Assemblages of endophagous insects on Asteraceae in São Paulo Cerrados

A comunidade de insetos endófagos de Asteraceae em Cerrados de São Paulo

ECOLOGY, BEHAVIOR AND BIONOMICS

Assemblages of endophagous insects on Asteraceae in São Paulo Cerrados

A comunidade de insetos endófagos de Asteraceae em Cerrados de São Paulo

Adriana M. Almeida^{I, II}; Carlos R. Fonseca^III; Paulo I. Prado^{I, IV}; Mário Almeida-Neto^I; Soraia Diniz^V; Umberto Kubota^I; Marina R. Braun^I; Rafael L.G. Raimundo^I; Luciano A. Anjos^I; Tehra G. Mendonça^I; Silvia M. Futada^I; Thomas M. Lewinsohn^I

^ILab. Interações Insetos-Plantas, Depto. Zoologia, IB, Unicamp. C. postal 6109, 13083-970, Campinas, SP [e-mail], [e-mail]

^IINew address: Depto. Botânica, Ecologia e Zoologia, CB, UFRN, Natal, RN

^IIILaboratório de Interação Animal-Planta, Unisinos, 93022-000, São Leopoldo, RS

^IVNew address: Depto. Ecologia Geral, UB, USP, São Paulo, SP

^VNew Address: Depto. Botânica e Ecologia, IB, UFMT, Cuiabá, MT

ABSTRACT

A survey of the endophagous insects fauna associated to Asteraceae capitula was carried out from 2000 to 2002 in eight cerrado sensu stricto sites located in the Brazilian state of São Paulo. Sixty-four endophagous species of Diptera and Lepidoptera were recorded from 49 asteracean host plants. Approximately half of the species were obtained from a single locality, with a large proportion emerging from a single sample (unicates). Thirty percent of the species were singletons (i.e. only one individual was recorded). The large proportion of rare species suggests a high species turnover among localities. Lepidopteran species were recorded on more host species than dipterans, confirming their more polyphagous food habit, also observed in other Brazilian biomes and in Europe. We conclude that the studied cerrado localities, all within São Paulo State, are isolated with its invertebrate fauna composed of many rare and exclusive species. We suggest that the maintenance of Asteraceae biodiversity and their endophagous insects depend on the conservation of all cerrado remnants in the state.

Key words: Insect-plant interaction, biodiversity, Diptera, Lepidoptera

RESUMO

No presente estudo foi realizado um levantamento da fauna de insetos endófagos de capítulos de Asteraceae (Compostas), de 2000 a 2002, em oito localidades de cerrados sensu stricto no estado de São Paulo. Foram obtidas 64 espécies de endófagos (dípteros e lepidópteros) de 49 espécies de hospedeiras. Aproximadamente metade das espécies foi obtida de apenas uma área, e destas uma grande proporção ocorreu em apenas uma amostra (unicatas). Trinta por cento das espécies foram consideradas "singletons" (apenas um indivíduo foi obtido). A grande quantidade de espécies raras sugere uma forte troca de espécies entre diferentes áreas. Lepidópteros foram registrados em mais espécies hospedeiras que dípteros, confirmando o já observado para o mesmo sistema de insetos e plantas em outros ambientes no Brasil e Europa. As áreas de cerrado sensu stricto estudadas no estado de São Paulo encontram-se isoladas, com uma grande parte da fauna de invertebrados composta por muitas espécies raras e exclusivas. Diante deste quadro, sugere-se que a manutenção da biodiversidade de Asteraceae e seus endófagos em seus níveis atuais depende da conservação de todo o conjunto de remanescentes de cerrado do estado.

Palavras-chave: Interação inseto-planta, biodiversidade, Diptera, Lepidoptera

About 23% of the Brazilian territory (2 million km²) was originally covered by cerrado vegetation (Ratter et al. 1997). Its core area, situated on the Brazilian Central Plateau, is the most preserved and continuous (Ratter et al. 1997). In the state of São Paulo there are a number of outlying cerrado areas representing the southernmost distribution of the biome. In the early 19^th century an estimated 14% of its territory was covered with cerrados, but nowadays only 7% of its original cover remains, representing less than one percent of its territory (Kronka et al. 1998, Durigan et al. 2003).

Efforts have been made to study and describe cerrado fragments in São Paulo state, mostly inventorying trees, or woody vegetation (e.g., Durigan 2003). Insect ecology in cerrados has not been sufficiently explored and studies tend to be restricted to its core area (e.g., Diniz & Morais 1997, Brown & Gifford 2002, Marquis et al. 2002), despite many recent studies in non-core areas (e.g. Ribeiro et al. 1998, Urso-Guimarães et al. 2003, Fernandes et al. 2005). Little is known about the insect fauna in São Paulo cerrados (but see Cytrynowicz 1991, Silveira & Campos 1995, Silva & Bates 2002). The study of insects in cerrado may bring surprises: information obtained so far shows that alternatively to accepted wisdom for vertebrates, the insect fauna in cerrados may be largely characteristic of the biome (MMA/SBF 2002, but see Camargo & Becker 1998).

The study of insect communities is often based on mass-collection methods such as insecticide fogging and light trapping. These methods provide a large number of individuals and species, but yield little or no information on the ecology of these species. It is uncertain which species are genuine members of the community and which are transient. An alternative approach in community ecology is the study of strongly interacting species, through resource-centered inventories, e.g. sampling live insects from host plants (Lewinsohn et al. 2001, Novotny et al. 2004).

Endophagous insects feed internally on host plants and can be borers, miners, or gall makers, and endophages are especially diverse and frequent in Asteraceae flower heads. Host plants act as microhabitat and food resource to herbivores (Southwood 1973). Surveys of endophagous insects yield more ecological information on the studied species, because once the endophages are sampled (in general as larvae), its trophic link to the host plant is ensured. This protocol provide a novel source of information on the structure of food webs and is very important in obtaining information on important and diversified groups that are not available if other sampling method is used (Lewinsohn et al. 2001, Novotny et al. 2004).

The present study used a resource-centered protocol to study the endophagous insect communities associated with asteracean flower heads in eight cerrado localities in the state of São Paulo. Flower heads are food resources with high structural diversity and are spatially and temporally fairly predictable. On the other hand, being restricted to a discrete structure, endophagous species face other competitor species and are easy and predictable targets for specialized natural enemies (Zwölfer 1983). From March to June for three years we sampled flower heads from 71 host species, belonging to different tribes (Almeida et al. 2005). In total, we found 64 endophagous species belonging to the orders Diptera and Lepidoptera, comprising 252 realized interactions. A high proportion of species was represented by one individual (singleton), and approximately half of the observed species of both hosts and endophages was restricted to a single locality.

In this paper we present a list of the species of flowerhead-feeding insects on cerrado Asteraceae, and an initial analyses on the occurrence and distribution across the studied localities in São Paulo. The following hypotheses were tested: 1. Is the total number of endophagous insect species positively correlated with local richness of its host community? 2. Is the geographical range of a host plant positively correlated with the size of its associated endophagous assemblage?

Material and Methods

Field sampling and insect rearing. We conducted 23 field trips to sample flower heads from eight remnant cerrado localities in São Paulo state. All sampled areas were covered by cerrado sensu stricto, characterized by dense cover of herbs and shrubs, and scattered trees (Oliveira & Marquis 2002). Fieldwork was performed in March to June, 2000 to 2002 (autumn in the Southern hemisphere), the main flowering period of most Asteraceae in cerrados (see Almeida et al. 2005 and Fonseca et al. 2005, for details of host sampling). In general, each remnant was sampled once a year, although Pedregulho, Martinópolis and Pé de Gigante were not sampled in 2001 (Table 1 in Almeida et al. 2005). In Itirapina in the last year, two distinct remnants 13 km apart were sampled. Since we are concerned with larger-scale geographical patterns, these two remnants were grouped as a single locality (Almeida et al. 2005).

For the present purpose, one sample is defined as a group of flower heads sampled from a population of one host species, in one remnant, in each sampling period. Only mature flower heads were sampled to rear endophagous adults. In each remnant, we obtained a variable number of samples, up to the total number of flowering Asteraceae species at the time. Some hosts were so rare that they could only be sampled for voucher specimens, and consequently no samples of these species could be obtained to rear herbivores.

In the laboratory, each sample was kept in transparent plastic containers of 500 ml or 1000 ml covered with a fine mesh lid. Samples were kept in the lab for a minimum of two months and were checked every two days for the presence of adult insects. Emerged insects were sedated with CO₂ to facilitate manipulation.

After adult emergence, each sample (one host plant in one locality) dry weight was measured as an approximate measure of sampling effort.

Species identification. All insects were assigned to morphospecies based on their external morphology. Morphospecies were subsequently verified by specialist taxonomists (see acknowledgements) and identified as far as possible. Our morphospecies correspond to species that have often been matched to named species and are the subject of further taxonomic research. Voucher specimens are deposited in the "Museu de História Natural da Unicamp" (MZUEC, Campinas, Brazil). Approximately half of the obtained species could be assigned to named species (28 species, 44%).

A high proportion of the morphospecies is probably of undescribed species. One example is the family Tephritidae (Diptera) in which several species have been described (e.g., Prado et al. 2004, Abreu et al. 2005) and others await description from our material (Allen Norrbom, personal communication). In every insect family, however, some groups show identification problems: (1) Xanthaciura sp.n.01 is undescribed and at present would be considered a morph of Xanthaciura chrysura S.G. Thomson (Allen Norrbom, pers. comm.); there are however some subtle differences in the puparium and adult external characters that will probably enable this morph to be described as a new species. (2) All agromyzids were aggregated as a single entity of Melanagromyza because species differentiation in this genus is based only on male genitalia, except for the individuals obtained in Mogi Guaçu, which were identified as Melanagromyza minimoides Spencer. (3) The microlepidopterans also have identification problems; the taxonomy of the genera and species groups represented on Asteraceae is based on small and scattered series of specimens, lacking almost any biological and host information. This group is comprised of small and fragile species that usually lose wing characters still in the sample container, hindering even more their identification. (4) The family Cecidomyiidae (Diptera) was excluded from the species list due to the difficulty in identifying species based only on adult characters; reliable identification in most important genera requires series of matched adults and immature stages, which are impossible to ensure with our rearing methods. Cecicomyiidae in flower heads can live freely or in floret-tube galls, and some are known to be predators of other larvae. The herbivorous species may present a high diversity, and some species are known to have a high potential impact on their hosts' populations and have been evaluated as potential biological control agents (e.g., Hinz 1998).

Data analyses. We performed path analyses to evaluate the importance of different variables in determining endophagous species richness per locality and per host species. Path analysis is a statistical analysis technique used primarily to test the fit of data to an a priori causal hypothesis about the . functioning of a system (McIntire 2004). This analysis is a generalized form of multiple regression, where the causal model is tested for consistency with the correlation structure among variables (Lewinsohn 1991, Wilkinson et al. 1996). To analyze local endophagous species richness we tested the effects of Asteraceae richness (potential hosts) and of the total flower head dry weight collected in each locality (a proxy of aggregate sampling effort). An additional analysis was performed to test a causal model for endophage richness on each plant species. Variables tested were the total sample dry weight and the number of localities where that host occurs (as a rough estimate of geographical distribution within the state). In both analyses, flower head dry weight was used as an estimate of sampling effort, but it is also directly affected by flower head availability in the field (Fonseca et al. 2005).

Results and Discussion

Occurrence and distribution along sampled localities. In all we recorded 89 Asteraceae species in the São Paulo cerrado remnants (Almeida et al. 2005). In total, we collected 358 flower head samples from 71 Asteraceae species throughout the three years of study (not all species could be sampled for rearing adult insects, see methods). The year with most samples was 2002 with 151 samples, followed by 136 samples in 2000 and 71 samples in 2001 (Almeida et al. 2005)

We identified 64 endophagous insect species or morphospecies from 229 samples from 49 host plants. No endophage emerged from 129 samples from 22 plant species (32% of species). Itirapina had the highest species richness, with 42 species, whereas Martinópolis presented the lowest richness with 12 species (Table 1). Itirapina, where two remnants were sampled in 2001, was the locality with highest plant richness and total sample volume (Table 1).

Forty endophagous species of Diptera (Tephritidae and Agromyzidae; Table 2) and 24 of Lepidoptera (Blastobasidae, Gelechiidae, Pterophoridae, Pyralidae, Tortricidae, Geometridae and Lycaenidae; Table 3) were reared from flower heads. Among dipterans, only X. chrysura and Melanagromyza spp. occurred in all sampled localities, and three other species occurred in seven localities (Xanthaciura sp.n.01, absent only from Martinópolis, and Neomyopites paulensis Steyskal and Cecidochares connexa Macquart absent from Pedregulho, Table 2). Among lepidopterans, Adaina bipunctata Möschler and Phalonidia squalida Raz. & Becker emerged from samples from all localities, and Lioptilodes parvus Walsingham was absent only in Pé de Gigante and Unadilla erronella Zeller was absent only in Assis (Table 3).

As already noted and described from Asteraceae host plants (Almeida et al. 2005), a high proportion of endophagous insects was restricted to one locality (45% of Diptera and 46% of Lepidoptera species; Fig.1). All dipteran species restricted to one locality emerged from a single sample (unicates), except for M. minimoides, which is recorded only in Mogi Guaçu, because till now was only identified to species from this locality (see Methods). One study in progress (Marina R. Braun, unpublished data) found that M. minimoides presents a broad geographical distribution. Among lepidopterans, species that occurred in one locality were also obtained from a single sample (unicates), with only one exception (Table 3).

Singletons. Thirty percent of the species were singletons, that is, represented by only one individual. We found proportionally more singletons among the Lepidoptera (10 species, 41.6%) than among the Diptera (9 species, 22.5%). Seven species (all Tephritids) were unicates (occurring in one sample) but not singletons (Table 2).

The occurrence of a great proportion of singletons is usually cited for herbivorous insects in tropical regions (Robinson & Tuck 1993, Novotny & Basset 2000), including previous studies in cerrados (Price et al. 1995, Diniz & Morais 1997). Diniz & Morais (1997) studied caterpillars from three host plant species in four cerrado sensu stricto areas in central Brazil. They reared 107 lepidopteran species, of which 46% were singletons. The authors conclude that Lepidoptera species in cerrados have a high incidence of rare species. Price et al. (1995) studied Lepidoptera larvae in three congeneric host plants in a cerrado near Brasilia, and also found a high richness of rare species.

Novotny & Basset (2000) made an exhaustive survey of herbivorous insects associated with 30 species of trees and shrubs in New Guinea and found that 30% of the insect species were represented by singletons. The average proportion of singleton herbivores per plant species was 45%, while an average of 8% were unique singletons (i.e., that did not occur on other hosts). They observed that a species could be a singleton in a particular host whilst being more common on other, often related, host species, or even relatively rare on numerous other host plants, so that its aggregate population was high. They hypothesized that the large number of singletons is due to the mass effect of Shmida & Wilson (1985). According to this effect, singleton species occur in "marginal" host plants, with its populations maintained by a constant influx of immigrants from the "main" host plants.

The tephritids studied here are usually restricted to one or few tribes within Asteraceae (Prado et al. 2002). We have also no information of our endophagous microlepidoptera species, but it is very unlikely that they would be able to use resources from families other than Asteraceae. The present study sampled flower heads from all asteracean flowering species that could be found in the area, so all known potential hosts were represented in our sample and the "mass effect" hypothesis among Asteraceae host plants is less likely to apply.

Host specificity is one of the most relevant traits influencing species rarity, since the patterns of host use determine the abundance and dynamics of resources available to herbivorous species (Novotny & Basset 2000). Nonetheless, Diniz & Morais (1997) found no relation between lepidopteran species frequency and abundance or specialization: the most frequent species can be either restricted to one plant genus or be generalist.

Abundance and specialization.X. chrysura was the most abundant species with 2,585 reared individuals, followed by Xanthaciura sp.n.01 with 1,760 individuals, Tetreuaresta obscuriventris Loew, with 782 individuals and Melanagromyza spp. with 505 individuals (Table 2). The most abundant Lepidoptera species was U. erronella with 125 individuals (Table 3).

More frequent endophagous species (i.e., that occurred in more samples) were also more abundant, both for Diptera (freq = 0.037 inds + 4.962; r² = 0.73; F_1,38 = 102.0; P < 0.001; n = 40) and Lepidoptera (freq = 0.271 inds + 1.956; F_1,22 = 87.1; r² = 0.80; P < 0.001; n = 24). Diniz & Morais (1997) found no such relationship between abundance and frequency in a local survey of cerrado leaf-feeding caterpillars.

Host plant number explained 57% of dipteran log-transformed abundance per species (loginds = 0.15 nhosts + 0.66; F _1,38 = 51.7; r² = 0.57; P < 0.001; Fig. 2). Lepidoptera showed a similar pattern species recorded on more hosts were most abundant (loginds = 0.09 nhosts + 0.46; F _1,22 = 45.8; r² = 0.68; P < 0.001; Fig. 2), although the slope was significantly shallower (analysis of covariance, test for homogeneity of slopes, F_1,60 = 5.58; P = 0.02). Lepidoptera larvae require several flower heads to complete their development, while more than one dipteran individual can develop in a single flower head (Almeida 1997, Almeida et al. unpublished). Moreover, the frequency of Diptera in flower heads is higher than of Lepidoptera (Almeida et al. unpublished). These two observations explain at least in part why more dipteran individuals are obtained than lepidopterans.

Only eight percent (252) of the 3,136 possible interactions between 49 host and 64 endophagous species were recorded. The three most frequently observed interactions were X. chrysura with Chromolaena pedunculosa (Hook & Arn.) K&R (28 records), Xanthaciura sp.n.01 with C. pedunculosa (19), and X. chrysura with Chromolaena squalida (DC.) K&R (15). Among lepidopterans, the most frequent interaction occurred between Lioptilodes parvus (Pterophoridae) and Baccharis dracunculifolia DC. (9 records). The community was dominated by rare interactions that were recorded once, and represented 65.5% of the total (Fig. 3).

The number of samples from which each herbivore species emerged was a good predictor of its number of host plants. Interestingly, among lepidopterans the increase in host plants with sample number is steeper (y = 0.54x + 0.86; F _1,22 = 414.1; r² = 0.95; P < 0.001; n = 24) than among dipterans (y = 0.18x + 1.48; F _1,38 = 138.4; r² = 0.78; P < 0.001; n = 40; Fig. 4); the difference among slopes is significant (covariance test for slope homogeneity, F_1,60 = 82.3; P < 0.001). This means that each new sample is more prone to reveal an additional host for a lepidopteran than for a dipteran species.

As noted above, endophagous dipterans are more specialized than lepidopterans, both in Brazil and in Europe. In the Mantiqueira Range, southeast Brazil, dipterans are associated to fewer host species than lepidopterans (Almeida 2001 Almeida et al. unpublished). An analysis of the main endophagous Tephritidae in Brazil revealed that 80% of species are oligophagous, found in host plants belonging to one genus or to related genera (Prado et al. 2002). In Europe, endophagous dipterans in Cynaroideae flower heads have a maximum host range of two genera, a quite high degree of specialization (Frenzel & Brandl 2000).

High specialization increases the risk of coextinctions in communities. When a host species is removed from its habitat, some associated species may become extinct, and this effect is more evident among specialists (Koh et al. 2004). This is a relevant concern for cerrado areas in São Paulo, which are already quite reduced (in area and number of remnants) and subject to major disturbances.

Factors influencing endophagous richness. We performed a path analysis in order to investigate whether local insect richness is affected by local host plant richness and by the total flowerhead sample size per locality (Fig. 5). The analysis revealed that total sample size had a significant direct effect on determining the number of endophagous species recorded per locality. The total effect of plant richness on insect richness (i.e., their correlation coefficient) is significant; however, each of its components – neither the direct path, nor the indirect one by way of its effect on sample size – is not significant by itself.

A second path analysis evaluated differences in endophage insect richness recorded on each host plant, rather than in each locality as in the preceding analysis. Potential factors included in this analysis were the number of localities, and the total sample size (i.e., flowerhead volume) for each host species. This analysis revealed that the sample size of flower heads was as important as the hosts' geographical spread within São Paulo cerrados in determining the size of their associated faunas (Fig. 6); however, since total sample size was highly correlated with geographical range, range can be identified as a prime factor in setting total endophagous species richness (i.e., insect gamma diversity) on different host plants.

Lewinsohn (1991) also found a strong relation between total sample size and asteracean endophagous species richness in southeastern Brazil. The effect of sample size can be ascribed to at least two main effects. The first one is that bulk sample reflects flower head abundance in the field (Fonseca et al. 2005). A greater resource pool should support more species (MacArthur 1972, Straw & Ludlow 1994). The second is the great number of rare species. Novotny and Basset (2000) found that a sizeable proportion of singletons in herbivore insects still occur even with a huge fieldwork effort. The greater the bulk sampled, the more the chance of finding rare species, increasing species richness. In the cerrado remnants studied, both effects are probably influencing the increase in endophagous richness with increase in total sample size per host.

Half of the endophagous insect species (Diptera or Lepidoptera) as well as their asteracean host plants (Almeida et al. 2005) were restricted to only one of the surveyed cerrado localities. We cannot estimate how many of these current singletons might eventually have been found with further sampling in each site but, given our highly focused and fairly effective sample scheme, there is undoubtedly a majority of scarce and infrequent species. This in turn leads us to infer that each sampled area bears not only unique plant or insect communities, but also unique insect-plant interaction webs (Lewinsohn et al. 2006). We cannot ascertain temporal turnover in species or in interaction links from our present data. However, if reciprocal colonization among cerrado areas has ever contributed to establish and maintain both local and regional diversity (see Thompson 2005), the current reduction of the cerrados in São Paulo to ever decreasing, scattered and isolated remnants is bound to threaten the future of this distinctive component of the neotropical biota.

Acknowledgments

This work was supported by research grant 98/05085 from the BIOTA-FAPESP program of the Fundação de Amparo à Pesquisa do Estado de São Paulo. Various specialists helped with insects' identification: A. Norrbom identified the Tephritidae, except for Tomoplagia identified by P.I. Prado; A.J. Santos identified the Pterophoridae. We are indebted to all project staff that participated the fieldwork and discussions in the laboratory, especially R. Fabiano, E.P. Anseloni, L.B. Mendes, A.E. Fraez, M. Portella, A.M. Rosa; the directors and staff of the Instituto Florestal de São Paulo (IEFSP), for permits and support in their conservation units; and AMBEV, for allowing us to work in the preservation area of their Agudos plant. A.M. Almeida and C.R. Fonseca received postdoctoral fellowships from FAPESP. U. Kubota received a graduate grant from FAPESP. M.R. Braun, T.G. Mendonça, S.M. Futada, R.L.G. Raimundo, L.A. Anjos and M. Almeida-Neto received technical grants from FAPESP.

Received 13/IX/05.

Accepted 16/II/06.

• Abreu, A.G., P.I. Prado, A.L. Norrbom & V.N. Solferini. 2005. Genetic and morphological diagnosis and description of two cryptic species of flower head-infesting Tephritidae (Diptera). Insect Syst. Evol. 36: 361-370.
• Almeida, A.M. 1997. Padrões de co-ocorrência em insetos endófagos associados a capítulos de Trichogoniopsis adenantha (DC) (Asteraceae). Instituto de Biologia, Unicamp, Campinas, 124p.
• Almeida, A.M. 2001. Biogeografia de interações entre Eupatorieae (Asteraceae) e insetos endófagos de capítulos na Serra da Mantiqueira. Tese de doutorado, Instituto de Biologia, Unicamp, Campinas, 161p.
• Almeida, A.M., C.R. Fonseca, P.I. Prado, M. Almeida-Neto, S. Diniz, U. Kubota, M.R. Braun, R.L.G. Raimundo, L.A. Anjos, T.G. Mendonça, S.M. Futada & T.M. Lewinsohn. 2005. Diversidade e ocorrência de Asteraceae em cerrados de São Paulo. Biota Neotropica 5 (2): Available at http://www.biotaneotropica.org.br/v5n2/pt/abstract?article+BN00105022005
• Brown Jr., K.S. & D.R. Gifford. 2002. Lepidoptera in the cerrado landscape and the conservation of vegetation, soil, and topographical mosaics, p.201-217. In P.S. Oliveira & R.J. Marquis (eds.), The cerrados of Brazil: Ecology and natural history of a neotropical savanna. Columbia Univ. Press, New York, 424p.
• Camargo, A.J.A. & V.O. Becker. 1998. Saturniidae (Lepidoptera) from the Brazilian cerrado: composition and biogeographic relationships. Biotropica 31: 696-705.
• Cytrynowicz, M. 1991. Resource size and predictability, and local herbivore richness in a subtropical Brazilian cerrado community, p.561-589. In P.W. Price, T.M. Lewinsohn, G.W. Fernandes & W.W. Benson (eds.), Plant-animal interactions: Evolutionary ecology in tropical and temperate regions. Wiley, New York, 639p.
• Diniz, I.R. & H.C. Morais. 1997. Lepidopteran caterpillar fauna of cerrado host plants. Biodiv. Cons. 6: 817-836.
• Durigan, G., M.F. Siqueira, G.A.D.C. Franco, S. Bridgewater & J.A. Ratter. 2003. The vegetation of priority areas for cerrado conservation in São Paulo State, Brazil. Edinb. J. Bot. 60: 217-241.
• Fernandes, G.W., M. Fagundes, M.K.B. Greco, M.S. Barbeitos & J.C. Santos. 2005. Ants and their effects on an insect herbivore community associated with the inflorescences of Byrsonima crassifolia (Linnaeus) H.B.K. (Malpighiaceae). Revta. Bras. Entomol. 49: 264-269.
• Fonseca, C.R., P.I. Prado, M.Almeida-Neto, U. Kubota & T.M. Lewinsohn. 2005. Flower-heads, herbivores, and their parasitoids: Food web structure along a fertility gradient. Ecol. Entomol. 30: 36-46
• Frenzel M. & R. Brandl. 2000. Phytophagous insect assemblages and the regional species pool: patterns and asymmetries. Global Ecol. Biogeogr. 9: 293-303.
• Hinz, H.L. 1998. Life history and host specificity of Rhopalomyia n. sp. (Diptera: Cecidomyiidae), a potential biological control agent of scentless chamomile. Environ. Entomol. 27: 1537-1547.
• Koh, L.P., R.R. Dunn, N.S. Sodhi, R.K.Colwell, H.C. Proctor & V.S. Smith 2004. Species coextinctions and the biodiversity crisis. Science 305:1632-1634.
• Kronka, F.J.N., M.A. Nalon, C.K. Matsukuma, M. Pavão, J.R. Guillaumon, A.C. Cavalli, E. Gianotti, M.S.S. Iwane, L.M.P.R. Lima, J. Montes, I.H. Del Cali & P.G. Haack. 1998. Áreas de domínio do cerrado no Estado de São Paulo. São Paulo: Secretaria de Estado do Meio Ambiente, Instituto Florestal, 84p.
• Lewinsohn, T.M. 1991. Insects in flower heads of Asteraceae in Southeast Brazil: A case study on tropical species richness, p.525-559. In P.W. Price, T.M. Lewinsohn, G.W. Fernandes & W.W. Benson (eds.), Plant-animal interactions: Evolutionary ecology in tropical and temperate regions. Wiley, New York, 639p.
• Lewinsohn, T.M., P.I. Prado & A.M. Almeida. 2001. Inventários bióticos centrados em recursos: Insetos fitófagos e plantas hospedeiras, p.174-189. In I. Garay & B.F.S. Dias (eds.), Conservação da biodiversidade em ecossistemas tropicais. Ed. Vozes, Petrópolis, 432p.
• Lewinsohn, T.M., P.I. Prado, P. Jordano, J. Bascompte & J.M. Olesen. 2006. Structure in plant-animal assemblages. Oikos 113: 174-184.
• MacArthur, R.H. 1972. Geographical Ecology. Harper & Row, New York, 269p.
• Marquis, R.J., H.C. Morais & I.R. Diniz. 2002. Interactions among cerrado plants and their herbivores: Unique or typical?, p.306-328. In P.S. Oliveira & R.J. Marquis (eds.), The cerrados of Brazil: Ecology and natural history of a neotropical savanna. Columbia Univ. Press, New York, 424p.
• McIntire, E.J.B. 2004. Understanding natural disturbance boundary formation using spatial data and path analysis. Ecology 85: 1933-1943.
• MMA/SBF. 2002. Biodiversidade brasileira: Avaliação de áreas e ações prioritárias para a conservação, utilização sustentável e repartição dos benefícios da biodiversidade nos biomas brasileiros. Brasília, 404p.
• Novotny, V. & Y. Basset. 2000. Rare species in communities of tropical insect herbivores: Pondering the mystery of singletons. Oikos 89: 564-572.
• Novotny, V., Y. Basset, S.E. Miller, R.L. Kitching, M. Laidlaw, P. Drodzd & L. Cizek. 2004. Local species richness of leaf-chewing insects feeding on woody plants from one hectare of a lowland rainforest. Cons. Biol. 18: 227-237.
• Oliveira, P.S. & R.J. Marquis (eds.). 2002. Ecology and natural history of a neotropical savanna: The cerrados of Brazil. Columbia University Press, New York, 424p.
• Prado, P.I., A.L. Norrbom & T.M. Lewinsohn. 2004. New species of Tomoplagia Coquillett (Diptera: Tephritidae) from capitula of Asteraceae in Brazil. Neotrop. Entomol. 33: 189-211.
• Prado, P.I., T.M. Lewinsohn, A.M. Almeida, A.L. Norrbom, B.D. Buys, A.C. Macedo & M.B. Lopes. 2002. The fauna of Tephritidae (Diptera) from capitula of Asteraceae in Brazil. Proc. Entomol. Soc. Wash. 104: 1007-1028.
• Price, P.W., I.R. Diniz, H.C. Morais & E.S.A. Marques. 1995. The abundance of insect herbivore species in the tropics: The high local richness of rare species. Biotropica 27: 468-478.
• Ratter, J.A., J.F. Ribeiro & S. Bridgewater. 1997. The Brazilian Cerrado vegetation and threats to its biodiversity. Ann. Bot. 80: 223-230.
• Ribeiro, S.P., M.A.A. Carneiro & G.W. Fernandes. 1998. Free-feeding insect herbivores along environmental gradients in Serra do Cipó: basis for a management plan. J. Insect Cons. 2: 107118.
• Robinson, G.S. & K.R. Tuck. 1993. Diversity and faunistics of small moths (Microlepidoptera) in Bornean rainforest: Ecol. Entomol. 18: 385393.
• Shmida, A. & M.R. Wilson. 1985. Biological determinants of species diversity. J. Biogeogr. 12: 1-20.
• Silva, J.M.C. & J.M. Bates. 2002. Biogeographic patterns and conservation in the south american cerrado: a tropical savanna hotspot. Bioscience 52: 225-233.
• Silveira, F.A. & M.J.O. Campos. 1995. A melissofauna de Corumbataí (SP) e Paraopeba (MG) e uma análise da biogeografia da abelhas do cerrado brasileiro (Hymenoptera, Apoidea). Revta. Bras. Ent. 39: 371-401.
• Southwood, T.R.E. 1973. The insect/plant relationship an evolutionary perspective, p.139-173. In H. F. van Emden (ed.), Insect/Plant relationships. Cambridge Univ. Press, Cambridge, 215p.
• Straw, N.A. & A.R. Ludlow. 1994. Small-scale dynamics and insect diversity on plants. Oikos 71: 188-192.
• Thompson, J.N. 2005. The geographic mosaic of coevolution. University of Chicago Press, Chicago. 400p.
• Urso-Guimarães, M.V., C. Scarelli-Santos & A.C. Bonifácio-Silva. 2003. Occurrrence and characterization of entomogen galls in plants from natural vegetation areas in Delfinópolis, MG, Brazil. Braz. J. Biol. 63: 705-715
• Wilkinson, L, G. Blank & C. Gruber. 1996. Desktop data analysis with SYSTAT. Prentice-Hall, New Jersey, 798p.
• Zwölfer, H. 1983. Life systems and strategies of resource exploitation in tephritids, p.16-30. In R. Cavalloro (ed.), Fruit flies of economic importance. Proceedings of the CEC/IOBC International Symposium, Athens, Greece, November 16-19, 1982. Rotterdam, A.A. Balkema, 624p.

Publication Dates

• Publication in this collection
  21 Sept 2006
• Date of issue
  Aug 2006

History

• Accepted
  16 Feb 2006
• Received
  13 Sept 2005