Use of Pheromones to Enhance Foraging Efficiency in Atta cephalotes Rachel Schiesher Biochemistry Department, University of Wisconsin Madison ABSTRACT Atta Formicidae: Myrmicinae are important herbivores in the Neotropics, cutting between 12 and 17% of the vegetation there Holldobler and Wilson 1990. Atta cephalotes uses pheromones to communicate and maximize efficiency during foraging. The presence and specificity of pheromones was studied on leaf segments, between trails in the same colony and be tween different colonies. All studies were completed at the Santuario EcolÃ³gico in Cerro Plano, Costa Rica. Leaves cut by ants were retrieved from the cache at a significantly faster rate than those cut by hand p value = <0.0001, suggesting pheromones we re laid on the plant material. The leaves were also retrieved at a significantly faster rate when compared to leaves collected from ants on a different trail within the colony or from a different colony p value = 0.0386 and 0.0010 respectively. Individua l ants also foraged at a significantly faster rate on their own trail when compared to when they were transferred to a different trail within the colony or to a trail of a different colony both p values = <0.0001. This suggests that the trail pheromones used for foraging are more complex and specifically recognized by workers on a certain trail. A higher specificity of pheromones may prove beneficial to the enhancement of worker specialization and colony foraging efficiency. RESUMEN Las hormigas Atta so n herbÃvoros muy importantes en los neotrÃ³picos, consumiendo entre el 12 y el 17% de la vegetaciÃ³n. Atta cephalotes usa feromonas para comunicarse y forrajear eficientemente. Se estudiÃ³ la presencia y especificidad de las feromonas en pedazos de hojas, ent re senderos en el mismo hormiguero y entre hormigueros diferentes. Todos los experimentos se realizaron en el Santuario EcolÃ³gico en Cerro Plano, Costa Rica. Las hormigas recolectaron hojas de grupos que fueron cortadas por las hormigas a una velocidad sig nificativamente mÃ¡s rÃ¡pida que las de grupos de hojas cortadas a mano p = <0.0001, sugiriendo que las feromonas fueron puestas en los pedazos de hoja. Las hojas tambiÃ©n fueron recogidas significativamente mÃ¡s rÃ¡pido cuando se les comparÃ³ con las hojas qu e fueron recolectadas de hormigas en un sendero diferente del mismo hormiguero o de un hormiguero completamente diferente p = 0.0386 y 0.0010, respectivamente. Las hormigas individuales tambiÃ©n forrajearon a una velocidad mÃ¡s rÃ¡pida en su mismo sendero e n comparaciÃ³n con hormigas que fueron trasladas a un sendero diferente del mismo hormiguero o a un sendero de un hormiguero diferente p = <0.0001 para los dos tratamientos. Este resultado sugiere que las feromonas de los senderos usados para buscar mater ial de plantas son mÃ¡s complicadas y reconocidas especÃficamente por las hormigas trabajando en un sendero particular. Una especificidad mÃ¡s alta de las feromonas puede ser beneficiosa para el mejoramiento de la especializaciÃ³n de las hormigas y la eficien cia de forrajeo para el hormiguero. INTRODUCTION The most successful organisms are those that can forage the most efficiently; being able to obtain the maximum net yield of energy creates a higher fitness and survival rate for that organism Rockwood an d Hubbell 1987. Group foraging, division into castes and mass communication are ways that have enhanced foraging efficiency in ants.
Atta Formicidae: Myrmicinae is a genus of leafcutting ants in the Neotropics. They are the dominant herbivores in thes e systems and cut between 12 and 17 percent of produced vegetation Holldobler and Wilson 1990. Organizing millions of individuals to accomplish such a feat can be complex and require specialization, motivation, and effective communication. A caste system is observed in leafcutting ants, allowing individuals to focus on one task and be proficient. Ants have reduced or no eyes and rely on chemical, tactile and mechanical signals for effective communication Holldobler 1995. Atta cephalotes find, cut and c arry leaf segments or flowers back to their nest to be used as a substrate for fungal gardens. Clearing a main trail increases foraging rates as much as four to ten times when compared to foraging through the leaf litter Rockwood and Hubbell 1987.Chemica l signals are also secreted and recognized by workers in order to coordinate the recruitment of plant material. In general, the trail pheromones are volatile compounds with a high molecular weight Holldobler and Wilson 1990. These characteristics allow t hem to be more complex than and not as volatile as alarm pheromones, which have a lower molecular weight and can be sometimes recognized by different species Hughes et al . 2001, Robinson et al. 1974. It is important to determine when and how the chemical si gnal gets used to increase the foraging efficiency of A. cephalotes . It has been suggested that the ants apply pheromones to the leaf when it is being cut, as well as along the trail, nest, and trunk to orientate the foragers Holldobler and Wilson 1990. Marking the supplies and foraging area with information may be the best way to organize group foraging. Pheromones on the plant material itself may help in recognition and retrieval if foraging is interrupted and the material dropped. Plant material select ion is very specific and can be time consuming Holldobler and Wilson 1990. Retrieving material that has already been chosen and carried a distance would therefore be more efficient than finding and retrieving new material. The specificity of pheromones is essential for efficient communication. Many species of ants use different concentrations and mixtures of compounds in order to elicit the correct response, for example recruitment/foraging, alarm, or attack Holldobler 1995. It may be helpful to use di fferent pheromones for each of the different trails within a colony while foraging. Optimal foraging theory predicts that the more energy expended on acquiring food should result in a larger reward Holldobler and Wilson 1990. Therefore, more ants may be allocated to a trail where the plant material being collected is in greater demand or of a higher quality. A trail specific pheromone may also be useful for the workers to specialize and remain on a single trail. A high rate of trail fidelity was found amo ng A. cephalotes ; an average of 94% of the marked ants remained on the same trail and an average of 74% returned to the original trail after being removed for 24 hours Romer 2004. A colony specific pheromone would be useful in foraging to assure allegia nce as well as exclusive resource accessibility. The workers would benefit the colony and therefore themselves if they were loyal and only foraged for their original colony. It is possible that there is competition for the same resources when there are mul tiple nests in the same area. If each colony within the same species had an unrecognizable odor, the trails of other colonies would not be exploited. Previous research suggested that the ants distinguished between plant material collected from a different colony, and they also
recognized it after retrieving unmarked material, and discarded it Berry 1994. A colony odor could also be helpful for protection against other conspecifics that may attack the nest Jaffe and Marcuse 1983. If an individual from a different colony is recognized because they have a distinguishing colony pheromone, it would be easier to organize a counter attack and remove the threat. It was hypothesized that A. cephalotes use specific pheromones to distinguish between plant material , different trails, and different colonies in order to increase their foraging efficiency. It may also be that there is a general recruitment pheromone used and recognized by all conspecifics. This would mean that there are other mechanisms that colonies o f A. cephalotes have developed in order to assure successful foraging. METHODS Study Site The study was carried out at the Santuario EcolÃ³gico in Cerro Plano, Monteverde Fig. 1a in the spring of 2006. Figure 1b depicts the layout of the study site, which used to be a farm and is now primarily secondary growth. Data were obtained from each of the four well established A. cephalotes nests located on the property. Most observations were made between 7:00 a.m. and 12:00 p.m., while the ants were still fo raging at a steady rate.
Figure 1a. Map of Monteverde, Cerro Plano and Santa Elena in northwestern Costa Rica in the Puntarenas province. All data were taken from the Santuario EcolÃ³gico highlighted, Cerro Plano.
Figure 1b. Map of Santuario E colÃ³gico located in Cerro Plano, Costa Rica. The trail system is depicted and the four A. cephalotes nests used for data collection labeled. Natural History A. cephalotes workers will follow the foraging trail to the resource, then carry the cut plant m aterial back to the nest Holldobler and Wilson 1990; therefore, if they are carrying plant material, they should be headed back toward the nest. A. cephalotes exhibits caching behavior during a disturbance such as rain, obstacles in the trail, or an atta ck on the nest. The ants will drop the leaf they are carrying into a pile if they are foraging, and attend to the more urgent matter. Hart and Ratnieks determined that this behavior makes it more probable that they will later retrieve the leaves 2000. Leaf Marking E XPERIMENT 1: The rate at which marked and unmarked plant material was retrieved from a cache was determined in order to see if pheromones are placed on the plant material itself. Leaves being carried by the ants were collected using tweezer s and assumed to have pheromones present. Then leaves from the same species of tree were hand cut to approximately the same size one cm 2 , taking care not to contaminate them with other scents. Every trial presented ten pieces of plant material for each t reatment on opposite sides of a main trail and observed for a maximum of twenty minutes. The rates of retrieval were then compared using a Wilcoxon sign rank test to see if there was a significant difference between the treatments.
Trail Marking E XPERIME NT 2: Leaves were collected from ants on two different trails of the same colony to determine if there were trail specific pheromones. The rate at which the leaves were retrieved from the different caches was recorded for 26 trials and compared using a Wil coxon sign rank test to see if there was a significant difference in recognition by the ants. E XPERIMENT 3: Ants carrying leaves were transferred using tweezers from one trail to a different trail of the same colony to determine if there were trail speci fic pheromones laid on the actual trail. Toothpicks were used to mark the replacement point and an arbitrary distance in either direction along the path 17 cm. The rate and direction that 131 individuals traveled on both trails was recorded and compared with the Wilcoxon sign rank test. Colony Odor E XPERIMENT 4: Methods and analysis from experiment 2 were performed using leaf samples collected from two different colonies for a total of 28 trials. E XPERIMNET 5: Methods and analysis from experiment 3wer e performed with 123 individuals moved between two different colonies using a compartmentalized box. RESULTS Leaf Marking E XPERIMENT 1: The leaves that were cut by the ants were retrieved from the caches at a significantly faster rate than those that w ere hand cut Fig. 2 z value = 4.043, p value = <0.0001. -0.5 0 0.5 1 1.5 2 2.5 3 With Pheromones Without Pheromones Rate Retrieved leaves/min FIGURE 2. Mean rate at which A. cephalotes retrieved cached leaves that were cut by the ants with pheromones and cut by hand without pheromones. Every treatment had a pile of ten leaves and a maximum observation time of 20 minutes. n = 24. The bars represent Â± 1 standard deviation of the means.
Trail Marking E XPERIMENT 2: Leaves from the original trail were retrieved at a significantly faster rate than those from a different trail in the same colony Fig. 3 z value = 2.068, p value = 0.0386. 0 0.5 1 1.5 2 Original Trial Other Trial Rate Retrieved leaves/min FIGURE 3. Mean rate at which A. cephalotes retrieved cached leaves from their trail and a different trail from the same colony. Every treatment had a pile of ten leaves and a maximum observatio n time of 20 minutes. n = 26. The bars represent Â± 1 standard deviation of the means. E XPERIMENT 3: Ants foraged at a significantly slower rate when placed on a different trail in the same colony Fig. 4 z value = 7.142, p value = <0.0001. Thirty fo ur percent of the ants placed on the foreign trail went away from the nest, while 100% of the ants replaced on their original trail went toward the nest. -0.5 0 0.5 1 1.5 2 2.5 3 Original Trail Other Trail Rate cm/sec FIGURE 4. Mean rate at which A. cephalotes foraged on their own trail and when transferred to a dif ferent trail within the same colony. n = 131. The bars represent Â± 1 standard deviation of the means.
Colony Marking E XPERIMENT 4: Leaves from the original colony were retrieved at a faster rate than those from a different colony Fig. 5 z value = 3.304, p value = 0.0010. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 Original Colony Other Colony Rate Retrieved leaves/min FIGURE 5. Mean rate at which A. cephalotes retrieved cached leaves from their own colony and those taken from a different colony. Every treatment had a pile of ten leaves and a maximum observation time of 20 minutes. n = 28 . The bars represent Â± 1 standard deviation of the means. E XPERIMENT 5: The ants foraged at a significantly faster rate within their own colony than when placed on the trail of a different colony Fig 6 z value = 9.511, p value = <0.0001. Forty six p ercent of the ants placed in a different colony went in the opposite direction while 100% of the individuals went toward their own colony. 0 0.5 1 1.5 2 2.5 Original Colony Other Colony Rate cm/sec FIGURE 6. Mean rate at which A. cephalotes foraged on their own trail as well as when transferred to a trail in a different colony. n = 123. The bars represent Â± 1 standard deviation of the means.
DISCUSSION This study was important in determining if there were specific pheromones used to mark the plant material, different trails within a colony, and different co lonies. This behavior would serve as effective mass communication between the workers in A. cephalotes to increase their foraging efficiency. It was found that plant material did get marked with pheromones because they were retrieved from caches at faster rates. The recognition of previously selected and partially transported plant material increases overall foraging efficiency. There was a distinction between leaves being carried on different trails within the same colony. However, this difference was not as significant as the experiment comparing leaves taken from different colonies. The foragers retrieved the leaves taken from their trail at a faster rate in both instances, which supports the idea that the workers can specialize on specific trails and re cognize the material being foraged. This specificity would help to distinguish pre selected plant material from leaf litter around the trails. The data also suggest that there are specific pheromones laid on different trails within the same colony as well as between different colonies. The ants removed from and replaced back onto their original trail recognized almost immediately which direction and where to bring their plant material. Although the individual ants foraged at a significantly slower rate on different trails and in other colonies, this was more than likely an underestimate of their effective foraging rate. There was visibly more uncertainty when an ant was transferred to a different colony, because many seconds were spent with either no moveme nt or little net movement. Ants on foreign trails would sometimes end up going away from the nest and occasionally straight off the trail. This reduced foraging rate may also be partially due to the fact that the ants from the colony would sometimes attack the foreign ant, occasionally seizing the leaf and debilitating the ant. This specificity in trail pheromones maintains exclusive resource accessibility and assures that workers specialize on one trail to maximize efficiency. More specific and therefore c omplex pheromones are used while A. cephalotes is group foraging in order to enhance communication and increase efficiency. A future study could focus on the directionality and specificity of trail pheromones on a single trail. It would be interesting to d etermine how ants know which direction to travel along the trail according to their current task. ACKNOWLEDGMENTS I would like to thank Javier MÃ©ndez and Alan Masters for the continuous guidance and suggestions. Also thanks to Ollie Hyman and Maria Jo st for general support and encouragement throughout the project. A special thanks goes to Mireya Salazar MÃ©ndez for letting me study on her land daily. LITERATURE CITED Berry, K. 1994. Collection Speeds of Marked, Unmarked, and Foreign Leaf Segments f rom the trails of Atta cephalotes . CIEE Program. Summer 1994. Hart, A and F.L. Ratnieks. 2000. Leaf caching in Atta leafcutting ants: discrete cache formation through positive feedback. Animal Behavior 59 :587 591.
Holldobler, B. 1995. The chemistry of s ocial regulation: Multicomponent signals in ant societies. Proceedings of the National Academy of Sciences of the United States of America 92 :19 22. Holldobler, B and E.O. Wilson. 1990. Ants . Harvard Universtiy Press, Cambridge, Mass. Hughes, W.O., P. Ho wse, and D. Goulson. 2001. Mandibular Gland Chemistry of Grass cutting Ants: Species, Caste, and Colony Variation. Journal of Chemical Ecology 27 : 109 124. Jaffe K., and M. Marcuse. 1983. Nestmate recognition and territorial behavior in the ant Odontomac hus bauri Formicidae: Ponerinae. Insecte s Sociaux 30 :466 481. Robinson, S.W., J.C. Moser, M.S. Blum, and E. Amante. 1974. Laboratory investigations of the trail following responses of four species of leaf cutting ants with notes on the specificity of a trail pheromone of Atta texana . Insectes Sociaux 21 :87 94. Rockwood, L and S.P. Hubbell. 1987. Host plant selection, diet diversity, and optimal foraging in a tropical leafcutting ant. Oecologia 74 :55 61. Romer, A. Trail Fidelity in Atta cephalotes . CIEE Program. Fall 2004.