
BioControl (2013) 58:815–820 DOI 10.1007/s10526-013-9541-z Ants defend coffee from berry borer colonization David J. Gonthier • Katherine K. Ennis • Stacy M. Philpott • John Vandermeer • Ivette Perfecto Received: 7 May 2013 / Accepted: 13 August 2013 / Published online: 29 August 2013 Ó International Organization for Biological Control (IOBC) 2013 Abstract Ants frequently prevent herbivores from bored, ant activity (that varied greatly among species) damaging plants. In agroecosystems they may provide was not a significant factor in models. This study is the pest control services, although their contributions are first field experiment to provide evidence that a diverse not always appreciated. Here we compared the ability group of ant species limit the berry borer from of eight ant species to prevent the coffee berry borer colonizing coffee berries. from colonizing coffee berries with a field exclusion experiment. We removed ants from one branch Keywords Biodiversity Á Ecosystem services Á (exclusion) and left ants to forage on a second branch Pest control Á Ant–plant defense Á Coffee Á (control) before releasing 20 berry borers on each Agroecology Á Ant Á Formicidae branch. After 24 h, six of eight species had signifi- cantly reduced the number of berries bored by the berry borer compared to exclusion treatment branches. While the number of berries per branch was a Introduction significant covariate explaining the number of berries Ants benefit plants (Styrsky and Eubanks 2007; Chamberlain and Holland 2009; Rosumek et al. Handling Editor: Arne Janssen. 2009; Trager et al. 2010). Humans have known this Electronic supplementary material The online version of for quite a long time. In fact, ants were described as this article (doi:10.1007/s10526-013-9541-z) contains supple- biological control agents in China around 304 AD (van mentary material, which is available to authorized users. Mele 2008). Many plants have also evolved to promote the activity of ants on their tissues. Surveys of tropical D. J. Gonthier (&) Á I. Perfecto School of Natural Resources and Environment, University forests show that up to one third of all woody plants of Michigan, 440 Church St., Ann Arbor, MI 48109, USA have evolved ant-attracting rewards (Schupp and e-mail: [email protected] Feener 1991). Some plants provide domatia as ant housing structures, while others attract ants to their K. K. Ennis Á S. M. Philpott Environmental Studies Department, University of tissues with extra-floral nectaries. Some plants are California, 1156 High St., Santa Cruz, CA 95064, USA hosts to honeydew-producing hemipterans that excrete honeydew, a sugary substance consumed by ants. Still J. Vandermeer other plants are simply substrates for ant foraging. The Ecology and Evolutionary Biology, University of Michigan, 830 N University Ave., Ann Arbor, MI 48109, majority of studies conducted across these ant–plant USA groups show that ants benefit plants by removal of 123 816 D. J. Gonthier et al. herbivores (Chamberlain and Holland 2009; Rosumek insecticide resistance (Brun et al. 1990). Several of the et al. 2009; Trager et al. 2010). Nonetheless, in many stages of the CBB life cycle make it vulnerable to agroecosystems, the benefits of pest control services by attack by ants (Damon 2000; Jaramillo et al. 2007). ants are not recognized. Agricultural managers often First, the CBB hatches from eggs within the coffee view them as pests or annoyances to agricultural berry, where it consumes the seeds (Damon 2000; production because some ants tend honeydew-produc- Jaramillo et al. 2007). Small ants may enter the berry ing insects that can damage crops (Styrsky and through the beetle entrance hole and predate the larvae Eubanks 2007). However, a review of the literature and adults inside (Larsen and Philpott 2010; Perfecto on ant-hemipteran associations suggests that even and Vandermeer 2013). Second, old berries infested these associations benefit plants indirectly because ants with the CBB may not be harvested because they often remove other, more damaging herbivores (Styrsky and turn black and remain on the coffee branches or may Eubanks 2007, 2010). Regardless, the literature lacks fall to the ground (Damon 2000; Jaramillo et al. 2007). studies investigating ant–plant interactions in agro- These old infested berries may act as a population ecosystems. Here, we broadly survey the pest control reservoir of borer populations and ant predation at this services provided by a suite of ant species to better stage could be very important for limiting CBB understand the role of ant defense of coffee. populations in the next season. Third, as adult borers Coffee is a tropical crop that occurs as an under- disperse (flying or crawling) to colonize new berries, story shrub in its native range, and coffee plants are ants may prevent them from entering new berries therefore often grown under a canopy of shade trees in (Pardee and Philpott 2011; Perfecto and Vandermeer agroforestry systems in some parts of the world 2006; Philpott et al. 2012). To date, no field experi- (Perfecto et al. 1996). This canopy layer provides ment has specifically investigated how coffee-forag- plantatsions with a forest-like vegetation structure that ing ants limit CBB colonization of berries. Here, we can help maintain biodiversity (Perfecto et al. 1996). studied the abilities of eight ant species to prevent Ant biodiversity is high in many coffee plantations and colonization of berries by the CBB. We hypothesized ants attack and prey on many coffee pests, including that only species with high activity on branches would the coffee berry borer (CBB; Hypothenemus hampei limit CBB colonization of berries. We show that six of [Ferrari] [Coleoptera: Scolytidae]) (Armbrecht and eight ant species limit CBB colonization of berries and Gallego 2007; Philpott and Armbrecht 2006; Vander- that the effect of ants is independent of ant activity on meer et al. 2010). For example, Azteca instabilis F. branches. This study is the first field experiment to Smith is a competitively dominant ant that aggres- provide evidence that a diverse group of ant species sively patrols arboreal territories in high densities and limits the CBB from colonizing coffee berries. previous research has found that it impacts the CBB (Larsen and Philpott 2010; Perfecto and Vandermeer 2006). Some laboratory and observational field studies have found that Pseudomyrmex spp., Procryptocerus Materials and methods hylaeus Kempf, and Pheidole spp. may limit the CBB (Jime´nez-Pinto et al. Unpublished work; Philpott et al. Our research was conducted on Finca Irlanda, a coffee 2012). However, other field experiments have not plantation in the Soconusco region of southern Mexico found ants to be biological control agents of the CBB and the site of much ongoing research regarding (Varo´n et al. 2004; Vega et al. 2009). Further, the pest community ecology of the arthropod interaction web control effects of many ant species on the CBB have (Vandermeer et al. 2010). In this region, the CBB is a not yet been evaluated and it could be that previously major pest of coffee (Vandermeer et al. 2010). We documented effects are specific to only a few species. searched for coffee bushes occupied by one of eight Natural ant pest control of the CBB is particularly species that were each abundant enough to obtain important because chemical insecticides used to sufficient replication for this experiment: A. instabilis control CBB are not always effective. This lack of (N = 20), Crematogaster spp. (N = 20), Pheidole effectiveness is in part because the CBB lifecycle synanthropica Longino (N = 19), Pseudomyrmex takes place largely hidden within coffee berries (Vega simplex Smith (N = 30), Pseudomyrmex ejectus et al. 2006) and also because the CBB has developed Kempf (N = 28), Solenopsis picea Emery (N = 31), 123 Berry borer colonization 817 Tapinoma sp. (N = 30), and Wasmannia auropuncta- the leaf platforms of the control and exclusion ta Roger (N = 28) (N = sample size used in exper- branches. After 24 h, we counted the number of berries iment). Our goal was to capture a broad survey of the per branch that had CBBs inside entrance holes. We did ant species that occupy the coffee vegetation in the not count partially bored holes in berries, nor CBBs coffee plantation. Within the plantation, five Crema- that had bored into twigs and leaves. Multiple bored togaster spp. forage in the coffee, however field entrance holes per berry were only counted as one identification at the time was not reliable therefore bored berry. We modified the experiment slightly for taxonomic resolution for Crematogaster spp. P. simplex and P. ejectus because of the difficulty in remained at the genus level. For P. simplex and P. locating these species within a bush using visual cues ejectus it was not always possible to find occupied (see above). For these two species, we used the living bushes by observation of ant foraging. Instead, for P. branch to which the nest was attached to as the control simplex and P. ejectus, we determined occupation by branch (with ants). This was done because we wanted removing all dead twigs on the coffee bush and to make sure that ants were actively foraging on control searching these for ant nests within the hollow branches after the disturbance of removing nests. branches (e.g. Philpott and Foster 2005). We reat- To statistically analyze experimental data, we opted tached the nested hollow branch to a living branch to use linear mixed models instead of paired t tests with thin wire and treated these bushes as bushes because mixed models allow inclusions of experi- occupied by P. simplex or P. ejectus. mental non-independencies through the incorporation To test the effects of each ant on CBB colonization of covariates. We included bush as a random effect in of berries, we performed an ant exclusion experiment.
Details
-
File Typepdf
-
Upload Time-
-
Content LanguagesEnglish
-
Upload UserAnonymous/Not logged-in
-
File Pages6 Page
-
File Size-