extending distally nearly to margin of apical hinge.
The genus Dyscinetus comprises 21 species distributed from North America south to Argentina and the West Indies (Fig. 58). Smooth, large, and mostly black Dyscinetus species superficially resemble hydrophilid beetles. Dyscinetus is generally not considered an aquatic or semiaquatic genus. However, some species in the genus have an intriguing association with moist, mucky soils and aquatic plants. Dyscinetus rugifrons and another Dyscinetus sp. attack water hyacinth in Uruguay (Silveira Guido 1965, Bennett and Zwolfer 1968, Perkins 1974. Dyscinetus morator also attacks water hyacinth in Florida (Perkins 1974, Buckingham andBennett 1989). These species are considered scavengers and enhancers of damage started by other arthropods on water hyacinth, though they are known to attack healthy tissues (Perkins 1974, Buckingham andBennett 1989). Feeding damage on water hyacinth occurs inside petioles, crowns, petiole bases, and submerged roots (Perkins 1974, Buckingham andBennett 1989).
Experiments indicated that D. morator can survive submerged in water for up to 36 hours (Buckingham and Bennett 1989). The mechanism allowing for this prolonged submersion is unclear. Air bubbles are visible along the elytral margin and on the metathorax in submerged individuals (Buckingham and Bennett 1989). Dyscinetus laevipunctatus Bates was also observed submerged in association with water hyacinth in Mexico (García-Rivera and Contreras-Ramos 2015). Unlike many other genera in the group, Dyscinetus adults are not known to visit flowers. A Brazilian Dyscinetus species was reportedly attracted to the floral odors of Annona sp., although these beetles were not encountered in any inflorescences (Gottsberger 1989). This is the only mention of Dyscinetus floral attraction in the literature.
Dyscinetus species can be recognized by the following combination of characters: 1) dorsal coloration dark piceous to black; 2) body convex, not strongly anteroposteriorly compressed or dorsoventrally flattened; 3) clypeus trapezoidal with apex truncate in dorsal view; 4) frontoclypeal suture complete medially; 5) males with anterolateral margin of the mandibles lacking weak tooth; 6) mandibular molar area with rows of circular micropunctures; 7) mandibular molar area on proximal margin with 2 semicircular depressed pits; 8) galea of maxilla on inner surface with 2 fused basal teeth, 2 free medial teeth, and 2 fused apical teeth (2-2-2 arrangement); 9) pronotum with broadly incomplete beaded basal margin; 10) males and females with 3 protibial teeth on lateral margin, basal tooth not greatly reduced, only slightly removed from the more apical 2 teeth, and oriented laterally; 11) protibial spur straight to weakly deflexed; 12) males with inner protarsal claw enlarged and narrowly cleft at apex; 13) mesocoxae not widely separated, nearly touching; 14) metacoxae on lateral edge with transverse, depressed sulcus; 15) metacoxae with lateral edge perpendicular to ventral surface; 16) meso-and metatibiae with distal, transverse carinae; 17) anterior edge of hindwing distal to apical hinge with erect setae and lacking produced, membranous border; 18) vein RA with single row of pegs proximal to apical hinge; 19) propygidium not expanded, with propygidium and pygidium not fused.
The five species of Erioscelis are distributed in South America north to Nicaragua (Fig. 59). Erioscelis species are remarkable among cyclocephalines for their well-characterized floral visitation syndromes. Erioscelis species are associated with nocturnally blooming genera in the family Araceae. Three Erioscelis species have been reported from the spathes of Dieffenbachia, Philodendron Schott, Syngonium Schott, Montrichardia Crueg., and possibly Xanthosoma Schott (Schrottky 1910, Gottsberger and Amaral 1984, Young 1986, Grayum 1996, Croat 1997, Morón 1997, Beath 1998. While the association between Erioscelis species and aroid flowers is firmly established, there is little evidence of species-or genus-level specificity in this pollination mutualism. For example, Erioscelis columbica Endrődi has been collected from the spathes of nine different Philodendron species in Heredia, Costa Rica (Grayum 1996, Croat 1997, Morón 1997, Moore and Jameson 2013. Based on feeding damage to Philodendron inflorescences by Erioscelis, it was hypothesized that this genus may be an interloper on the cyclocephaline/aroid mutualism (Goldwasser 1987). Other observations seem to indicate that Erioscelis species are part of this mutualism.
The descriptions of Erioscelis spp. visitation of Dieffenbachia and Philodendron inflorescences are some the most detailed available for Cyclocephalini. In Costa Rica, E. columbica is a pollinator of Dieffenbachia nitidipetiolata Croat & Grayum (Young 1986, 1988a, 1988b. Erioscelis columbica arrive at receptive female-phase inflorescences during nightfall, where they feed on staminodia and mate (Young 1986). The beetles exit the spathe after 24 hours when the spadix is in the male-phase and shedding pollen (Young 1986). Erioscelis columbica are covered in sticky pollen grains while exiting the spathe, and they may also feed on some of the pollen (Young 1986). Erioscelis proba (Sharp) displays similar behavior in the inflorescences of two other Dieffenbachia species in French Guiana (Gibernau 2015a).
Observational and experimental evidence suggests that Erioscelis emarginata (Mannerheim) prefers to feed upon sterile staminate flowers on the spadix in two Philodendron species (Maldonado et al. 2015). Furthermore, analyses of nutritional and defensive compound (calcium oxalate) content of sterile and fertile flowers in these Philodendron species suggested that sterile staminate flowers have lower amounts of defensive compounds (Maldonado et al. 2015). Erioscelis species are seemingly attracted to the strong floral scents that are volatilized during thermogenesis and receptivity of the staminate flowers in these aroids. The dynamics of floral scent attraction are mostly unexplored for Erioscelis. In the case of Philodendron adamantium Mart. ex Schott, a single dominant flower scent compound (Dihydroβ-ionone) extracted from this species was sufficient to attract E. emarginata to scent traps (Pereira et al. 2014).
Erioscelis was first revised by Saylor (1946) and again by Endrődi (1966Endrődi ( , 1985a. These works provide a strong foundation for species-level identification, but characters that separate Erioscelis from other cyclocephalines are largely undiscussed. For example, Saylor (1946) commented, "When compared with such species as Cyclocephala (Stigmalia) mafaffa Burmeister, or C. (Aclinidia) castanea (Fabricius), the only character definitely to separate Erioscelis is the unenlarged front tarsal claws of both sexes". Unique protibial (2 teeth on the lateral margin in both sexes, subapical position of reduced protibial spur) and abdominal (bisinuate margin of 6 th abdominal sternite, terminal spiracle not positioned on pleural suture) characters of Erioscelis emarginata also complicate recognition of the genus and may be reasons to doubt the monophyly of the group. These characters (except for the bisinuate margin of 6 th abdominal sternite) are associated with Anomalini (Rutelinae) and are absent in all other members of Erioscelis and Cyclocephalini more broadly. Sister-relationships of Erioscelis have not been hypothesized and the immature stages are unknown for the genus.
The monotypic genus Harposceles was erected for the species H. paradoxus. This striking, relatively large cyclocephaline occurs in lowland forests in Brazil, Ecuador, French Guiana, Peru, Suriname, and possibly Colombia (Burmeister 1847, Harold 1869b, Arrow 1937b, Blackwelder 1944, Endrődi 1966, 1985a, Endrődi and Dechambre 1976, Lachaume 1992, Couturier and Kahn 1992, Andreazze 2001, Andreazze and da Silva Motta 2002, Touroult et al. 2010, Ponchel 2011, Saltin and Ratcliffe 2012 (Fig. 60). Males display dramatic, and unique, characters of the protibia. Harposceles paradoxus males have elongated, arcuate protibia with the protibial spurs fused to the base of the tibia. Females are much less common than male specimens in collections, and males are readily attracted to lights at night, especially between midnight and 4 am (Andreazze 2001, Andreazze and   (Couturier and Kahn 1992). The larvae and pupae were found in the organic litter accumulated between leaf sheaths of A. carnosum (Couturier and Kahn 1992). The immature stages are undescribed.
Harposceles species can be recognized by the following combination of characters: 1) dorsal coloration dark piceous to black; 2) body dorsoventrally flattened; 3) clypeus rounded in dorsal view; 4) frontoclypeal suture incomplete medially; 5) apical margin of mentum truncate; 6) anterolateral margin of mandible lacking tooth; 7) mandibular molar area with surface lacking circular pits, with large, disorganized, canal-like invaginations; 8) galea of maxilla dorsoventrally flattened; 9) galea on inner surface at base with large, flattened, blade-like, tooth (less produced than in Surutu species); 10) galea on inner surface with 7 teeth in 2-1-1-1-2 arrangement from base to apex; 11) apical and basal beaded margins of pronotum complete at middle; 12) anterior membrane of the pronotum interrupted before lateral pronotal margins; 13) males with protrochanter with ventrally produced tooth; 14) protibia with 3 teeth in both sexes; 15) males with protibia elongated and arcuate; 16) protibial spur straight to weakly reflexed; 17) males with protibial spur fused to protibia, not articulated at its base; 18) males with inner protarsal claw thickened and not cleft at apex; 19) mesocoxae not widely separated, nearly touching; 20) metacoxae with lateral edge perpendicular to ventral surface; 21) apices of the meso-and metatibiae with a corbel; 22) anterior edge of hindwing distal to apical hinge lacking membranous border; 23) anterior edge of hindwing distal to apical hinge with decumbent setae surrounding vein and originating away from apical hinge; 24) vein RA with single row of pegs proximal to apical hinge.
The relationship of Harposceles to other cyclocephalines has not been elaborated upon in the literature. However, H. paradoxus shares some characters with Surutu that may be indicative of a close relationship between the two genera. The rounded shape of the clypeal apex in H. paradoxus is like the clypeal form in S. dytiscoides. The single row of RA pegs in H. paradoxus is shared between Ancognatha and Surutu, though Ancognatha species lack setae on the anterior edge of the hindwing distal to the apical hinge. The decumbent setae of the hindwing leading edge (distal to apical hinge) found in H. paradoxus is also found in Surutu species and the "Cyclocephala cribrata species group" (which included species previously placed in Mononidia and Surutoides) (Dechambre 1997). These groups also all share corbeled meso-and metatibial apices and entirely black coloration. Harposceles paradoxus shares other interesting characters with Surutu species. These shared characters include: 1) body strongly dorsoventrally flattened; 2) dorsoventrally flattened maxillary galea; 3) a seven-toothed maxillary galea in a 2-1-1-1-2 arrangement from the base to apex; 4) an incomplete frontoclypeal suture; and 5) the apical pronotal membrane interrupted before the lateral margins of the pronotum. The large basal tooth of the maxillary galea is much smaller and less produced in H. paradoxus than in Surutu species. Several male characters of H. paradoxus are autapomorphic in Cyclocephalini: 1) the protibial spur fusion to the protibial; 2) the arcuate, elongated protibia (seen also in some Dynastini); and 3) the ventrally produced protrochanter teeth. Peltonotus species are distributed throughout Southeast Asia, southern China, and the eastern portion of the Indian Subcontinent (Fig. 61). Peltonotus is currently considered the sole Asian lineage of Cyclocephalini, though its subfamilial classification has been unstable. The genus is remarkable for its confounding combination of morphological and behavioral traits that blurred the lines between historical concepts of the subfamilies Dynastinae and Rutelinae. For example, the sexual dimorphism of the protarsi in Peltonotus species has long been compared to that found in Cyclocephala (e.g., see Burmeister 1847). In contrast, the labral morphology of Peltonotus species matches that found in Asian parastasiine and fruhstoferiine (Rutelinae) scarabs (Arrow 1908(Arrow , 1910. The floral feeding behavior of Peltonotus species on Araceae is also shared between cyclocephalines and Asian parastasiines, adding a further layer of intrigue to unresolved evolutionary relationships between the groups at the subfamilial-and tribal-level (e.g., see Moore and Jameson 2013, Kumano-Nomura and Yamaoka 2006, Kumano-Nomura and Yamaoka 2009, Tung et al. 2010, Hoe et al. 2011, 2016.
Peltonotus was described by Burmeister (1847), and he included it within the Chalepidae division of Cyclocephalidae. The classification of Peltonotus was stable until Arrow (1908Arrow ( , 1910 transferred the genus to Rutelinae based upon the exposed (in dorsal view, produced apically beyond the clypeus) and chitinized labrum. Arrow (1917) later erected the "division" Peltonotini for Peltonotus within his classification of Rutelinae. Ohaus (1918Ohaus ( , 1934b and Machatschke (1972) rejected Peltonotini and included Peltonotus in Pelidnotina (Rutelini) in their catalogs of Rutelinae. Morphological phylogenetic analysis of Rutelina (Rutelinae: Rutelini) suggested that Peltonotus were more closely related to Cyclocephalini than Rutelini (Jameson 1998). Subsequent works on the genus have treated Peltonotus as a member of Cyclocephalini (Jameson and Wada 2004, Jameson and Jákl 2010, Jameson and Drumont 2013. Little is known about the biology and natural history of Peltonotus species. The immatures are undescribed. Adults are attracted to lights at night (Jameson and Wada 2004). Peltonotus malayensis Arrow was collected from the spathes of Epipremnum falcifolium Engl. (Araceae), where males and females were observed mating and feeding (Jameson and Wada 2004). In Thailand, P. nasutus visit the large inflorescences of the terrestrial aroid Amorphophallus paeoniifolius (Dennst.) Nicolson, where adult beetles feed and mate (Grimm 2009). Peltonotus nasutus can be attracted to the inflorescences in high numbers (over 70 individuals) (Danell 2010).
Peltonotus species can be recognized by the following combination of characters: 1) dorsal coloration brown to black with variable presence of maculae; 2) body convex, not dorsoventrally flattened; 3) clypeal apex rounded to straight in dorsal view; 4) frontoclypeal suture incomplete medially; 5) apical margin of mentum variably shaped with weak emargination; 6) anterolateral margin of mandible lacking tooth; 7) mandibular molar area with rows of circular micropunctures; 8) galea of maxilla not strongly dorsoventrally flattened; 9) galea of the maxilla on inner surface with 3 fused basal teeth, a free median tooth, and 2 fused apical teeth (3-1-2 arrangement); 10) galea with articulated medial tooth; 11) labrum extending apically beyond clypeal apex (obvious in dorsal view); 12) apical and basal margins of pronotum with beaded margin complete or incomplete at middle; 13) protibia of males with 2 or 3 teeth, females with 3 teeth; 14) protibial spur straight to weakly reflexed; 15) males with inner protarsal claw thickened and not cleft at apex (nib variably present or absent); 16) mesocoxae not widely separated, nearly touching; 17) metacoxae with lateral edge perpendicular to ventral surface; 18) anterior edge of hindwing distal to apical hinge lacking membranous border; 19) anterior edge of hindwing distal to apical hinge with row of long setae extending from apical hinge along length of the costal vein; 20) vein RA with single row of pegs proximal to apical hinge. Arrow, 1908 Types species. Ruteloryctes tristis Arrow, 1908: 336, by monotypy. Valid taxa. Two species.

Ruteloryctes
The two species of Ruteloryctes are distributed in the Guinea-Congo lowland rainforests of West and Central Africa. Ruteloryctes specimens have been collected in Angola, Benin, Cameroon, Chad, Côte d'Ivoire, Democratic Republic of the Congo, Guinea, Guinea-Bissau, Nigeria, Senegal, Sierra Leone, and The Gambia (Burgeon 1947, Paulian 1954, Endrődi 1960, 1985a, Krell et al. 2003, Ervik and Knudsen 2003 (Fig. 62). Ruteloryctes morio is a pollinator of nocturnally blooming Nymphaea lotus L., and this floral association has been reported from Côte d'Ivoire, Senegal, and Nigeria (Fabricius 1798, Krell et al. 2003, Ervik and Knudsen 2003. The immature stages of Ruteloryctes are undescribed. Ruteloryctes species can be recognized by the following combination of characters: 1) dorsal coloration black to dark brown; 2) body convex, not strongly anteroposteriorly compressed or dorsoventrally flattened; 3) clypeal apex truncate or rounded in dorsal view; 4) frontoclypeal suture incomplete medially; 5) males with anterolateral margin of the mandibles lacking weak tooth; 6) mandibular molar area with rows of circular micropunctures; 7) apex of mentum weakly emarginated at middle; 8) galea of maxilla on inner surface with 3 fused basal teeth, a free median tooth, and 2 fused apical teeth (3-1-2 arrangement); 9) pronotum with broadly incomplete beaded basal margin; 10) males and females with 3 protibial teeth on lateral margin, basal tooth not greatly reduced, slightly removed from apical 2 teeth, and oriented laterally; 11) protibial spur straight to weakly deflexed; 12) males with inner protarsal claw enlarged and narrowly cleft at apex; 13) mesocoxae not widely separated, nearly touching; 14) meso-and metatibiae with distal, transverse carinae; 15) metacoxae with lateral edge perpendicular to ventral surface; 16) anterior edge of hindwing distal to apical hinge lacking setae and with produced, membranous border; 17) vein RA with single row of pegs proximal to apical hinge.
The original description of Ruteloryctes compared the genus to New World Dyscinetus species, and it was hypothesized to have "strayed across the Atlantic" (Arrow 1908). Endrődi (1966) thought that Ruteloryctes was one of the most "primitive" cyclocephaline genera. The 3-1-2 arrangement of the teeth on the maxillary galea in Ruteloryctes is most similar to Arriguttia, Augoderia, and many Cyclocephala species. The membranous border of the hindwing present in Ruteloryctes is also shared with Arriguttia, Acrobolbia, Ancognatha, Aspidolea, and Cyclocephala. However, the single row of pegs present on the hindwing RA vein in Ruteloryctes is present in Ancognatha, Surutu, Harposceles, Stenocrates, Dyscinetus, Erioscelis, and Chalepides.
The enigmatic genus Stenocrates comprises 52 taxa distributed from Mexico south throughout South America (except Chile) and Jamaica (Fig. 63). Species diversity in the group is highest in the tropical forests of Brazil, especially the northern and western states of Amazonas, Pará, Acre, and Rondônia. Many Stenocrates species are also known from eastern Brazil, especially Bahia, Espírito Santo, São Paulo, and Santa Catarina. Stenocrates species are problematic to identify due to conserved external morphology among species, making the group, "…possibly the most difficult genus of Dynastinae in the Americas with which to work" . Male paramere morphology is diagnostic for species-level identification in the genus, and females not associated with males at the time of collection cannot be reliably identified with existing literature. Nothing is known about the natural history and biology of Stenocrates species. Adults can be collected at lights at night (Endrődi 1969a, Ratcliffe and Cave 2006, Ratcliffe 2014. Immature stages are undescribed for the genus. Stenocrates was erected by Burmeister (1847) for species that he considered highly similar to the historical concept of Chalepus, except for the lack of dimorphic protarsi. Burmeister (1847) included 4 species in Stenocrates and speculated that Melolontha rufipennis Fabricius could also be a member of the genus. Descriptions of new species of Stenocrates were slow to accumulate in the 19 th and early 20 th century. Kirsch (1870) described the sixth Stenocrates species from Colombia. Bates (1888) examined S. laborator specimens from Mexico and noted that the simple protarsi of the males and dorsoventrally flattened tibiae separated diagnosed Stenocrates within Cyclocephalini. Stenocrates was compared to Euetheola by Bates (1888) stating that the form of the mandibles and the proximal tarsomeres served to separate these genera. Arrow (1911Arrow ( , 1913 added two new species to Stenocrates, but he did not offer a diagnosis for the genus or make meaningful character comparisons for the genus. Stenocrates was revised by Endrődi (1966Endrődi ( , 1985a, and many new species have been described since that work, which have not been incorporated into a comprehensive identification key. Stenocrates species can be recognized by the following combination of characters: 1) dorsal coloration black or dark brown and without maculae; 2) body convex, not strongly anteroposteriorly compressed or dorsoventrally flattened; 3) clypeus trapezoidal with apex truncate in dorsal view; 4) frontoclypeal suture complete medially; 5) males with anterolateral margin of the mandibles lacking weak tooth; 6) mandibular molar area with rows of circular micropunctures; 7) mandibular molar area on proximal margin without semicircular depressed pits; 8) galea of maxilla on inner surface with 2 fused basal teeth, 2 fused medial teeth, and 2 fused apical teeth (2-2-2 arrangement); 9) pronotum with broadly incomplete beaded basal margin; 10) pronotum with narrowly incomplete beaded apical margin; 11) males and females with 3 protibial teeth on lateral margin, basal tooth not greatly reduced, only slightly removed from apical 2 teeth, and oriented laterally; 12) protibial spur straight to weakly deflexed; 13) males and females with protarsal claws simple (not cleft) and not enlarged; 14) mesocoxae not widely separated, nearly touching; 15) metacoxae on lateral edge without transverse, depressed sulcus; 16) metacoxae with lateral edge perpendicular to ventral surface; 17) meso-and metatibiae with distal, transverse carinae; 18) meso-and metatibiae dorsoventrally flattened and laterally expanded; 19) anterior edge of hindwing distal to apical hinge with erect setae and lacking produced, membranous border; 20) vein RA with single row of pegs proximal to apical hinge; 21) propygidium not expanded, propygidium and pygidium not rigidly fused.
The five species of the South American genus Surutu are distributed in Colombia, Bolivia, and Brazil (Martínez 1955, D'Andretta and Martínez 1956, Endrődi 1966, 1985a, Ratcliffe 1981, Andreazze 2001, Otavo et al. 2013 (Fig. 64). These spectacular black species are truly the monsters of the Cyclocephalini, with some specimens of Surutu seabrai D'Andretta and Martínez measuring over 4 cm in length. Nothing is known about the biology of Surutu species. At least some species are attracted to lights at night (Ratcliffe 1981). The immature stages are undescribed for the genus as currently circumscribed.
Surutu species can be recognized by the following combination of characters: 1) dorsal coloration dark piceous to black; 2) body dorsoventrally flattened; 3) clypeus rounded to parabolic in dorsal view; 4) frontoclypeal suture incomplete medially; 5) apex of mentum narrowly and deeply emarginated (in S. dytiscoides and S. seabrai; other species unknown); 6) anterolateral margin of mandible lacking tooth; 7) galea of maxilla dorsoventrally flattened (in S. dytiscoides and S. seabrai; other species unknown); 8) galea on inner surface at base with large, flattened, blade-like, bifurcated tooth (in S. dytiscoides and S. seabrai; other species unknown); 9) galea on inner surface with 7 teeth in 2-1-1-1-2 arrangement from base to apex (in S. dytiscoides and S. seabrai; other species unknown); 10) apical and basal beaded margins of pronotum incomplete at middle (in S. dytiscoides and S. seabrai; other species unknown); 11) anterior membrane of the pronotum interrupted before lateral pronotal margins (in S. dytiscoides and S. seabrai; other species unknown); 12) protibia with 3 teeth in both sexes; 13) protibial spur straight to weakly reflexed; 14) males with protibial spur articulated at base, not fused to protibia; 15) males with inner protarsal claw thickened and narrowly cleft at apex (claw apex entire in S. fenni Ratcliffe and S. schulzei Endrődi); 16) mesocoxae not widely separated, nearly touching; 17), metacoxae with lateral edge perpendicular to ventral surface; 18) apices of the meso-and metatibiae with a corbel (in S. dytiscoides and S. seabrai; other species unknown); 19) anterior edge of hindwing distal to apical hinge lacking membranous border; 20) anterior edge of hindwing distal to apical hinge with decumbent setae surrounding the vein and originating away from the hinge; 21) vein RA with single row of pegs proximal to apical hinge.
Some characters of the head, mouthparts, and elytra of Surutu have been compared to Ancognatha, Cyclocephala, and Mimeoma (Martínez 1955, D'Andretta and Martínez 1956). The parabolic and rounded clypeal apex in Surutu species is like the clypeal form in several Ancognatha species. Surutu dytiscoides and S. seabrai, at least, have a deeply emarginated apex of the mentum that is also shared with Ancognatha species. The single row of RA pegs is also shared between Ancognatha and Surutu, although Ancognatha species lack setae on the anterior edge of the hindwing distal to the apical hinge. Instead, Ancognatha have a hindwing membrane like that found in Cyclocephala, Augoderia, Arriguttia, Aspidolea, and Acrobolbia. The dramatic dilations and knobs on the elytral epipleuron of S. seabrai are similar to those found in some Ancognatha and Cyclocephala species.
The distinctive setae of the hindwings found in Surutu are also found in Harposceles and species of the "Cyclocephala cribrata species group" (which included species previously placed in the genera Mononidia and Surutoides) (Dechambre 1997). These groups also share corbeled meso-and metatibial apices and entirely black coloration. Harposceles paradoxus shares other interesting characters with Surutu species, suggestive of a close relationship between the two genera. These shared characters include: 1) body strongly dorsoventrally flattened; 2) dorsoventrally flattened maxillary galea; 3) a 7-toothed maxillary galea in a 2-1-1-1-2 arrangement from the base to apex; 4) an incomplete frontoclypeal suture; and 5) the apical pronotal membrane interrupted before the lateral pronotal margins.
Platyphileurus felscheanus Ohaus (Dynastinae: Oryctini) warrants special discussion here. This species was described twice. Platyphileurus felscheanus was described from specimens collected from Santa Catarina, Brazil (Ohaus 1910). This new genus was compared to Phileurus Latreille and later included in the tribe Phileurini (Ohaus 1910, Arrow 1937b. Endrődi (1975) later described Surutu jelineki from Rio de Janeiro based on two female specimens. Comparison of the types of these species revealed that they are conspecific, with the name Platyphileurus felscheanus having priority over Surutu jelineki (Grossi et al. 2010).
The immatures of Platyphileurus felscheanus are associated with bromeliads (Grossi et al. 2010, Albertoni et al. 2014. Based on examination of larval, pupal, and adult characters, P. felscheanus was excluded from Phileurini and proposed to be a member of Oryctini (Albertoni et al. 2014). However, there are some intriguing adult character similarities between P. felscheanus and other Surutu species. For example, P. felscheanus is black, dorsoventrally flattened, and has dimorphic protarsal claw morphology (enlarged in males, simple in females) (Endrődi 1975, Grossi et al. 2010, Albertoni et al. 2014. The apices of the metatibiae in P. felscheanus are "weakly dentate" (Albertoni et al. [2014]: figure 30). Alternatively, the outer edge of the metatibia figured in Albertoni et al. (2014) could be considered not to be "weakly dentate", but corbeled (outer edge produced beyond the inner edge of the tibial apex). This tibial character is found in Surutu, Harposceles, and in the "Cyclocephala cribrata species group". The venter of the meso-and metatarsi in P. felscheanus is covered with dense, reddish, flattened setae (Albertoni et al. 2014). Similar flattened, scale-like setae are also found on the venter of the meso-and metatarsi of S. seabrai and S. dytiscoides. Future analyses of the tribal placement of P. felscheanus should focus on adult character comparisons with Surutu species and H. paradoxus, especially characters of the mandibles, maxillary galea, tibiae, tarsi, parameres, and hind wings.