Relationships between host plant architecture and gall abundance and survival

Correlação entre a arquitetura da planta hospedeira e a abundância e sobrevivência de um inseto galhador

BIOLOGY, ECOLOGY AND DIVERSITY

Relationships between host plant architecture and gall abundance and survival

Correlação entre a arquitetura da planta hospedeira e a abundância e sobrevivência de um inseto galhador

Daniela P. Lara^I; Lázaro A. Oliveira^I; Islaine F. P. Azevedo^I; Márcia F. Xavier^I; Fernando A. O. Silveira^{I, III}; Marco Antonio Alves Carneiro^{II, III}; G. Wilson Fernandes^III

^IDepto. de Ciências Biológicas, Faculdade de Filosofia, Ciências e Letras do Alto São Francisco - Av. Laerton Paulinelli, 153, 35.595.000 Monsenhor Parreiras, Luz-MG, Brazil

^IIInstituto de Ciências Exatas e Biológicas, Universidade Federal de Ouro Preto, Campus Morro do Cruzeiro, Ouro Preto-MG, Brazil

^IIIEcologia Evolutiva & Biodiversidade/DBG, ICB/Universidade Federal de Minas Gerais; Caixa Postal 486, 30161 970 Belo Horizonte-MG, Brazil. [e-mail]

ABSTRACT

The plant architecture hypothesis predicts that variation in host plant architecture influences insect herbivore community structure, dynamics and performance. In this study we evaluated the effects of Macairea radula (Melastomataceae) architecture on the abundance of galls induced by a moth (Lepidoptera: Gelechiidae). Plant architecture and gall abundance were directly recorded on 58 arbitrarily chosen M. radula host plants in the rainy season of 2006 in an area of Cerrado vegetation, southeastern Brazil. Plant height, dry biomass, number of branches, number of shoots and leaf abundance were used as predicting variables of gall abundance and larval survival. Gall abundance correlated positively with host plant biomass and branch number. Otherwise, no correlation (p > 0.05) was found between gall abundance with shoot number or with the number of leaves/plant. From a total of 124 galls analyzed, 67.7% survived, 14.5% were attacked by parasitoids, while 17.7% died due to unknown causes. Larvae that survived or were parasitized were not influenced by architectural complexity of the host plant. Our results partially corroborate the plant architecture hypothesis, but since parasitism was not related to plant architecture it is argued that bottom-up effects may be more important than top-down effects in controlling the population dynamics of the galling lepidopteran. Because galling insects often decrease plant fitness, the potential of galling insects in selecting for less architectural complex plants is discussed.

Keywords: Cerrado; insect galls; plant size; structural complexity.

RESUMO

A hipótese da arquitetura de planta prediz que variações na arquitetura da planta hospedeira influenciam a comunidade de insetos herbívoros, sua dinâmica e performance. Neste estudo o efeito da arquitetura de Macairea radula (Melastomataceae) na abundância de galhas induzidas por uma mariposa (Lepidoptera: Gelechiidae) foi avaliado. A arquitetura da planta e a abundância de galhas foram diretamente registradas em 58 indivíduos de M. radula arbitrariamente selecionados durante a estação chuvosa de 2006 em uma área de Cerrado no sudeste brasileiro. A altura da planta, biomassa seca, número de ramos e abundância de galhas foram utilizados como variáveis para prever a abundância de galhas e sobrevivência das larvas. A abundância das galhas foi positivamente correlacionada com a biomasssa seca e o número de ramos secundários. Por outro lado, nenhuma correlação (p > 0.05) foi encontrada entre a abundância de galhas e ramos terciários e número de folhas por planta. De um total de 124 galhas analisadas, 67,7% sobreviveram, 14,5% foram parasitadas e 17,7% morreram por causas indeterminadas. No entanto, o parasitismo e sobrevivência das larvas não foram influenciados pela complexidade arquitetônica da planta hospedeira. Os resultados corroboram parcialmente a hipótese de arquitetura da planta, mas como o parasitismo não foi relacionado com a arquitetura da planta, os efeitos reguladores de níveis inferiores da cadeia alimentar possivelmente são mais importantes do que os efeitos de níveis superiores da cadeia alimentar na dinâmica populacional do galhador.

Palavras-chave: Cerrado; complexidade estrutural; galhas de insetos; tamanho de planta.

The preference for host plant species, individuals within a species, or traits of host plants, can influence the performance, distribution and abundance of insect herbivores and many trends have been reported worldwide. Not surprisingly, several hypotheses have been proposed to understand the described patterns of attack by herbivorous insects at several ecological scales. The "plant architecture hypothesis" states that the physical structure of the aerial parts of the host plant influences the community structure of herbivorous insects (Lawton 1983; Strong et al. 1984; Araújo et al. 2006). The plant architecture encompasses plant size, growth form, variety of above-ground parts and persistence of these parts (Lawton 1983). A pattern of increased insect herbivore species richness and/or abundance with architecture host plant has been supported by several studies (e.g., Lawton 1983; Strong et al. 1984; Alonso & Herrera 1996; Haysom & Coulson 1998; Marquis et al. 2002; Rudgers & Whitney 2006; Woodcock et al. 2007). The mechanisms driving the reported pattern are argued to be i) the increase in size per se; ii) increase in resource diversity that provide herbivores with greater amount and diversity of resources offered by high architectural complex plants; iii) lower herbivore mortality because larger plants provide enemy-free space (Jeffries & Lawton 1984); and/or reduced searching efficiency of parasitoids (Gingras et al. 2002; Riihimäki et al. 2006).

Based on a broad scale study, Espírito-Santo et al. (2007) stressed that plant architecture is usually a relevant factor influencing herbivore distribution and abundance, but that the determinant plant traits may vary between studies depending on the life history of the species involved. Indeed, the responses of gall-inducing insects to plant architecture is highly variable. As a result of their sessile habit, gall-inducing insects rely on the availability of young, undifferentiated tissue for gall induction and development (Mani 1964; Weis et al. 1988; Rohfritsch 1992). Therefore, plant architectural traits should have stronger effects on this guild when compared to free-feeding herbivores, which are allowed to leave the plant. Galling sawfly species richness does not increase with plant height, and are richer on shrubs than on trees, independently on the geographic range of the host plant (Price et al. 1998). Moreover, galling species richness is higher on woody than on herbaceous plants, but do not differ between shrubs and trees (Fernandes & Price 1988; Gonçalves-Alvim & Fernandes 2001; Lara et al. 2002). At the species level, Araújo et al. (2006) found that plant architecture positively influenced richness, abundance and survival of galls induced on the shrub Baccharis pseudomyriocephala Teodoro, while a similar result was obtained for several other Baccharis species (Espírito-Santo et al. 2007). Since no conclusions can yet be drawn at this moment on whether galling insects positively respond to increase in plant architecture complexity, new studies are required to verify whether this hypothesis holds for different phylogenetically unrelated species. Therefore, the goal of this study was to test the hypothesis that gall abundance and larval survival increase with plant size and architectural complexity of the host Macairea radula.

METHODS

This study was conducted in the rainy season of 2006 in a 20-ha area of Cerrado vegetation in Dores do Indaiá (19º 27' 16.4'' S; 45º 36' 09.5'' W), Minas Gerais, southeastern Brazil at 716m a.s.l. The vegetation is characterized by scattered and sclerophyllous shrubs occurring among a well-developed herbaceous layer dominated by grasses and sedges. The climate is seasonal with rainy summers and dry winters. Mean annual temperature is 22ºC and precipitation is 1.300mm.

Macairea radula DC. (Melastomataceae) is an evergreen shrub commonly found along the borders of gallery forests, but also occurs away from watercourses in the vegetation of Cerrado in southeastern Brazil (Gonçalves-Alvim et al. 1999). This species bears a conspicuous gall induced by an unidentified microlepidopteran (Lepidoptera, Gelechiidae) on the leaf axillary buds. Galls are spherical and covered by long green trichomes that become orange-reddish as gall matures. These authors also observed that galls are more abundant on hosts in xeric habitats when compared to mesic ones. Moreover, number of leaves, inflorescence length, dry shoot biomass and length were significant smaller on galled than on ungalled shoots, suggesting that galls decrease host fitness (Gonçalves-Alvim et al. 1999). Galls are large and conspicuous facilitating their counting and evaluation of their survival on hosts differing in architectural complexity.

Plant architecture and gall abundance were directly recorded on 58 arbitrarily chosen M. radula host plants in the field. The host plants were cut at soil level, placed in labeled plastic bags and then taken to the laboratory for measurements. Second-order branches were branches emerging from the trunk while shoots represented branches emerging from second-order branches. The total number of expanding leaves was recorded on all individuals. Number of second-order branches, shoots and leaves were used to evaluate M. radula architecture complexity, while plant biomass (weight of above-ground plant parts) and plant height indicated plant size (Araújo et al. 2006). After the removal of all lepidopteran galls, plants were dried in an oven at 70ºC for seven consecutive days until they reached constant weight.

All galls were dissected in order to determine larvae survival. Mortality factors were classified into predation, parasitism, pathogens, and undetermined factors, following Fernandes & Price (1992).

All analyses were carried out using the procedure 'glm' in R_2.4.0 (R Development Core Team 2006). To test the relationship of gall abundance (y-variables) with architectural complexity and plant size (x-variables), data were fitted into a multiple regression analysis with Poisson error corrected for over-dispersion. The relationship between gall-inducing larvae survival and plant architecture were tested through regression analysis with binomial error corrected for over dispersion. In order to adjust for overdispersion a F test was used instead of a chi square test (Tjur 1998, Crawley 2002, Faraway 2006). The response variable was the proportion of surviving larvae and total number of galls, and the explanatory variables were the architectural complexity and plant biomass (x variables). A regression model with binomial error was used to test for the response of the proportion of parasitized / total galls per plant. Complete models were fitted and then simplified by removing non-significant variables in turn, and verifying the effect of removal on the deviance (Crawley 2002, Araújo et al. 2006). The analyses were followed by residual inspection to test for the suitability of the models and error distribution.

RESULTS

A significant relationship between architectural complexity of the host plant and gall abundance was observed (Fig. 1). Gall abundance was positively correlated with plant biomass (F = 8.897; P = 0.004) and with the number of branches (F = 5.025; P = 0.03). However, no correlation was found between number of galls with shoot number (F = 3.154; P = 0.08) or with the number of leaves of the host plant (F = 2.359; P = 0.13).

From a total of 124 analyzed galls, 67.7% survived, 14.5% were attacked by parasitoids, while the remaining 17.5% died due to unknown factors. The number of surviving larvae did not vary statistically with architectural complexity of the host plant. The number of surviving larvae was not influenced by plant biomass (F = 1.248; P = 0.27), number of leaves (F = 2.854; P = 0.10), number of branches (F = 0.0180; P = 0.90), or by shoot number (F = 0.78; P = 0.38). Lack of correlation between parasitized larvae and plant biomass (F = 0.297; P = 0.59), number of leaves (F = 1.190; P = 0.28), number of branches (F = 0.135; P = 0.71), and shoot number (F = 0.006; P = 0.94) was also observed.

DISCUSSION

Gall abundance increased with increasing plant structural complexity and size. Plants with the lowest structural complexity (number of branches) and lowest size (plant biomass) supported lower gall abundance. Increasing the size of a host plant directly influence the size per se and resource diversity (Lawton 1983). As galling insects depend on young, undifferentiated plant tissue for gall induction and development, higher gall abundance on high architectural complex plants should be expected. In fact, several studies have found an increase in the richness and abundance of galling insects with increasing host plant biomass (Cornelissen & Fernandes 2001; Araújo et al. 2006; Rudgers & Whitney 2006). By assuming that galling insects rely on meristem availability, the estimated number of active meristems has been used as a strong predictive variable of gall abundance (Larson & Whitham 1997; Espírito-Santo et al. 2007).

Our data partially corroborate the "plant architecture hypothesis". However, the mechanisms underlying the pattern are still uncertain. Host plant size per se should account for higher insect abundance on larger host plants. Insect density should increase with increase in host plant biomass per unit area, because of the relative rates of encounters between insects and plants (Haysom & Coulson 1998). Thus, taller M. radula individuals should be more apparent (sensu Feeny 1976) for gravid females and hence support higher gall abundance.

High meristem availability would explain higher colonization in more architectural complex plants. The increase in gall abundance with plant architecture complexity could be the result of an increase in the availability of oviposition sites in more complex plants. Furthermore, plants with higher architectural complexity should provide herbivores with an enemy-free space (Heisswolf et al. 2005; Riihimäki et al. 2006). Otherwise, our results did not support this assumption because the Gelechiidae larvae survival and mortality were not correlated with plant architecture. This result suggests that an increase in architectural complexity and plant size did not offer niches for escape from parasitoids or predators. Araújo et al. (2006) found similar results and proposed that bottom-up mechanisms are more important than top-down in determining the distribution and abundance of galling insects. Host plant architecture differentially determines the predation and parasitism of Epirrita autumnata (Borkhausen, 1794) (Lepidoptera, Geometridae) on birches (Riihimäki et al. 2006). An unidentified passerine was observed predating galls on M. radula (L.A. Oliveira, pers. obs.) but the effect of plant architecture on larval predation remains unclear, as predated galls were not collected.

In a broader scale our study has an important implication. If galling insects negatively impact host fitness (e.g., Gonçalves-Alvim et al. 1999; Muniappan & McFadyen 2005), and gall abundance is higher on more complex plants, it should be expected that natural selection to favor low architectural complexity. Models that predict the evolution of plant architecture (Sussex & Kerk 2001) usually neglect the role of herbivores in plant architecture. Marquis et al. (2002) observed that leaftying caterpillar abundance increased on Quercus alba L. individuals with increased number of touching leaves. They suggest that architectural traits that minimize leaf-to-leaf contact may be defensive against leaftying caterpillar. Thus, further investigations should be conducted to verify whether herbivores represent an important selective pressure on plant architecture (Whitham & Mopper 1985).

Acknowledgements. We thank G.E.V. Padilha for encouraging the study and two anonymous reviewers for their comments and criticisms. CAPES/PICDT provided a Ph.D. scholarship to MAAC. This study was supported by the CNPq (304851/2004-3, 472811/2006-1) and Fapemig.

Received 13/07/2007; accepted 05/11/2007

• Alonson, C. & C. M. Herrera. 1996. Variation in herbivory within and among plants of Daphne Laureola (Thymelaeaceae): correlation with plant size and architecture. The Journal of Ecology 84: 495502.
• Araújo A. P. A.; J. D. Paula; M. A. A. Carneiro & J. H. Schoereder. 2006. Effects of host plant architecture on colonization by galling insects. Austral Ecology 31: 343348.
• Cornelissen T. G. & G. W. Fernandes. 2001. Patterns of attack by herbivores on the tropical shrub Bauhinia brevipes (Leguminosae): vigour or chance? European Journal of Entomology 98: 3740.
• Crawley, M. 2002. Statistical Computing: an introduction to data analysis using S-plus Baffins Lane, John Wiley & Sons, ix + 761 p.
• Espírito-Santo, M. M.; F. S. Neves; F. R. Andrade-Neto & G. W. Fernandes, 2007. Plant architecture and merystem dynamics as the mechanisms determining the diversity of gall-inducing insects. Oecologia 153: 353364.
• Faraway, J. J. 2006. Extending Linear Models with R: Generalized Linear, Mixed Effects and Nonparametric Regression Models Boca Raton, Chapman & Hall/CRC, x + 229 p.
• Feeny, P. 1976. Plant apparency and chemical defense. Recent Advances in Phytochemistry 10: 140.
• Fernandes, G. W. & P. W. Price. 1988. Biogeographical gradients in galling species richness: tests of hypotheses. Oecologia 76: 161167.
• Fernandes, G. W. & P. W. Price. 1992. The adaptative significance of insect gall distribution: survivorship of species in xeric and mesic habitats. Oecologia 90: 1420.
• Gingras, D.; P. Dutilleul & G. Boivin. 2002. Modeling the impact of plant structure on host-finding behavior of parasitoids. Oecologia 66: 6873.
• Gonçalves-Alvim, S. J. & G. W. Fernandes. 2001. Biodiversity of galling insects: historical, community and habitat effects in four neotropical savannas. Biodiversity and Conservation 10: 7098.
• Gonçalves-Alvim, S. J; E. C. Landau; M. Fagundes; V. G. Silva; Y. R. F. Nunes & G. W. Fernandes. 1999. Abundance and impact of a lepidopteran gall on Macairea sp. (Melastomataceae) in the Neotropics. International Journal of Ecology and Environmental Sciences 25: 115125.
• Hayson, K. A. & J. C. Coulson. 1998. The Lepidoptera fauna associated with Calluna vulgaris: effects of plant architecture on abundance and diversity. Ecological Entomology 23: 377385.
• Heisswolf, A.; E. Obermaier & H. J. Poethke. 2005. Selection of large host plants for oviposition by a monophagous leaf beetle: nutritional quality or enemy-free space? Ecological Entomology 30: 299306.
• Jeffries, M. J & J. H. Lawton. 1984. Enemy-free space and the structure of ecological communities. Biological Journal of the Linnean Society 23: 269286.
• Lara, A. C. F.; G. W. Fernandes & S. J. Gonçalves-Alvim. 2002. Tests of hypotheses on patterns of gall distribution along an altitudinal gradient. Tropical Zoology 15: 219232.
• Larson, K. C. & T. G. Whitham. 1997. Competition between gall aphids and natural plant sinks: plant architecture affects resistance to galling. Oecologia 109: 14321439.
• Lawton, J. H. 1983. Plant architecture and the diversity of phytophagous insects. Annual Review of Entomology 28: 2329.
• Mani, M. S. 1964. Ecology of Plant Galls Junk, The Hague, vii + 343 p.
• Marquis, R. J.; J. T. Lill & A. Piccinni. 2002. Effect of plant architecture on colonization and damage by leaftying caterpillars of Quercus alba Oikos 99: 531537.
• Muniappan, R. & R. E. C. McFadyen. 2005. Gall-inducing arthropods used in the biological control of weeds, p. 709730. In: A. Raman, C. W. Schaefer & T. M. Withers (eds.). Biology, Ecology and Evolution of Gall-inducing Arthropods New Hampshire, Science Publishers Inc, xxi + 817 p.
• Price, P. W.; H. Roininen & A. Zinovjev. 1998. Adaptative radiation of gall-inducing sawflies in relation to archictecture and geographic range of willow host plants, p. 196203. In: G. Csóka, W. J. Mattson, G. N. Stone & P. W. Price (eds.). Biology of gall-inducing arthropods, St. Paul, USDA Tech. Rep. NC-199, v + 329 p.
• R Development Core Team (2006). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org
• Riihimäki, J.; H. Vehvilänen; P. Kaitaniemi & J. Koricheva. 2006. Host tree architecture mediates the effect of predators on herbivore survival. Ecological Entomology 31: 227235.
• Rohfritsch, O. 1992. Patterns in gall development, p. 6086. In: J. D. Shorthouse & O. Rohfritsch (ed.). Biology of insect-induced galls. Oxford, Oxford University Press, xiii + 285 p.
• Rudgers, J. A. & K. D. Whitney. 2006. Interactions between insect herbivores and a plant architectural dimorphism. Journal of Ecology 94: 12491260.
• Strong, D. R.; J. H. Lawton & T. R. E. Southwood. 1984. Insect on plants: community patterns and mechanisms London Backwell, Science, vi + 313 p.
• Sussex, I. M. & N. M. Kerk. 2001. The evolution of plant architecture. Current Opinion in Plant Biology 4: 3337.
• Tjur, T. 1998. Nonlinear regression, quasi likelihood, and overdispersion in generalized linear models. The American Statistician 52: 222227.
• Weis, A. E.; R. Walton & C. L. Creco. 1988. Reactive plant tissue sites and the population biology of gall makers. Annual Review of Entomology 33: 467486.
• Whitham, T. G. & S. Mopper. 1985. Chronic herbivory: impacts on architecture and sex expression of Pinyon pine. Science 228: 10891091.
• Woodcock, B. A.; S. G. Potts; D. B. Westbury; A. J. Ramsay; M. Lambert; S. J. Harris & V. K. Brown. 2007. The importance of sward architectural complexity in structuring predatory and phytophagous invertebrate assemblages. Ecological Entomology 32: 302311.

Publication Dates

• Publication in this collection
  17 Apr 2008
• Date of issue
  2008

History

• Accepted
  05 Nov 2007
• Received
  13 July 2007