Hispines (Chrysomelidae, Cassidinae) of La Selva Biological Station, Costa Rica

Abstract Survey work from 1992–2001 identified 139 species of hispines at the lowland part of La Selva Biological Station, Costa Rica. The tribe Cephaloleiini was the most speciose with 58 species (41.7%) followed by the Chalepini with 55 (39.5%). The fauna is most closely related to that in South America but with some genera which are more speciose in the Nearctic Region. Plant associations are known for 88 (63.3%) of the species but many of these are merely collecting records, not host plant associations. The first plant associations are reported for Alurnus ornatus, Alurnus salvini, and Acentroptera nevermanni.


Introduction
Hispines comprise half of the subfamily Cassidinae (sensu lato) in the family Chrysomelidae within the order Coleoptera (Staines 2002b). Until recently, most authors treated the group as a separate subfamily but recent work has shown that there is no biological or morphological reason to retain sub-familial status (Staines 2002b). The combined subfamily consists of 6000 species placed in 42 tribes (Staines 2002b). See Staines (2002b) for a detailed history of the classification of the two groups.
The combination of the Hispinae with the Cassidinae (s. str.) has created difficulty in having a handy term to use for these beetles. Several have been proposed but they Daily operations of ALAS were conducted by the parataxonmists under the direction of the principal investigators John T. Longino, Evergreen State College, and Robert K. Colwell, University of Connecticut. Over the course of the project there were over 100 collaborating taxonomists.

Survey Methods
The ALAS survey consisted of both structured and directed sampling. Structured sampling consisted of black-lights, Malaise and flight intercept traps, and canopy fogging (see Furth et al. 2003 for summary).
Passive black-lights were utilized from 1993 to 1999 at twelve sites, six on the ground and six in the canopy. Malaise traps were used at sixteen sites from 1993 to 2000. Specimens were collected directly into ethanol and the traps emptied every two weeks. Flight intercept traps were place at sixteen sites and samples were collected every two weeks.
Canopy fogging was conducted in 1993-1994, 1996, and 2000. Sixteen trees were fogged: six trees of the most common species at La Selva, six trees of an intermediate abundant species, and trees of six different species. The tree selected had large crowns with little crown overlap and with good climbing access.
Directed collecting for chrysomelids used beating, sweeping, visual observation, known host plant observation, and use of a mid-canopy shaker net.

Species richness at La Selva
As of the end on 2001, a total of 139 hispine species have been collected at La Selva Biological Station (see Table 1).
Quantitative inventory by non-specialists using standard sampling techniques can capture about half of the fauna. Individual methods are needed to sample the rest of the community. Sweeping, beating, and host plant sampling are the best methods. Fogging, Berlese funnels, and Malaise traps capture a few species usually not otherwise collected but are not sufficient in themselves to indicate the actual fauna.

Major lineages
The most recent classification of hispines is by Würmli (1975) and Staines (2002b). There are 24 extant tribes of hispines, of which six have been found at La Selva (see Table 1).
Over 40% of the 139 hispine species and 25% of the genera are in the tribe Cephaloleiini. The Cephaloleiini is a New World tribe of 16 genera and 382 species (Staines 2002b). Over 200 species are in the genus Cephaloleia Chevrolat (Uhmann 1957, Staines 1996.  Uhmann 1930;Maulik 1932;Strong 1977bStrong ,1982aMeskins et al. 2008;Descampe et al. 2008;García-Robledo et al. 2010 Heliconia At La Selva Cephaloleia is the most speciose genus with 44 species from La Selva. The biology of various Cephaloleia species has been studied by Strong (1977aStrong ( , b, 1982aStrong ( , 1983, Seifert and Seifert (1976), Strong and Wang (1977), Auerbach and Strong (1981), and Morrison and Strong (1981). Since the only identification aid available to these workers was Baly (1885), which covered less than half of the species known from Central America, some of the published names are not associated with the correct species. However the published information does give valuable data on the general biology and ecology of Cephaloleia species. Staines (2004a) attempted to associate the biological data with the correct species. Additional biological work and host plant associations have been done by Johnson (2004a, b), Johnson and Horvitz (2005), McKenna and Farrell (2005), Descampe et al. (2008), Meskins et al. (2008), García-Robledo &Horvitz (2009, in press), andGarcía-Robledo et al. (2010).
Cephaloleia eggs are flat, with a thin chorion; hence they are subject to desiccation. Eggs are laid on host surfaces. Oviposition sites vary among beetle species and host plant. The most common oviposition sites are leaf surfaces, petioles of immature leaves or inflorescence bracts. Eggs hatch in 10 to 20 days. Larvae begin feeding immediately upon the part of the plant on which the egg was laid. Cephaloleia larvae have a water penny-like appearance. They are flat and well adapted to moving between the wet surfaces of Zingiberales leaves, stems, and flowers. Larvae grow very slowly and go through up to eight molts depending on the size of the species and the part of the plant fed on. During their development, larvae of leaf and stem-feeding species utilize several leaves or even leaves on adjacent plants. Inflorescence-feeding larvae are restricted to a single inflorescence. Larvae of Cephaloleia species feed on the plant by dragging their mandibles across the plant surface while they crawl forward. This results in an irregularly shaped feeding scar and a trail of frass. Pupation occurs above ground, usually on the stalk of the host plant and lasts about 20 days. Adult Cephaloleia are found in the same habitat as larvae and cause similar feeding damage. Several different Cephaloleia species as well as other genera may utilize the same leaf, so larval associations require rearing (Strong 1977a(Strong , b, 1982a(Strong , 1983Strong and Wang 1977;Auerbach and Strong 1981;and Morrison and Strong 1981).
Seven other genera of Cephaloleiini containing 14 species are known from La Selva. Most of these species are poorly known and not associated with their host plant.
The tribe Arescini consists of four genera and 17 species from the Neotropics (Staines 2002b). One genus and two species are known from Mesoamerica. None of the genera have been revised and little work has been done on the biology. Chelobasis bicolor Gray and C. perplexa Baly are found at La Selva. Strong (1977aStrong ( , 1983 reported the larval host plants of C. bicolor as Heliconia latispatha Benth. and H. tortuosa Griggs (Heliconiaceae). Strong (1983) reported on the biology of this species indicating that eggs are laid on wet, tender tissue of the host plant and hatch in about 20 days. Larvae begin feeding in rolled leaves immediately after hatching. Development is slow, requiring at least eight months until pupation. Larvae require more than one leaf-roll to complete development and move from maturing leaf-rolls to more tender ones at night. If they are between leaf-rolls at daylight, they hide between the petiole and stalk until nightfall. Adults are polymorphic (in color and size) and long-lived; in markrecapture studies adults were found 18 months after marking.
Chelobasis perplexa is known to feed on Calathea insignis Hort. & Bull. (Marantaceae) and H. imbricata (Kuntze) Baker in Costa Rica (Maulik 1932). Strong & Wang (1977) and Auerbach and Strong (1981) reported H. latispatha as a larval host plant. The biology of this species is similar to that of C. bicolor.
The tribe Alurnini consists of six genera and 29 species (Staines 2002b) and contains some of the largest chrysomelids (25-45 mm). The tribe was revised by Fischer (1935) and I am in the process of revising it. Published life histories record various genera and species feeding on palms (Arecaceae) (Fischer 1935, Villacis Santos 1968, Macedo et al. 1994. Both Mesoamerican species, Alurnus ornatus Baly and A. salvini Baly, have been collected at La Selva. Alurnus salvini is the more commonly collected species. The New World tribe Prosopodontini contains the genus Prosopodonta Baly with 26 species found from Nicaragua to Ecuador (Staines 2002b). The genus is in need of revision.
Two species, P. distincta (Baly) and P. dorsata (Baly), have been collected at La Selva. McCoy (1984 reported P. dorsata (as Cheirispa) adults and larvae feeding in accumulated leaf debris on the top of Heliconia leaves in Costa Rica and Ecuador. All other species of Prosopodonta have been reported as leaf-miners on various Arecaceae (Jolivet and Hawkeswood 1995). The photograph in McCoy (1984) is a Prosopodonta adult however the larval photograph does not resemble the known Prosopodonta larvae (Maulik 1931). All other species of Prosopodonta are associated with Arecaceae and I have only found P. dorsata on unfurled palm fronds, never on Heliconia.
The tribe Sceloenoplini contains five genera and 299 species, with 154 species in the genus Sceloenopla Chevrolat (Staines 2002b). They are leaf-miners in a variety of plant families. This tribe is represented at La Selva by four genera and 20 species (see Table 1). There are 17 species of Sceloenopla known from La Selva. The biology is unknown for all species.
The tribe Chalepini consists of 55 genera and nearly 1000 species in the New World (Staines 2002b). Very few genera have been revised. All species studied are leaf-miners and appear to prefer dicots (Jolivet and Hawkeswood 1995). This tribe is represented at La Selva by 18 genera and 55 species (see Table 1). Chalepus is the most speciose genus with 14 species.

Habitat specificity
Hispines can be found in most non-aquatic habitats at La Selva. There are 46 species which feed on rolled leaves and inflorences of Zingiberales. This one feeding guild accounts for 33% of the hispine species known from La Selva.
Most hispines species seem to be restricted to understory to mid-canopy level plants. Work on hispines has shown many species to be monophagous or narrowly oligophagous. These species are found mostly in relation to their host plants. Other hispines are broadly oligophagous or polyphagous and can be found in many habitats.
A continuing problem in inventory work is determining if the specimen collected was actually on its host plant or was a transient. Much of the earlier literature on host associations does not specify whether the insect was feeding as an adult, was breeding on the plant, or merely resting on it.
Relatively few species have only been collected from canopy fogging but these have been almost always undescribed species. Some of these species may actually be breeding in epiphytes rather than the fogged tree. Calliaspis rubra (Olivier) and Acentroptera pulchella Guérin-Méneville have been associated with bromeliads (Bromeliaceae) in South America (Lowman et al. 1996;Mantovani et al. 2005).

Biogeography
Most of the La Selva hispine fauna is closely related to South American species. Some species have distributions throughout the Neotropics such as Aslamidium impurum (Boheman), Charistena ruficollis (Fabricius), and Imatidium thoracicum Fabricius. However, the genera Anisostena Weise and Glyphuroplata Uhmann are most speciose in the Nearctic and the La Selva specimens are part of the southern extension of the genera (Staines 2002b). No La Selva hispines are exotic.
Many species appear to be Central American Atlantic lowland wet forest endemics but with congeners in South America. Ocnosispa humerosa Staines, Sceloenopla bicolorata Staines, S. bidentata Staines, S. lutena Staines, and S. nigropicta Staines appear to fall into this category.

Specimen identification
Of the 139 hispine species known from La Selva, 125 (89.9%) are described species with published names, one is a morphospecies which is known to be new, and 14 (11.2%) are morphospecies in groups whose taxonomy is too poorly known to determine whether they are new or not.
La Selva hispine species can be identified using the key to the genera in Staines (2002b). All genera and species of La Selva hispines are in the "hispines of La Selva" web site (http://viceroy.eeb.uconn.edu/ALAS/ALAS.html). This site includes a summary of hispines, species lists, keys to species, references to revisions and other taxonomic publications, and individual species accounts with images and natural history data.

Suggestions for future Research
What do hispines eat? A little more than half (63.3%) of La Selva hispines have any host plant association. Many of these have only been noted as being collected on a plant rather than actually feeding on it (listed as adult on Table 1). Since hispines are intimately tied to their host plant, determining the food plant will give a much better picture of their distribution and abundance. Additional leaf-miner rearing work such as that of Hespenheide & Dang (1999) is needed to make larval host plant associations. Johnson (2004a, b), Johnson & Horvitz (2005), García-Robledo and Horvitz. (2009, in press), andGarcía-Robledo et al. (2010) worked on the biology and ecology of several Cephaloleia species at La Selva.
What is the biology and ecology of hispines? Very little work has been done on the biology and ecology of La Selva hispines. Kirkendall (1984) studied the mating behavior of the North American Odontota dorsalis (Thunberg). Eberhard (1994) mentioned a hispine in his study of insect and spider courtship behavior. Staines and Staines (2001) and Flowers and Hanson (2003) suggested chrysomelids as potential indicator species assemblages for natural area monitoring. Farrell and Erwin (1988) showed that chrysomelids are a good indicator of local species richness. None of these ideas have been applied to hispines at La Selva.
What are the hispine host plant interactions? Strauss (1988) demonstrated that chrysomelids are a useful group for studing these interactions. Some work by Strong and his students (Strong 1977a, 1977b, 1981, 1982a, 1982b, Strong and Wang 1977, Horvitz and Schemske (2002), García-Robledo andHorvitz. (2009, in press), andGarcía-Robledo et al. (2010) have added to our knowledge of this but much remains to be done.
How do pathogens, predators, and parasitoids influence hispine populations? Hispines are parasitized by various wasps and flies (Cox 1994) and mites (Santiago-Blay and Fain 1994). They also have a few recorded predators (Cox 1996) and pathogens (Balazuc 1988, Hazarika andPuzari 1990). , , and Lewis et al. (2002) developed food and parasitism webs for some hispine species. A great deal of work needs to be done on how these organisms interact and what effect they have on hispine populations and distribution.
How do hispine populations and distributions change over time? Staines (2004b) studied the changes in chrysomelid populations over time on Plummers Island, Maryland. With the baseline inventory data and local knowledge at La Selva, a similar project could be started.