An annotated checklist of the leaf beetles (Coleoptera, Chrysomelidae) from El Salvador, with additions from the Bechyné collection in the Royal Belgian Institute of Natural Sciences

Abstract A checklist of the species of leaf beetles (Coleoptera: Chrysomelidae) of El Salvador is presented based on data from literature and a digitization project of the Bechyné collection of the Royal Belgian Institute of Natural Sciences (RBINS). The RBINS collections contain a total of 2797 individual chrysomelid specimens from El Salvador, sorted into 89 species and 132 genera. In total, the current checklist contains 420 species, of which 33 are new records for El Slavador from the Bechyné collection. In these collections, there are also ten nomina nuda named by Bechyné, which need further study. The leaf beetle diversity in El Salvador, partly due to the country’s unstable political history, remains poorly studied, and many (new) species await discovery. This checklist provides a baseline for further study in El Salvador and nearby region.


Introduction
The description and inventory of biodiversity is facing hard times due to budgetary problems and the decline of taxonomists (Drew 2011). This is also certainly the case for leaf beetles (Coleoptera: Chrysomelidae). Nonetheless, leaf beetles can be of economic significance (Livia 2006), often possess interesting life histories (e.g., the semi-aquatic lifestyle of most Donaciinae, Kleinschmidt and Kölsch (2011)), and can display complex behavior like eusociality (Windsor et al. 2013). Leaf beetles are a taxonomically complex group and, although some regions where the diversity of this group is relatively well known (like the Palearctic and Nearctic regions), other areas like the Neotropics are in desperate need of more study.
One of the most important early studies on Neotropical Chrysomelidae was done for the multipartite series of the Biologia Centrali Americana (Baly andChampion 1885-1894;Jacoby 1880Jacoby -18921888-1892. The series remains the most complete source of information today on insect fauna of Central America. This work collated information from previous taxonomists, e.g., Crotch, Illiger, Harold and Baly, who published many monographs and species descriptions within Alticini, Eumolpinae, Cryptocephalinae and other subfamilies. In approximately the same period (1850-1862), Boheman had published his 'Monographia Cassididarum' dealing with the subfamily Cassidinae (Boheman 1850(Boheman -1862. These early works were followed by the 'Coleopterorum Catalogus', published in multiple volumes, which represented the first actual checklist (Clavareau 1913a(Clavareau , 1913b(Clavareau , 1914Csiki 1939, 1940;Pic 1913;Spaeth 1914;Weise 1911Weise , 1916Weise , 1924. Further checklists include Blackwelder (1946), but, as stated by Furth and Savini (1996), this list contained mainly information derived from the 'Coleopterorum Catalogus'. After the 1940's, the works on Neotropical Chrysomelidae done by Jan Bechyné and his wife Bohumila Bechyné produced an impressive list of 188 publications about the Chrysomelidae from the Neotropical, Afrotropical and Palearctic regions, with a main focus on the subfamilies Eumolpinae and Galerucinae including Alticini (Seeno et al. 1976). Jan and Bohumila Bechyné described many species, and therefore significantly contributed to the knowledge of Neotropical Chrysomelidae. Another significant work is Scherer (1962) with an identification key to the genera of Neotropical Alticini (Scherer 1962). One of the most prolific taxonomists working on Neotropical leaf-beetles was Francisco Monrós (1922Monrós ( -1958. The Bechnynés' works focused mainly on describing new taxa, and less on revising previous works and often ignored biological aspects of the taxa. Francisco Monrós focused his studies on detailed complex revisions of particular groups and include numerous ecological observations on host plants and behavior of the species. He published mainly on Criocerinae, Clytrini, Fulcidacini, and Lamprosomatinae. He was in the process of writing his opus magnum 'Los géneros de Chrysomelidae', but unfortunately due to his premature death only the first volume was published covering Sagrinae, Donaciinae, and Criocerinae (Monrós 1960).
Records from El Salvador are relatively rare (e.g., see Rodrigues and Mermudes 2016)). The country is undersampled for insects and this can be demonstrated by the fact that only two chrysomelid beetles are recorded from El Salvador in the 'Biologia Centrali Americana' series. The most intensive effort to construct a national list for the family dates back to 1960 with three main works by the Bechynés Bechyné and Bechyné 1960;1963). Another list of Chrysomelidae from El Salvador is found in the two-part series 'Lista de Insectos Clasificados de El Salvador' Salazar 1957, Berry 1959). However, these are inadequate because most Chrysomelidae species identifications include the statements "probably" and "proximately", lacking information on who identified the species and depository of the specimens. Moreover, there are many misspellings and misinterpretations in species and author names and some species listed in Berry and Salazar (1957) are not included in Berry (1959) without any explanation. The occurrence data of these listed species should be verified by examination of material. Part of the material collected by Berry is deposited in the National Museum of Natural History (UNSM), Washington DC, USA. The Berry lists were unknown to most of subsequent authors, who explicitdly recorded species mentioned in the lists as new to El Salvador. This is particularly obvious in Cassidinae, which represent one of the best known chrysomelid subfamilies of the world.
Much of the Bechyné material from El Salvador was unfortunately lost during their trip from Europe to Brazil (Furth 2018, in litteris), but a total of 32 boxes remains in the Royal Belgian Institute of Natural Sciences (RBINS), and they contain hundreds of specimens of the Bechyné collection. These include numerous paratypes of the subfamily Eumolpinae and tribe Alticini.
We carried out a literature search, seeking any records of leaf beetles for El Salvador, with the main goal of constructing an updated and annotated checklist. Additionally, we added species data from the Bechyné El Salvador collection from the RBINS. The results of this work are presented here.

Literature search
During our literature study, we accessed multiple historical references from eminent contributors to chrysomelid taxonomy, such as Bechyné, Stål, Jacoby, Baly, Blake, Harold, Illiger, and others as mentioned in the introduction. We also scouted some existing checklists from neighboring countries for references of El Salvadorean Chrysomelidae (e.g., Furth (2006), Furth (2013, Maes and Staines (1991)) as well as subfamily checklists for Central America level (e.g., Furth and Savini (1996)). A complete list of references containing records for El Salvador are indicated in our Results section. Species in Berry and Salazar (1957) and Berry (1959) which used the statements "probably" and "proximately" were not added in the present checklist to avoid incorrectly identified species.

Specimen digitization
Eighteen full and 14 partially-filled insect drawers with material from El Salvador were digitized by volunteers (following the protocol described in Merckx et al. (2018)). For this, at least one specimen per species was photographed in dorsal, lateral and frontal views, using an Olympus TG4 digital camera with focus stacking functionality (see Mertens et al. (2017), where this method is discussed thoroughly). Pictures of these specimens were stacked using the Helicon Focus software (HeliconSoft Ltd, Kharkiv, Ukraine). A total of 42 specimens from El Salvador belonging to the genus Calligrapha Chevrolat that had already been digitized in a previous project were added to the database (Merckx et al. 2018). Furthermore, a high-resolution picture was taken of every insect drawer; such that all specimens have a clear dorsal picture, making intraspecific variation clear. The original labels of all specimens in the Bechyné El Salvador collection were also digitized, however for this manuscript the months of dates were changed to roman numerals to avoid confusion. The pictures and accompanying data are publicly available on-line at the RBINS online database (http://collections.naturalsciences.be/ssh-entomology).

Taxonomy
We followed the system of Bouchard et al. (2011) for division of subfamilies and tribes. Names found in literature were screened for any nomenclatural changes in their relevant literature and/or were checked by experts on the respective groups (see acknowledgements). If a species record in literature had an outdated nomenclature, the name under which it was recorded is given under remarks at the relevant species checklist. mens included in the RBINS can be found below. Included in the collection were ten Walterianella sp. (Manuscript species -Galerucinae, Alticini). We chose to not give the names and photographs of these undescribed species here to avoid the cluttering of invalid names in the literature.
We noted a severe undersampling of Cryptocephalinae and Criocerinae. For example, Vencl et al. (2004) (Criocerinae) indicated the range of a species distribution (e.g., Mexico to Panama), but lacked specific records from El Salvador of which the authors are aware. Such unspecific records were left out of the present checklist. In total, the literature search led to a total of 385 chrysomelid species know to occur in El Salvador. Together with the Bechyné collection, this led to a combined total of 420 species (Figure 1 and Suppl. material 1: Table S1).
Suppl. material 1: Table S1 displays the number of species and genera per subfamily for each district. The departments with the highest species count to date are San Salvador (182 species), La Libertad (105 species), Santa Ana (114 species), Ahuachapán (54 species), La Paz (50 species) and Chalatenango (46 species). No records from both San Miguel and Cabañas could be found, nor were there any specimens from these departments included in the RBINS collections. Below is the checklist of species from El Salvador taken from literature (the reference indicated between brackets after every record) and from the Bechyné collection in the RBINS. Additionally, some extra records of Cassidinae were added from other sources known to one of the authors. If the authors were aware of any nomenclatural changes, this is indicated under "remarks".

Acanthoscelides brevipes (Sharp, 1885)
Remarks. Johnson (1990) mentions this species as being present in El Salvador, but does not give specific records.

Acanthoscelides clitellarius (Fahraeus, 1839)
Remarks. Johnson (1990) mentions this species as being present in El Salvador, but does not give specific records. Johnson, 1983 Remarks. Johnson (1990) mentions this species as being present in El Salvador, but does not give specific records.

Acanthoscelides difficilis (Sharp, 1885)
Remarks. Johnson (1990) mentions this species as being present in El Salvador, but does not give specific records.

Acanthoscelides griseolus (Fall, 1910)
Remarks. Johnson (1990) mentions this species as being present in El Salvador, but does not give specific records.

Acanthoscelides guazumae Johnson & Kingsolver, 1971
Remarks. Johnson (1990) mentions this species as being present in El Salvador, but does not give specific records. 8. Acanthoscelides guerrero Johnson, 1983 Remarks. Johnson (1990) mentions this species as being present in El Salvador, but does not give specific records. 9. Acanthoscelides hectori Kingsolver, 1980 Remarks. One specimen from El Salvador was intercepted at USDA Plant Quarantine, Washington DC (Kingsolver 1980 Remarks. Berry (1959) mentions this species as being present in El Salvador, but does not provide specific records. Certainly species of the subgenus Xenocassis Spaeth, 1936 must occur in El Salvador but so far formally none was recorded. The species are very similar to each to other and the identification is not easy without comparative material. Charidotella (X.) ambita is distributed in southern part of Central America (S Nicaragua to Panama), however, its occurrence in El Salvador cannot be excluded but is improbable.

Charidotis vitreata (Perty, 1830)
Remarks. Berry (1959)  Remarks. The three specimens from El Salvador have the black stripe on their pronotum rather thin and only apical 1/5 of elytra black while the holotype of O. atroterminata has a thick medial stripe on pronotum and apical 1/3 of elytra black and the dark colouration extending along suture forwards. However, Salvadorian specimens show variability in black colouration: one has the pronotal stripe only slightly indicated; two have the apical black colouration on elytra rather emarginate near suture anteriorly while the third specimen has a clearly projecting black stripe along suture. Therefore, we consider these as intraspecific variability as other characters (e.g. general shape, punctation of elytra and pronotum) are similar to the holotype.

Physonota attenuata Boheman, 1854
Published records. SONSONATE: Sonsonate (Berry and Salazar 1957 Xenochalepus (s. str.) omogerus (Crotch, 1873) Published records. El Salvador, without further data (Berry 1959). LA LIBER-TAD: Santa Tecla (Berry and Salazar 1957). SAN SALVADOR: San Salvador env., 17/VIII/1958, 3 spec. (Uhmann 1961 (Baly, 1885), which is considered a synonym of X. rufithorax. Remarks. Bechyné and Bechyné (1965a) also report C. argus from El Salvador, but do not state specific localities. Remarks. Bechyné and Bechyné (1965a) also report C. fulvipes from El Salvador, but do not state specific localities. Most possibly this originates from the localities from  which are cited here. For a full inventory of the Calligrapha spec. in the RBINS collections, one should consult Merckx et al. (2018), where the records cited here from El Salvador are also included.

Discussion
The digitization of the Bechyné collection in the RBINS revealed that there were a total of 2797 individual specimens sorted into 89 species. Among these were ten apparent nomina nuda which could not be found in any publication from J. Bechyné Walterianella sp. (Manuscript species -Galerucinae, Alticini). These are species which Bechyné either forgot to formerly describe, did not have time to describe, or the validity of which he questioned. Regarding the former list, the authors believe that Walterianella sp. (Manuscript species) could just be a variation of W. venustula (Schaufuss) which it closely resembles, but this should be confirmed by comparison with the type material and especially the structure of the genital structures. These and all of the other "paratypes" of the nomina nuda should be examined and revised in the future by experts.
The study of relevant literature led to a checklist of 385 species known from El Salvador. A total of 43 species from these 309 were also present in the collections in the RBINS. Material from the Bechyné collection added a further 33 species (excluding the ten nomina nuda) to the literature-based checklist of chrysomelids. This leads to a preliminary checklist of a total of 420 species of Chrysomelidae currently known for El Salvador (see table 1 for a full overview). Incorporated were also records from Berry and Salazar (1957) and Berry (1959), references which were frequently overlooked in the past.
Surprisingly few records of the subfamilies Criocerinae, Lamprosomatinae, and Cryptocephalinae could be found, despite their high prevalence in Central America. This is most likely due to the fact that most chrysomelid research in El Salvador has been done by J. and B. Bechyné, who focused mostly on Eumolpinae and Galerucinae including Alticini (of which respectively 16, 9, and 2 species could be newly added to the El Salvador checklist by Bechyné's collection in the RBINS). The latter are relatively well represented in comparison with the currently known number of species from neighbouring countries (see Furth and Savini (1996)), mainly because of the extensive work of the Bechynés.
We noted a strong bias towards collection efforts in the departments San Salvador (182 species), La Libertad (105 species) and Santa Ana (114 species), and to a lesser extent in the districts Ahuachapán (54 species), La Paz (50 species) and Chalatenango (46 species). No records from San Miguel or Cabañas could be found. Future surveys in the country should thus also be focussed on the two latter departments.

Conclusions
Our study reveals a preliminary total of 420 species of Chrysomelidae known to El Salvador. However, this number should be approached with caution, since the taxonomy of some subfamilies is not yet fully clear (e.g., Eumolpinae), some subfamilies seem to lack sampling effort in the country (e.g., Cryptocephalinae, Criocerinae and Lamprosomatinae), and in general there has been little study on the fauna of El Salvador, possibly due to its political instability and safety issues for field research (Bourgois 2001). Nonetheless, we believe that this checklist, although almost certainly incomplete, will serve as a baseline for further study in the area.