The park covers an area of 20,024 ha of well-preserved Atlantic Rain Forest (see Veloso et al. 1991 for more on local vegetation) and is located ca. 100 km from Rio de Janeiro, in a mountainous area ranging from 80 m to 2263 m elevation. The climate is tropical, with a colder drier season from May to August, and a rainy warmer period from November to February (Flinte et al. 2009b). Mean annual temperature is around 18 °C, maximum of 38 °C and minimum of 0 °C. Annual precipitation varies between 1250 and 1500 mm (Flinte et al. 2008).

Species were sampled by a combination of manual collecting, sweep nets and malaise traps, during the conduction of other projects with Chrysomelidae in the park. When a species was initially found in the field, individuals in as many different developmental stages as possible were brought to the laboratory and reared in plastic containers for host plant confirmation and observations on behavior and biology. In an attempt to describe species seasonal distribution in the area, considering data on labels of specimens from all collections, we recorded the different months on which they were collected and summed the number of species per month (independent of year).

Chrysomelinae species were identified by Mauro Daccordi. Solanaceae host plants were identified by Lucia d’Ávila Freire de Carvalho (Jardim Botânico do Rio de Janeiro) and Luciano Bianchetti (Embrapa/Brasília), Asteraceae by Roberto Lourenço Esteves (Universidade do Estado do Rio de Janeiro), Convolvulaceae by Rosângela Simão-Bianchini (Herbário SP - Instituto de Botânica) and Malvaceae by Massimo Bovini (Jardim Botânico do Rio de Janeiro). Thiago Marinho Alvarenga (Universidade de Campinas) identified parasitoids. Species collected at Serra dos Órgãos National Park are deposited at CLEI-UFRJ, Rio de Janeiro, Brazil.

The Chrysomelinae Neotropical fauna is thought to comprise ca. 38 genera (Daccordi 1996) and 1,020 species (Blackwelder 1944), but these are outdated numbers and no such information could be found specifically for Brazil. One hundred species occurring in the State of Rio de Janeiro were recorded, belonging to 21 genera in one tribe (Chrysomelini) and three subtribes: Chrysolinina, Chrysomelina and Entomoscelina (Table 1). Chrysolinina was represented by 91 species, followed by Chrysomelina with eight and Entomoscelina with only one species (Table 1). According to Daccordi (1996), there are many endemic Chrysolinina and Chrysomelina taxa in the Neotropical region, where they reach their maximum diversity. Only nine species are from previously published sources, the other 91 species we found are new records for the state. The genus with most species records was Platyphora (n = 39) representing 42.4% of Chrysolinina found and 40% of total species records, followed by Stilodes (n = 13, 12.9% of all species recorded) and Deuterocampta (n = 11, 10.9%), genera restricted to the Neotropical region (Daccordi 1996). Indeed, Platyphora is the most species-rich genus in South America (Daccordi 1994), with approximately 500 species (Chaboo et al. 2014 and references therein).

The findings presented here also revealed a high diversity of species and genera, typical for the Atlantic rain forest, in comparison to other studies in South and Central America. Flowers (2004) documented 67 species in 11 genera for Costa Rica and, similar to our work, Platyphora and Stilodes were the most species-rich genera. During a six-year field study in a Mexican state, 47 species and eight genera were found; Leptinotarsa, Calligrapha and Zygogramma were the genera with most species records (Burgos-Solorio and Anaya-Rosales 2004). Chaboo and Flowers (2015) found 158 species and 18 genera for Peru, based on species catalogues.

Species were recorded from only 17 (18.5%) of the state’s municipalities, and 62 species were known from only one location (Table 1). We found a similar pattern in an inventory of Cassidinae for the same state (Flinte et al. 2009a), with most records concentrated near the city of Rio de Janeiro and in large protected areas, such as Petrópolis and Teresópolis (Serra dos Órgãos National Park) and Itatiaia (Itatiaia National Park). The high number of single locality records is probably due more to sampling effort than to endemism, considering that species normally are not very abundant and are more easily collected manually than with traps.

A total of 43 species were recorded from Serra dos Órgaõs National Park (Table 1, under SONP; Figure 1), all Chrysomelini, 42 occurring within the subtribe Chrysolinina and only one from Chrysomelina (Pixis columbina). Within Chrysolinina, Platyphora was the genus with most species records (23 species) out of the 10 genera found, followed by Stilodes (7) and Deuterocampta (3), much like the pattern found generally over the state (Table 1). Species showed an enormous variation in color. Adult polymorphism expressed by variation in pronotum color was observed in P. fervida, (Fig. 1–L1, L2), while the degree of fusion in stripes on the elytra varied greatly among individuals in Zygogramma appendiculata (Fig. 1–Y1). Other species, such as Platyphora axillaris (Fig. 1–E1), P. dejeani (Fig. 1–I1) and P. fraterna (Fig. 1–N1), displayed strikingly similar coloration to the leaves of their host plant, while other species including P. congener (Fig. 1G), Calligrapha polyspila (Fig. 1A) and Elytrosphaera xanthopyga (Fig. 1C) were highly conspicuous to the human eye.

Figure 1. 

Chrysomelinae species in Rio de Janeiro. Some Chrysomelinae species occurring in Serra dos Órgãos National Park, State of Rio de Janeiro, Brazil. Calligrapha polyspila (A); Deuterocampta sedula adult (B1) and larva (B2); Elytrosphaera xanthopyga (C); Grammodesma rubroaenea adult (D1) and larva (D2); Platyphora axillaris adult (E1) and larva (E2); Platyphora bigata (F); Platyphora congener (G); Platyphora curticollis adult (H1), larval cannibalism (H2) and larval aggregation (H3); Platyphora dejeani adult (I1) and larva (I2) Calligrapha polyspila (Platyphora difficilis) (J); Platyphora fasciatomaculata adult (K1) and larva (K2); Platyphora fervida yellow-pronotum adult and larva (L1) and red-pronotum female ovipositing (L2); Platyphora flavovittata (M); Platyphora fraterna adult (N1) and larval aggregation (N2); Platyphora itatiayensis adult (O1) and larvae (O2); Platyphora jucunda adult (P1) and larval aggregation (P2); Platyphora langsdorfi adult (Q1) and larva (Q2); Platyphora pastica (R) Platyphora vidanoi (S); Platyphora zikani adult (T1) and young larvae (T2); Platyphora zonata (U); Stilodes thetis (V); Stilodes (Grammomades) impuncticollis adult (W1), eggs (W2) and larva (W3); Trichomela xantholoma (X); Zygogramma appendiculata polymorphic adults in copula (Y1), larvae feeding (Y2), larval cycloalexy (Y3), adult aggregation (Y4), egg mass (Y5), larva attacked by hemipteran nymph (Y6).

The subfamily in SONP exhibited a clear seasonal pattern (Fig. 2), with more species recorded in the hot rainy season, from October to January, than during the drier and colder months, between June and August. This seasonal pattern is well-established for the family Chrysomelidae in the area, with annual variation in temperature and precipitation and effects on host plant phenology being likely the main drivers of the temporal dynamics in these beetles (Flinte et al. 2009b, 2011, 2015). This is particularly so because many of the records were made at altitudes above 1000 m, where the pattern normally more closely resembles that found in the subtropical zone (e.g. Medeiros and Vasconcellos-Neto 1994, Nogueira-de-Sá et al. 2004). However, the present results are, to our knowledge, the first to examine the seasonal pattern for such a large number of Chrysomelinae species in a single area. Ideally, a standardized collecting effort across the year would better describe the seasonal differences we observe here. However, as we have conducted research in the area over many years, doing the same surveys at least once a month every year, we are confident that this represents the seasonal pattern of chrysomeline species occurrence in the area. Moreover, the Chrysomelinae species which were intensively studied over the year, Platyphora axillaris (Flinte et al. 2015), P. fervida and Stilodes (Grammomades) impuncticollis (Flinte et al. 2009b) exhibited the same low densities during the drier and colder months.

Figure 2. 

Seasonal distribution of Chrysomelinae. Number of Chrysomelinae species recorded on each month, obtained for 40 species from collections and fieldwork, in Serra dos Órgãos National Park, southeast Brazil.

Host plant and/or biological information were found for almost half of the species (n = 16) (Table 2) that we (VF, AA, MVM, RFM) collected in the park (n = 35). Solanaceae was the most common host plant family, followed by Convolvulaceae, Asteraceae, Malvaceae and Apocynaceae. As expected, this is a pattern that reflects Platyphora preference for Solanaceae (Jolivet and Hawkeswood 1995, Chaboo et al. 2014 and references therein). Intense host plant defoliation was observed in Stilodes (Grammomades) impuncticollis (Fig. 1–W1, W2, W3) on Capsicum mirabile (Solanaceae), Platyphora fraterna (Fig. 1–N1, N2) on Solanum swartzianum (Solanaceae) and Zygogramma appendiculata (Fig. 1–Y1, Y2, Y3) on Callianthe rufinerva (Malvaceae).

Maternal care was not recorded for any species in this study, although subsocial behavior is known in the subfamily for several species, including some Doryphora and Prosicela species (Windsor et al. 2013, Chaboo et al. 2014). Other interesting behavioral defenses were recorded, including larval cycloalexy in Z. appendiculata (Fig. 1–Y3) and P. curticollis (Fig. 1–H3), a defensive behavior of gregarious circular formation at rest (Jolivet et al. 1990, Vanconcellos-Neto and Jolivet 1994, Dury et al. 2014). Additionally, we recorded larvae of P. fraterna (Fig. 1–N2) attaching trichomes from Solanum host plant leaves to hairs on their backs, a behavior already described in P. zonata (Bernardi and Scivittaro 1991), which may contribute to larval camouflage. Larval aggregations were observed in many species (Table 2), but also for adults of Z. appendiculata on young folded leaves in the field (Fig. 1–Y4). Larval gregarious behavior may serve to reduce individual risk against small invertebrate parasitoids and predators, and promote defense against larger predators through the cumulative effect of individuals’ toxins (Grégoire 1988). Thanatosis (“feigning death”) was observed in adults of P. axillaris (Fig. 1–E1), P. fervida (Fig. 1–L1, L2), and P. fraterna (Fig. 1–N1), and both in adults and larvae of Stilodes (Grammomades) impuncticollis (Fig. 1–W1, W3).

Seven oviparous and seven larviparous species were found, most being new records of reproductive biology (Table 2). Chrysomelinae is the subfamily of leaf beetles with the most diversity in reproductive biology, containing oviparous, ovoviviparous and viviparous species (Bontems 1988), sometimes in the same genus, and also different levels of social behavior (Chaboo et al. 2014). The last two types of development may be more costly to the mothers, but ensure a quicker development of the vulnerable larval stage, among other advantages, as proposed by Jolivet and Hawkeswood (1995) and Chaboo et al. (2014 and references therein), which is why it is sometimes considered to be a parental care preceding birth (Hinton 1981). Interestingly, viviparous species may result in solitary larvae, as in P. axillaris (Fig. 1–E2), or larval aggregations, as in P. jucunda (Fig. 1–P2). Oviparous species may also have solitary or gregarious larvae, as in P. dejeani (Fig. 1–I2) and Z. appendiculata (Fig. 1–Y2), respectively, although larval aggregations seem rarer in this type of reproduction. In their work on subsocial neotropical Doryphorini, Windsor et al. (2013) found, among Platyphora species, two with solitary larvae and nine which formed larval aggregations, but all eleven species were larviparous. We observed a single case of larval cannibalism in the viviparous P. curticollis during laboratory rearing (Fig. 1–H2), a behavior already described for some Chrysomelinae genera (Wade 1994, Mafra-Neto and Jolivet 1996, Windsor et al. 2013) that grants nutritional benefits.

Except for the eggs of Z. appendiculata, which are laid in masses on the underside of its host plant leaves (Fig. 1–Y5), no other eggs of oviparous Chrysomelinae were found in the field. This is probably because chrysomelids often lay their eggs in the soil or in secluded parts of plants (Selman 1994). All oviparous species reared in laboratory laid chorion-covered yellowish eggs on the bottom of the vials or on leaves, normally grouped in clutches (Fig. 1–W2). In the field, the number of eggs of Z. appendiculata varied from 80 to 100 per group (90.4 ± 8.3 SD; n=7 egg masses), and larval aggregations comprised between 10 and 233 individuals per group (49.2 ± 38.7; n=50 groups). Larvae of different egg masses may cohabit the same aggregation of this species, since differently sized larvae were observed in the same aggregation. Platyphora fraterna larvae (Fig. 1–N2) were grouped in aggregations of 24.9 ± 13.1 SD individuals (n=14 groups), with a minimum of seven and maximum of 44 larvae per group. No pupa has yet been found in the field, but in the laboratory, prepupae always buried themselves in earth layer at the bottom of the vials. While pupation in Chrysomelinae may be arboreal or underground (Takizawa 1976), it seems that underground pupation is most common in our taxa as indicated by laboratory rearing.

Only a few observations on natural enemies of Chrysomelinae were made. Phoretic wasps of Pteromalidae (Hymenoptera) were found on adults of Grammodesma rubroaenea (Fig. 1–D1) and Deuterocampta sedula (Fig. 1–B1). Pteromalidae are well known parasitoids of chrysomeline larvae (Cox 1994). On one occasion, a Podisus (Hemiptera) nymph was seen preying on a larva of Z. appendiculata (Fig. 1–Y6). Many chrysomeline species presented unprotected larvae without any apparent behavioral defense, but several gain chemical defenses by the sequestration of host plants toxins or by synthesizing defensive compounds from plant precursors, especially in Platyphora (Pasteels et al. 2001, Termonia et al. 2002).