Abstract

Spatio-temporal variation in the quantity and quality of resources available to insects influence the strength and direction of interactions with plants, ultimately affecting the preference of insects and also plant performance. Florivory encompasses the consumption of floral resources and has both direct and indirect effects on plant reproduction and performance. In this study, we evaluated how synchronic flowering of phylogenetic-related plant species blooming in an urban landscape affected insect florivores. Flowers from six plant species belonging to the Bignoniaceae family were sampled and florivory was measured as the frequency of attacked flowers (florivory incidence) as well as the amount of floral area removed (florivory intensity). We detected species-specific variation in florivory incidence and intensity in the urban landscape and our spatial analysis indicated that plants that were close (< 200 m apart) and in simultaneous blooming had significantly correlated levels of florivory than plants that were far apart (almost 1 km). Our data indicates complex relationships between insects and flowers and we suggest that a throughout evaluation of insect community and dispersal ability is necessary to understand the patterns of florivory in trees, as well as its effects on plant reproduction and insect attraction.

Keywords:
florivores; insects; insect-plant interactions; flower damage; negative interactions.

Introduction

Interactions between insects and plants comprise processes ranging from positive and mutualistic interactions, such as pollination, to negative and antagonistic interactions, such as herbivory (Strauss et al. 1996Strauss SY, Conner JK, Rush SL. 1996. Foliar herbivory affects floral characters and plant attractiveness to pollinators: Implications for male and female plant fitness. The American Naturalist 147: 1098-1107.; Bronstein et al. 2007Bronstein J, Huxman T, Davidowitz G. 2007. Plant-mediated effects linking herbivory and pollination. In: Ohgushi T, Craig T, Price P (eds.). Ecological Communities: Plant Mediation in Indirect Interaction Webs. Cambridge, Cambridge University Press. p. 75-103.; Roddy et al. 2021Roddy AB, Martínez‐Perez C, Teixido AL et al. 2021. Towards the flower economics spectrum. New Phytologist 229: 665-672.). Variations in plant quantity and quality as well as in the resources offered to insects influence the strength and direction of such interactions (Bronstein et al. 2007Bronstein J, Huxman T, Davidowitz G. 2007. Plant-mediated effects linking herbivory and pollination. In: Ohgushi T, Craig T, Price P (eds.). Ecological Communities: Plant Mediation in Indirect Interaction Webs. Cambridge, Cambridge University Press. p. 75-103.; Albor et al. 2020Albor C, Arceo‐Gómez G, Parra‐Tabla V. 2020. Integrating floral trait and flowering time distribution patterns help reveal a more dynamic nature of co‐flowering community assembly processes. Journal of Ecology 108: 2221-2231.). Pollination and herbivory have been well documented over the last decades (Strauss et al. 1996Strauss SY, Conner JK, Rush SL. 1996. Foliar herbivory affects floral characters and plant attractiveness to pollinators: Implications for male and female plant fitness. The American Naturalist 147: 1098-1107.; McCall & Irwin, 2006McCall AC, Irwin RE. 2006. Florivory: The intersection of pollination and herbivory. Ecology Letters 9: 1351-1365.; Bronstein et al. 2007Bronstein J, Huxman T, Davidowitz G. 2007. Plant-mediated effects linking herbivory and pollination. In: Ohgushi T, Craig T, Price P (eds.). Ecological Communities: Plant Mediation in Indirect Interaction Webs. Cambridge, Cambridge University Press. p. 75-103.; Solga et al. 2014Solga MJ, Harmon JP, Ganguli AC. 2014. Timing is everything: An overview of phenological changes to plants and their pollinators. Natural Areas Journal 34: 227-234.; Mendes et al. 2021Mendes GM, Silveira FA, Oliveira C et al. 2021. How much leaf area do insects eat? A data set of insect herbivory sampled globally with a standardized protocol. Ecology 102: e03301.) but florivory has been far less commonly evaluated in the ecological literature (see Boaventura et al. 2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.). Florivory is defined as any type of damage caused by a floral consumer, occurring at floral buds, bracts, floral parts of the perianth or peritoneum and at male and female reproductive organs (Burgess, 1991Burgess KH. 1991. Florivory: The ecology of flower feeding insects and their host plants. PhD Thesis, Harvard University, Cambridge.; McCall & Irwin, 2006McCall AC, Irwin RE. 2006. Florivory: The intersection of pollination and herbivory. Ecology Letters 9: 1351-1365.; Boaventura et al. 2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.). However, little is yet known about its direct and indirect effects on plants (see McCall & Irwin, 2006McCall AC, Irwin RE. 2006. Florivory: The intersection of pollination and herbivory. Ecology Letters 9: 1351-1365.; Bronstein et al. 2007Bronstein J, Huxman T, Davidowitz G. 2007. Plant-mediated effects linking herbivory and pollination. In: Ohgushi T, Craig T, Price P (eds.). Ecological Communities: Plant Mediation in Indirect Interaction Webs. Cambridge, Cambridge University Press. p. 75-103.; Boaventura et al. 2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.). Even though, recent studies demonstrated that florivory by chewing insects can affect plant fitness through direct effects on flowers via a reduction in reproductive investment and also indirectly by reducing the attractiveness of flowers to pollinators (Moreira et al. 2019Moreira X, Castagneyrol B, Abdala-Roberts L, Traveset A. 2019. A meta-analysis of herbivore effects on plant attractiveness to pollinators. Ecology 100: e02707.; Boaventura et al. 2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.; Ortiz et al. 2023Ortiz GL, Columbano Y, Melo MV et al. 2023. Among-species variation in flower size determines florivory in the largest tropical wetland. American Journal of Botany 110: e16186. ).

Insects that consume plant material can feed upon different plant parts and insect resource preference or choice can vary throughout its life cycle or plant phenological phases (Matter et al. 1999Matter SF, Landry JB, Greco AM, LaCourse CD. 1999. Importance of floral phenology and florivory for Tetraopes tetraophthalmus (Coleoptera: Cerambycidae): Tests at the population and individual level. Environmental Entomology 28: 1044-1051.; Silva & Oliveira, 2010Silva DP, Oliveira PS. 2010. Field biology of Edessa rufomarginata (Hemiptera: Pentatomidae): Phenology, behavior, and patterns of host plant use. Environmental Entomology 39: 1903-1910. ). Resource availability for insects varies due to changes in plant phenology, such as asynchronies that occur in plant phenophases (Thompson & Gilbert, 2014Thompson K, Gilbert F. 2014. Phenological synchrony between a plant and a specialised herbivore. Basic and Applied Ecology 15: 353-361.; Fagundes et al. 2018Fagundes M, Xavier RCF, Faria ML, Lopes LGO, Cuevas‐Reyes P, Reis‐Junior R. 2018. Plant phenological asynchrony and community structure of gall‐inducing insects associated with a tropical tree species. Ecology and Evolution 8: 10687-10697.). Still, the temporality of the supply of resources (such as flowers and leaves) has the potential to structure communities of both herbivores and florivores. Plant phenological variation plays a fundamental role in insect-plant interactions since such variations create temporal and spatial patches of resources of different qualities (Matter et al. 1999Matter SF, Landry JB, Greco AM, LaCourse CD. 1999. Importance of floral phenology and florivory for Tetraopes tetraophthalmus (Coleoptera: Cerambycidae): Tests at the population and individual level. Environmental Entomology 28: 1044-1051.; Solga et al. 2014Solga MJ, Harmon JP, Ganguli AC. 2014. Timing is everything: An overview of phenological changes to plants and their pollinators. Natural Areas Journal 34: 227-234.; Thompson & Gilbert, 2014Thompson K, Gilbert F. 2014. Phenological synchrony between a plant and a specialised herbivore. Basic and Applied Ecology 15: 353-361.), which in turn influence the preference and performance of insects that use flowers as a resource (Solga et al. 2014Solga MJ, Harmon JP, Ganguli AC. 2014. Timing is everything: An overview of phenological changes to plants and their pollinators. Natural Areas Journal 34: 227-234.).

Recent studies on flower-pollinator interactions indicate that species in synchronous flowering (i.e., co-flowering) increase pollinator attraction by facilitating the detection of plants, resulting in greater success in visitation and pollination (Shrestha et al., 2019Shrestha M, Dyer AG, Garcia JE, Burd M. 2019. Floral colour structure in two Australian herbaceous communities: It depends on who is looking. Annals of Botany 124: 221-232.; Albor et al. 2020Albor C, Arceo‐Gómez G, Parra‐Tabla V. 2020. Integrating floral trait and flowering time distribution patterns help reveal a more dynamic nature of co‐flowering community assembly processes. Journal of Ecology 108: 2221-2231.), as predicted by the resource concentration hypothesis (RCH) (Root, 1973Root RB. 1973. Organization of a plant‐arthropod association in simple and diverse habitats: The fauna of collards (Brassica oleracea). Ecological Monographs 43: 95-124.). The RCH predicts that a greater aggregation of resources would serve as a display of attraction for consumers, both mutualists and antagonists. Given that floral traits that evolved as attractants to pollinators may also attract antagonistic species such as florivores (Adler & Bronstein, 2004Adler LS, Bronstein JL. 2004. Attracting antagonists: Does floral nectar increase leaf herbivory? Ecology 85: 1519-1526., Boaventura et al. 2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.), plants in synchronous flowering that form patches can act as a display of attraction for florivores. Furthermore, these patches can influence the movement of florivorous insects, so that plants that are spatially close and with different quality and abundance of resources due to interspecific synchronous flowering would maximize the search and collection of resources by these insects per unit of time. Spatio-temporal variation in resources is particularly important in urban landscapes, where the use of trees as ornamental species is common, but not necessarily planned from the perspective of animals that use such resources, especially in very fragmented landscapes (see Pena et al. 2017Pena JC, Martello F, Ribeiro MC, Armitage RA, Young RJ, Rodrigues M. 2017. Street trees reduce the negative effects of urbanization on birds. PLoS One 12: e0174484. ). Therefore, studies regarding plant-animal interactions are crucial to understanding the impacts of tree planting in urban green areas in insect foraging and plant reproduction, being essential for the conservation of urban biodiversity (Aronson et al. 2017Aronson MF, Lepczyk CA, Evans KL et al. 2017. Biodiversity in the city: Key challenges for urban green space management. Frontiers in Ecology and the Environment 15: 189-196.; Dylewski et al. 2020Dylewski L, Maćkowiak L, Banaszak-Cibicka W. 2020. Linking pollinators and city flora: How vegetation composition and environmental features shapes pollinators composition in urban environment. Urban Forestry & Urban Greening 56: 126795.; Cornelissen et al. 2023Cornelissen TG, Lourenço GM, Costa et al . 2023. Insects in the cities: Patterns of biodiversity, interactions and ecosystem services in urban green areas. In: Angeoletto F, Tryjanowski P, Fellowes M (eds.). Ecology of Tropical Cities: Natural and Social Sciences Applied to the Conservation of Urban Biodiversity. Cham, Springer. (in press).).

In order to understand how synchronous flowering affects spatial and temporal insect-flower interactions, this study tested the hypothesis that plants that flower synchronously represent massive resources to insects, and such resources influence florivorous insects. We predicted that levels of florivory among plant species that flower synchronously are influenced by plant distance, as plants that are close together, regardless of species, would show similar levels of florivory incidence and intensity. We tested this hypothesis by studying six species of Bignoniaceae plants in an urban landscape in a large tropical city. The Bignoniaceae family has several deciduous species (Seibert, 1948Seibert RJ. 1948. The use of glands in a taxonomic consideration of the family Bignoniaceae. Annals of the Missouri Botanical Garden 35: 123-137.) that lose their leaves just before they flower in the dry season (Boaventura et al. 2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.) and this characteristic allowed us to understand the space-time complexity of florivory, as there is a temporal separation of the resources used by insects (leaves or flowers) allowing attention to each interaction (herbivory and florivory) in different space-time dimensions.

Material and Methods

Study area

The study was conducted in the urban landscape of Belo Horizonte, Minas Gerais, a planned metropolis in southeastern Brazil. Sampling occurred during the dry season (August-September) of 2021 in several urban green spaces in the northern section of the city (W 19°51' 57'', S 43°58' 06'') in the Pampulha Lake region.

The metropolitan region of Belo Horizonte is located in a transition region between the Atlantic Forest and the Cerrado biomes (Rezende et al. 2021Rezende JC, Rocha BM, Abrahão KCDFJ. 2021. Fachadas Vegetadas em áreas urbanas: Estudo de caso em Belo Horizonte. Periódico Técnico e Científico Cidades Verdes 9: 83-99.). The climate, according to the Köppen classification, is type Cw _a - high altitude tropical with dry winter and rainy summer (Meyer et al. 2004Meyer ST, Silva AF, Marco-Júnior PD, Meira-Neto JAA. 2004. Composição florística da vegetação arbórea de um trecho de floresta de galeria do Parque Estadual do Rola-Moça na Região Metropolitana de Belo Horizonte, MG, Brasil. Acta Botanica Brasilica 18: 701-709.; Rezende et al. 2021Rezende JC, Rocha BM, Abrahão KCDFJ. 2021. Fachadas Vegetadas em áreas urbanas: Estudo de caso em Belo Horizonte. Periódico Técnico e Científico Cidades Verdes 9: 83-99.), and annual average temperature and relative humidity of 21.1ºC and 72.2%, respectively (Ferreira et al. 2017Ferreira DG, Assis ESD, Katzschner L. 2017. Construction of an analytical climate map for the city of Belo Horizonte, Brazil. Brazilian Journal of Urban Management 9: 255-270.).

Study System

The Bignoniaceae family presents species in synchronous flowering in the urban landscape during the dry season in the city. Bignoniaceae is a family of pantropical plants, present in all tropical flora, especially in South America. In Brazil, this family has 34 genera and 420 species, of which 212 are endemic (Lohmann et al. 2020Lohmann LG, Kaehler M, Fonseca LHM et al. 2020. Bignoniaceae in Flora do Brasil 2020. https://floradobrasil.jbrj.gov.br/reflora/floradobrasil/FB11230. 11 Jan. 2022.
https://floradobrasil.jbrj.gov.br/reflor... ).

To evaluate the effects of synchronous flowering on florivory levels, six sympatric tree species distributed in three distinct genera were selected: Handroanthus, Jacaranda and Tabebuia. All species evaluated in our study present flowers while leafless, usually during the dry season. In total, 48 trees belonging to six species were studied: Handroanthus chrysotrichus (Mart. ex DC.) Mattos, H. heptaphyllus (Vell.) Mattos, H. ochraceus (Cham.) Mattos, Jacaranda mimosifolia D. Don, Tabebuia aurea (Silva Manso) Benth. & Hook.f. ex S.Moore and T. roseoalba (Bertol.) Bertero ex A.DC. A description of each species can be found in Supplementary Material S1.

Data collection

To assess whether there is spatial variation in florivory levels in phylogenetically related plant species with synchronous flowering, individuals of the six species evaluated were marked in the field during the dry season of 2021, between August and September. The studied area comprehends the green surroundings of the campus of the Federal University of Minas Gerais in the Pampulha Lake county and all the individuals flowering between August 28th and September 29th of 2021 were tagged in the field using plastic tags. At the moment of sampling, we found 48 individuals of Bignoniaceae with flowers (Supplementary Material S2), Handroanthus chrysotrichus (n=11), H. heptaphyllus (n=15), H. ochraceus (n=5), Jacaranda mimosifolia (n=8), Tabebuia aurea (n=3) and T. roseoalba (n=6). The individuals were identified according to reference material deposited in the BHCB Herbarium at the Federal University of Minas Gerais (voucher numbers: H. chrysotrichus - BHCB-68094, H. heptaphyllus - BHCB-42.217, H. ochraceus - BHCB-32.957, J. mimosifolia - BHCB-21532, T. aurea - BHCB-217827, and T. roseoalba - BHCB-32952). All individuals were georeferenced based on SIRGAS 2000 (Geocentric Reference System for the Americas) using a Promark 3 GPS coupled to an antenna. The coordinates of each plant were used to build a matrix of the distance (in meters) between all individuals using AutoCAD 2022 (Autocad, 2022Autocad. 2022. Autodesk Autocad version S.051.0.0. Available at: https://help.autodesk.com/view/ACD/2022/ENU/
https://help.autodesk.com/view/ACD/2022/... ).

From each tree, five inflorescences were collected with the aid of pruning shears (individuals smaller than 2.5 m) or an adapted telescope cable (individuals taller than 2.5 m). The collected flowers were placed in identified plastic bags, taken immediately to the laboratory using dry ice and refrigerated. In the laboratory, all flowers sampled were evaluated and flowers with signs of florivory by chewing insects were immediately numbered and digitized. Flowers without visible damage and/or with signs of nectar robbing only (small perforations on the corolla but without tissue removal) were discarded. Flower collection and processing followed the protocol proposed by Boaventura et al. (2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.).

Florivory was quantified using ImageJ (Schneider et al. 2012Schneider CA, Rasband WS, Eliceiri KW. 2012. NIH Image to ImageJ: 25 years of image analysis. Nature Methods 9: 671-675.) after all digital images were calibrated at 0.01 mm. Florivory incidence was quantified as the proportion of attacked flowers in each plant and florivory intensity indicates the level of damage and was quantified by adding all the floral areas removed divided by the total area of each flower, multiplied by 100 to express florivory in percentage of tissue lost. Subsequently, the mean florivory incidence was calculated using each plant per species as a replicate. Florivory intensity was calculated using all damaged flowers by individual as replicates and individuals as replicates per species.

Data analysis

Effects of simultaneous flowering on florivory

To assess whether synchronous flowering influenced the levels of florivory of the six studied species, the average florivory was calculated using the inflorescences and flowers nested within individual trees as replicas of each species and the individual plants as replicas of the plant species. The average levels of florivory of each species were compared using generalized linear mixed models (GLMMs). In all models, florivory incidence and intensity were used as response variables and plant species were used as explanatory variables. To account for variation in florivory due to species-specific traits, we added the plant's species as a random variable. Post-hoc differences among plant species were tested using Tukey's test. Error distribution was tested and fitted a gamma distribution for florivory intensity and a negative binomial for florivory incidence. All analyses were performed using the software R and the package lme4 (Bates et al. 2015Bates D, Mächler M, Bolker B, Walker S. 2015. Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software 67: 1-48.).

Spatial variation in florivory in species of Bignoniaceae

To assess whether florivory intensity exhibits spatial structure across the distribution of plants in the urban landscape, a spatial autocorrelation analysis was performed using the mean value of florivory intensity per individual plant. The spatial structure of florivory intensity was determined by calculating Moran's I index (Legendre & Legendre 1988). Each plant was used as a spatial point and the linear distance between the plants that form pairs was calculated from the geographic coordinates recorded for each of the individual trees. The geographic distance matrix was created using geodesic distances, i.e., distances that take into account the curvature of the Earth's surface. Distance values were then grouped into distance classes following Legendre & Legendre (1998Legendre P, Legendre L. 1998. Numerical Ecology - Developments in Environmental Modeling. Amsterdam, Elsevier Science Ltd.) and Moran's I (d) was calculated for 11 distance classes, equally spaced at every 100 m. A spatial correlogram was plotted using Moran's coefficients as a function of the different distance classes between plants. The Moran coefficient has values between [- 1, + 1], where positive values indicate positive autocorrelations and negative values indicate negative autocorrelations (Legendre & Legendre, 1998Legendre P, Legendre L. 1998. Numerical Ecology - Developments in Environmental Modeling. Amsterdam, Elsevier Science Ltd.). The significance of the autocorrelation coefficient was tested by calculating confidence intervals for each distance class, and the overall significance of the correlogram was tested after Bonferroni's correction for multiple comparisons (Legendre & Legendre, 1998Legendre P, Legendre L. 1998. Numerical Ecology - Developments in Environmental Modeling. Amsterdam, Elsevier Science Ltd.). All analyses were conducted in R using lctools and spdep (Bivand, 2022Bivand R. 2022. R Packages for Analyzing Spatial Data: A Comparative Case Study with Areal Data. Geographical Analysis 54: 488-518.).

Results

A total of 2,467 flowers belonging to the 48 individuals of the six species flowering synchronously were evaluated. More than a third (31.3%) of the individuals sampled belonged to the species Handroanthus heptaphyllus, followed by H. chrysotrichus (22.9%) and Jacaranda mimosifolia, which represented 16.7% of the individuals sampled. Of the total flowers collected, 34.01% were of H. heptaphyllus, which presented the highest number of flowers evaluated in the present study. Tabebuia aurea was the species with the lowest number of individuals sampled (6.5%), and also the lowest number of available flowers, corresponding to only 2.11% of the sampled flowers.

The proportion of intact flowers sampled in the present study was high compared to the proportion of flowers attacked by florivores (1,568 and 899 flowers, respectively), and intact flowers represented about 64% of all flowers evaluated. Among the studied species, the ratio of attacked flowers varied between 29 and 78%, with T. aurea being the species with the highest incidence of florivory (79% of the flowers sampled with damage, Fig. 1), followed by H. ochraceus (68.2%), T. roseoalba (41.2%) and H. heptaphyllus (40%). H. chrysotrichus and J. mimosifolia were the species with the lowest florivory incidence (around 30% of flowers with damage by florivores on each species).

Figure 1.
Proportion of attacked flowers and intact flowers in each of the six species of Bignoniaceae sampled in synchronous flowering. Light gray bars indicate intact flowers and dark gray bars indicate flowers with some sign of florivory. The numbers inside the bars indicate the number of flowers in each category.

Effects of simultaneous flowering on florivory

The six plant species showed significantly different levels of florivory. Florivory incidence (X ² =3.28, d.f.=5, P=0.014, Fig. 2 A ) was significantly different among the studied species and the highest frequency of florivory was found in the two species of yellow ipês, T. aurea (78.84%) and H. ochraceus (68.18%), which were heavily attacked by florivores. In contrast, H. chrysotrichus, another species of yellow ipê, presented the lowest frequency of florivory amongst the studied species (31%).

Figure 2.
A) Florivory incidence (frequency of attacked flowers) and B) florivory intensity (percentage of floral area removed by insects) in each of the six evaluated species in synchronous flowering. Differences were tested using a GLMM and different letters indicate significant differences between species after the Tukey post-hoc test. Dots indicate each one of the plants sampled for each plant species and solid horizontal lines indicate the mean and vertical lines around the means indicate standard errors.

Florivory intensity also differed among plant species (X ²=3.86, d.f.=5, P=0.006, Fig. 2 B ) and most plants studied exhibited high levels of floral damage. Handroanthus chrysotrichus was the only plant species with a very low level of florivory intensity among the evaluated species (about 1% of floral tissue removal), whereas T. aurea had nine times higher levels of floral tissue removal (about 10%), being the species most intensely consumed by florivores (Tab. 1).

Spatial variation in florivory in species of Bignoniaceae

The smallest distance recorded between pairs of plants in the urban landscape was 5 m (individuals 21 and 22) and the greatest distance between pairs of plants in synchronous flowering was more than 1 km (1.115,37 m; individuals 1 and 26).

The correlogram constructed to evaluate the spatial distribution of florivory intensity in the landscape indicated positive autocorrelations at small distances, i.e., pairs of plants that were close and with distances between 5 and 200, i.e., distance classes 1 and 2, exhibited similar values of attack on flowers (Fig. 3) when compared to plants at greater distances. The distance between zones of significant similarity and dissimilarity in florivory intensity is indicated in the correlograms at the distance at which the first significant negative autocorrelation was found (P≤0.05), at approximately 750 m between pairs of plants.

Figure 3.
Spatial correlogram of florivory intensity along the distribution of flowering plants in the urban landscape, as a function of distance between plants. Distance classes were calculated every 106 m, with class 1 including plants from distances between 0 and 106.75 m, class 2 between 106.75 and 213.50 m, up to class 11, which included plants between 1067.25 and 1173.75 m). Colored blue circles indicate significant positive or negative autocorrelations ( α≤0.05) and gray circles indicate non-significant values.

Discussion

Our study showed that phylogenetically related plant species in synchronous flowering show different levels of florivory, but individual plants share similar levels of florivory when they are geographically close. Distances between individuals ranged from 5 m to 1,115 m, but a positive spatial correlation between distance and florivory was found only in those plants that were within 200 m of each other. Our data also indicate that plant species in synchronous flowering in the urban landscape exhibit different levels of florivory, both in terms of frequency of flower attack (incidence) and in terms of the amount of floral tissue consumed (intensity) by insect florivores.

Although synchronous flowering between different species promotes massive resource availability for florivorous insects, our study with six species spatially and phylogenetically close indicates that synchronous flowering alone is not able to explain the differences in levels of florivory observed in these plant species in the urban landscape. According to the RCH (Root 1973Root RB. 1973. Organization of a plant‐arthropod association in simple and diverse habitats: The fauna of collards (Brassica oleracea). Ecological Monographs 43: 95-124.), plants that are spatially and geographically close would be more easily found by herbivores because they function as an "attractive display" for these insects. Moreover, spatially close plants that form patches and offer resources simultaneously would keep insects longer in these high-resource islands, increasing therefore damage levels, especially when insects are generalists (Boaventura et al. 2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.). The RCH has already been tested in several studies evaluating herbivory by chewers (Long et al. 2003Long ZT, Mohler CL, Carson WP. 2003. Extending the resource concentration hypothesis to plant communities: Effects of litter and herbivores. Ecology 84: 652-665.; Stephens & Myers, 2012Stephens AE, Myers JH. 2012. Resource concentration by insects and implications for plant populations. Journal of Ecology 100: 923-931.), gall-formers (Boaventura et al. 2018Boaventura MG, Pereira CC, Cornelissen T. 2018. Plant architecture influences gall abundance in a tropical montane plant species. Acta Botanica Brasilica 32: 670-674.;) and sap-suckers (Jaworski et al. 2022Jaworski CC, Thomine E, Rusch A et al. 2022. At which spatial scale does crop diversity enhance natural enemy populations and pest control? An experiment in a mosaic cropping system. Agronomy 12: 1973.), but this was the first study evaluating florivore-flower interactions. Considering that flowers are an ephemeral and potentially valuable resource for florivorous insects and that levels of florivory in individual plants tend to be higher than those of foliar herbivory (see Boaventura et al. 2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.), synchronous flowering in spatially close patches in the urban landscape - where there is great fragmentation and loss of habitats - represents a crucial and patchy resource for florivores. However, it is important to highlight that plant characteristics such as height, number of flowers, number of days the flowers are open and variation in the diversity of the florivore community can alter the levels of florivory by influencing how florivores find the resources and stay onto those patches. Thus, although this study evaluated spatial dimensions of urban florivore-flower interactions, it has been suggested that other plant traits, the flowers themselves and plant arrangement in the urban environment can influence the attractiveness and also the consumption of flowers. We, therefore, suggest that the interaction between insects and plants with synchronous flowering should also be evaluated in the context of functional traits of flowers. Recent studies have demonstrated, for example, that the size of the flowers (Teixido et al. 2018Teixido AL, Dayrell RLC, Arruda AJ et al. 2018. Differential gender selection on flower size in two Neotropical savanna congeneric species. Plant Ecology 219: 89-100.) and their longevity (Zhao et al. 2020Zhao Z, Hou M, Wang Y, Du G. 2020. Phenological variation of flower longevity and duration of sex phases in a protandrous alpine plant: Potential causes and fitness significance. BMC Plant Biol 20: 137.) are important functional traits for pollinator attraction and effectiveness, and these traits should also be evaluated in the context of florivory. The Floral Economic Spectrum (FES) (Roddy et al. 2021Roddy AB, Martínez‐Perez C, Teixido AL et al. 2021. Towards the flower economics spectrum. New Phytologist 229: 665-672.) suggests the combined importance of pollinators, abiotic factors and antagonists in the evolution of flower shape, but florivores have been commonly neglected as important agents of selection in floral traits. We suggest that the identity of the florivores in the studied species of Bignoniaceae should be examined in the context of diet breadth and ability to disperse in the urban landscape to better understand the patterns of flower selection and resource use.

Our study demonstrated variation in florivory intensity and incidence among plant species, but only Handroanthus chrysotrichus exhibited significantly lower levels of florivory compared to the other Bignoniaceae species. Handroanthus heptaphyllus, H. ochraceus and Tabebuia roseoalba showed levels of florivory similar to those already reported in the literature (see Boaventura et al. 2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.), reinforcing the suggestion that urban plants might represent patches of resources to both antagonistic and mutualistic insects. The low levels of florivory experienced by H. chrysotrichus may be related to the ontogenetic state of the sampled individuals, as only young individuals with a height of less than 2.5 m were flowering during the time of our study. These individuals might be more difficult to be found by florivores in the urban landscape, as proposed by the plant apparency hypothesis (Feeny, 1976Feeny P. 1976. Plant apparency and chemical defense. In: Wallace JW, Mansell RL (eds.). Biochemical interaction between plants and insects, Recent Advances in Phytochemistry, vol 10. Boston, Springer. p. 1-40.; Smilanich et al. 2016Smilanich AM, Fincher RM, Dyer LA. 2016. Does plant apparency matter? Thirty years of data provide limited support but reveal clear patterns of the effects of plant chemistry on herbivores. New Phytologist 210: 1044-1057.) or in an alternative scenario, flowers in young individuals may be better protected against insects, indicating the important role of secondary chemistry for florivores, as has been demonstrated for herbivores (Smilanich et al. 2016Smilanich AM, Fincher RM, Dyer LA. 2016. Does plant apparency matter? Thirty years of data provide limited support but reveal clear patterns of the effects of plant chemistry on herbivores. New Phytologist 210: 1044-1057.). Moreover, individuals of this particular plant species were located in streets with high traffic and public, which might have precluded florivores from reaching the flowers and consuming flower tissue. Handroanthus ochraceus and T. aurea, two other species of yellow ipê trees presented the highest levels of florivory (8.74 and 9.84%, respectively), reaching more than 65% of their flowers attacked by insects.

Pioneering studies on insect-flower interactions have already indicated that flower color can be used as a proxy to predict the type of secondary metabolite that is predominantly expressed in plants (e.g., Irwin et al. 2003Irwin RE, Strauss SY, Storz S, Emerson A, Guibert G. 2003. The role of herbivores in the maintenance of a flower color polymorphism in wild radish. Ecology 84: 1733-1743.), and such functional trait could help to explain the levels of florivory found in the studied species with yellow flowers. Although there is yet no consensus on how corolla color can influence florivore-flower interactions, it is known that yellow flowers have lower concentrations of anthocyanins and glucosinolates compared to purple and pink flowers (Irwin et al., 2003Irwin RE, Strauss SY, Storz S, Emerson A, Guibert G. 2003. The role of herbivores in the maintenance of a flower color polymorphism in wild radish. Ecology 84: 1733-1743.). Yellow flowers show recessive anthocyanin (A-) color morphs (Stanton et al. 1986Stanton ML, Snow AA, Handel SN. 1986. Floral evolution: Attractiveness to pollinators increases male fitness. Science 232: 1625-1627.; Stanton et al. 1989Stanton ML, Snow AA, Handel SN, Bereczky J. 1989. The impact of a flower‐color polymorphism on mating patterns in experimental populations of wild radish (Raphanus raphanistrum L.). Evolution 43: 335-346.; Irwin et al. 2003Irwin RE, Strauss SY, Storz S, Emerson A, Guibert G. 2003. The role of herbivores in the maintenance of a flower color polymorphism in wild radish. Ecology 84: 1733-1743.), whereas pink flowers have the dominant anthocyanin (A+) morph. Furthermore, chemical analyses of foliar glucosinolates showed that (A+) morphs produce higher concentrations of indole glucosinolates when compared to morphs (A-) in the presence of foliar herbivores. These compounds have the potential to negatively affect the preference and performance of insects (Irwin et al. 2003Irwin RE, Strauss SY, Storz S, Emerson A, Guibert G. 2003. The role of herbivores in the maintenance of a flower color polymorphism in wild radish. Ecology 84: 1733-1743.) and lower concentrations in yellow flowers might aid in explaining higher florivory incidence and intensity in these particular plant species. However, it is known that flower color and chemical defenses are just a few traits that might explain the preference and performance of insects and it is unlikely that these isolated mechanisms can explain the patterns of florivory exhibited by all individuals (Kevan, 1972Kevan PG. 1972. Floral colors in the high arctic with reference to insect - flower relations and pollination. Canadian Journal of Botany 50: 2289-2316.; Irwin et al. 2003Irwin RE, Strauss SY, Storz S, Emerson A, Guibert G. 2003. The role of herbivores in the maintenance of a flower color polymorphism in wild radish. Ecology 84: 1733-1743.). Tabebuia roseoalba, the white ipê tree, also presents the morph (A-), with lower concentrations of glucosinate, which might also explain its high levels of florivory (6.34%) compared to the pink ipê tree H. heptaphyllus, which exhibited lower levels (4.42%) of flower damage. Moreover, flower longevity should also be considered when examining florivory levels (see Boaventura et al., 2022Boaventura MG, Villamil N, Teixido AL et al. 2022. Revisiting florivory: An integrative review and global patterns of a neglected interaction. New Phytologist 233: 132-144.) as this trait might also be related to flower investment in chemical defenses and flower apparency.

Our study is pioneer in investigating antagonistic interactions between insects and plants in urban landscapes and indicates that variation in levels of florivory between plant species that are phylogenetically close exist and should be interpreted considering environmental cues and floral traits that influence the way florivores find and use these resources. We also suggest that plant proximity should also be considered when trees are planted in the cities, as this factor can influence flower damage and insect dispersion and movement patterns in the urban landscape.

Acknowledgments

The authors would like to thank the Graduate Program in Ecology (ECMVS) at UFMG and the Center for Ecological Synthesis and Conservation (CSEC) for their support. Marina Andrade acknowledges the research scholarship from Conselho Nacional de Desenvolvimento Científico e Tecnológico - CNPq/UFMG. Maria Gabriela Boaventura acknowledges the research scholarship from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES (Financial code 001) and Giselle Lourenço acknowledges the research scholarship from CAPES (88887.692955/2022-00). TC thanks the continuous support of Fundação de Amparo à Pesquisa do Estado de Minas Gerais - FAPEMIG, CAPES (grant 88887.910715/2023-00) and CNPq (grants 313007/2020-9 and 311243-2023).

References

Supplementary Material

The following online material is available for this article:

Edited by

Editor Chef:

Publication Dates

History