Bridge Building and Foraging Efficiency in the Army Ant Eciton b urchellii Daniel Brunelle Department of Ecology and Evolutionary Biology, University of Colorado at Boulder Abstract Efficiency in food transport is vital for all organisms. Eusocial insects are efficient because of decentralized colony control, caste specialization, and unique behaviors. Bridge building in the army ant Eciton burchellii has been widely studied as a mechanism that augment s prey transport efficie ncy. Little data exist on why bridges confer an advantage and how their dimensions facilitate rapid food delivery. It is also less well studied how efficiency on a bridge compares to the average efficiency of ant traffic on transport pathways that do not have ant bridges. Bridg e lengths and widths were recorded and patterns were looked for between bridge dimensions and ant velocity, traffic density, and collision rate as proxies of food transport efficiency . Bridges were also compared to non bridge sites and velocity, traffic de nsity, and collision rates were compared between sites. Bridge dimensions were found t o correlate with an increase in all three parameters . Non bridge sites had significantly higher velocities and fewer collisions, and thus were more efficient. Larger brid ges confer more efficiency up to the point where too many ants in bridges reduces the potential number of foragers. Areas without bridges were far more efficient than sites with bridges. This is possibly explained by the use of brid ges only in uneven terrain; when compared to uneven terrain with no bridge , efficiency may be increased. Resumen La eficiencia en transporte de comida es vital para todos los organismos. Insectos eusociales son eficientes por algunas razones como control descentralizado de l a colonia, especializaciÃ³n de las castas, y comportamientos Ãºnicos. Han estudiando bastante construyendo puentes en la hormiga arriera Eciton burchellii como un mecanismo de aumentaciÃ³n del transporte de la presa. Pocos datos existen por las razones que lo s puentes son ventajas por la colonia y como los dimensiones del puente permiten que la comida esta movido mÃ¡s rÃ¡pido al nido. Tampoco estÃ¡ estudiado como la eficiencia en un puente compara a la eficiencia en una columna normal sin puente. Los largos y anc hos de los puentes estaban medidos y miraba patrones entre los dimensiones del puente y la velocidad d e las hormigas, densidad del trÃ¡ fico de las hormigas, y colisiones entre hormigas como ejemplos de eficiencia de transporte de comida. TambiÃ©n, los puente s estaban comparado s a sitios sin puentes, comparando velocidad, densidad de trÃ¡fico, y colisiones . La s dimensiones de los puentes mostraron una correlaciÃ³n positivamente con los tres parÃ¡metros. Los sitios sin puentes tenÃan velocidades mÃ¡s altas con meno s colisio nes, y por estas razones eran mÃ¡s eficientes. Los puentes mÃ¡s largos son mÃ¡ s eficientes hasta el punto donde hay bastantes hormigas en el puente que reduzca el nÃº mero potencial de hormigas forraj eando. Zonas sin puentes eran mÃ¡ s eficientes que sit ios con puentes. Una explicaciÃ³n es que los puentes estÃ¡n usado s solamente en terrenos desiguales; si hacemos un a comparaciÃ³n a terreno desigual sin puentes, es posibl e que eficiencia pueda ser aumentado.
Introduction Efficiency is the keystone of a successful species. Optimizing energy use is important across all taxa and plays a central role in natural selection. It is thus both a driving force and a result of specialization and evolution. Suboptimal resource use redu ces fitness and lowers competitive ability against other species and conspecifics. Food harvesti ng is one of the most fundamental actions that all organisms do; the behavior of food energy intake should be specialized to be as efficient as possible, gaining the most amount of food energy for the least amount of time spent foraging (Pyke et al. 1977). This is known as optimal foraging theory. Eusocial insect colony interactions are fundamentally complex with emergent properties that arise from selectiv e pressure to act efficiently , both for overall specialization and optimal foraging in general . There is a correlation between behavioral specialization and colony size, as well as intra colonial conflict and colony size (Anderson and McShea 2001) , and col onies with fewer conflicts should be more efficient at foraging . Fewer conflicts allow for more individuality in the actions of the workers. Simple individual interactions happening at the same time is one of the characteristics of complex insect societies, where control becomes decentralized. F or example , different large scale patterns in movement can be driven by the same individual actions (Deneubourg and Goss 1989). Allelomimesis, or imita ting a neighboring conspecific, is a self perpetuating p rocess that raises the decision of one individual to a colony sized reaction (Deneubourg and Goss 1989) . If there is a strong tre nd in individual reactions, the entire group will quickly be galvanized to one task. Ants, for example, must constantly forage in order to su stain the ir colony, and r andom food searching leads on a greater scale to directional hunting (Couzin and Franks 2003) , which increases the success of finding food due to higher recruitment . Foraging efficiency is important to ants because th eir bodies are not as derived for movement as more recently evolved taxa. Hurlbert et al. (2008) calculated that ants run slower than expected for their size; c ompared to small mammals, they have less efficient locomotion, so to support an entire colony they must find other ways to be efficient foragers than just selecting for speed . One of the most widely studied predatory ant species that uses highly specialized foraging behaviors is the army ant Eciton burchellii (Formicidae: Ecitoninae) . Eciton burch ellii is an obligate nomadic predator with central foraging tendencies ( i.e. they bring all prey back to the bivouac instead of eating it immediately) that hunts in a straight line from the bivouac in a swarm often over 5 m wide between dawn and dusk , bringing in about 30,000 prey items from up to 105 m away on an average raid (Franks 1982 ). Their life cycle consists of a 20 day stationary period, where they hunt in a radial pattern from the bivouac, and a 15 day nomadic phase, where the bivouac moves i (Franks and Fletcher 1983 ) . Each swarm uses about one third of the colony (Franks and Fletcher 1983), with an average colony consisting of up to 700,000 workers (Franks 1985; Powell and Franks 2006) . F ood is brought back f rom the swarm to the bivouac via a single pathway, or column , which always links the swarm to the bivouac (Powell and Franks 2007) . When a worker has brought foot back to the bivouac, it returns to the swarm front to continue hunting. Each facet of their h unting behavior requires extreme energy input and organization; efficiency is vital for the success of their lifestyle. E. burchellii has many different specialized behaviors to increase the efficiency of food transport.
On e adaptation specific to E. burchellii is building bridges made of living ants over uneven terrain to smooth prey delivery (Powell and Franks 2007 ). Bridges are thought to be set up in areas of high terrain variability in order to smooth out the column or bypass obstacles more effici ently (Powell and Franks 2007) . Bridging has been shown to effectively increase food intake by the colony , up to a 31 percent net gain in prey (Powell and Franks 2007 ). There is very little information, however, for how bridge structure s increase colony efficiency. In this investigation I will study how E. burchelli bridge structure influences ant foraging efficiency and efficiency of bridges and non bridge sites. The dimensions of the bridges may have an impact on the ease of movement o f workers , making them more efficient at foraging and directly influencing the fitness of the colony. I predict that longer and wider bridges will allow ants to more efficiently transport food to the bivouac. Because both ants returning from the swarm and ants leaving the bivouac use the main column, inbound and outgoing ants collide with each other while trying to get to their destination, and bridge dimensions may influence how often this happens. E fficiency can best be estimated by measuring the velocity of the ants moving over the bridge, the number of col lisions they hav e with other ants , and the rate of ants crossing the bridge. Bridges that are longer and wider should increase the velocity and traffic an d reduce the number collisions. Because the bridges are created to increase foraging efficiency, there should also be a distinct difference in efficiency between bridges and non bridge sites along the column. Methods This study was carried out at the University of Georgia Research Station in the San LuÃ s Valley, Costa Rica 1000 m eters above sea level ) from April 12 May 5, 2011. The research station comprises 62 hectares , 60 percent of which is protected premontane moist forest. An estimated five E. burchellii colonies maximum were observed . I me asured the length and width of 35 bridge s and recorded ant velocity (measured as the bridge length divided by the time taken to cross the bridge for four ants per bridge) , traffic density (number of ants crossing the bridge in 30 sec ) , and collision rate ( the number of ant collisions on each bridge in 30 sec, collision defined as an interaction between two ants which slowed down or altered the course of one or both ants) , on each bridge . The se data were also collected for ten non bridge locations from the main column. Additional Observations M any ants stopped at the edge of a bridge momentarily before they crossed . Traffic density was variable along the non bridge column and ants often moved in groups, regardless of whether they were carrying food . Not all collisions had the same impact: s ome collisions stopped one ant while other times both ants were interrupted , and larger ants frequently crawled directly over smaller ants without any hindrance to either; these interactions were not counted as collisions . It was co mmon for one ant to collide with multiple ants heading the other way or stalling many ants that followed behind.
Results Bridge Structure and Efficiency Bridge dimensions significantly affected v elocity, number of c ollisions, and t raffic density . Length was positively correlated with velocity ( N = 33, R 2 = 0.183, P = 0.01 ; Fi g . 1a ) and traffic density ( N = 33, R 2 = 0.1699, P = 0.0139 ; Fig. 1b ), whereas width was positively correlated with collision rate ( N = 33, R 2 = 0.197 , P = 0.0076 ; Fig . 1c ) and traffic density ( N = 33, R 2 = 0.3377, P < 0.000 5 ; Fig . 1d ). There was also a positive correlation between length and width ( N = 33, R 2 = 0.25, P = 0.0076). Bridge and Control S ites Non t test = 2.4077, df = 43, P = 0.02042; Fig. t test = 9.986, df = 52.245, P < 0.0001; Fig. 2b) and lower collision rates ((Fig. 2). Ants were over three times faster in non bridge sites (mean +/ SD = 42.58 +/ 17.79 versus 12.36 +/ 7.03 mm/sec), and experienced half as many collisions (0.67 +/ 0.56 versus 1.3 +/ 0.73 collisions /sec). Width and traffic density did not differ between bridge and non t test = 0.4774, df = 43, P = 0.6355, and traffic density: t test = 0.7592, df = 43, P = 0.4519). FIGURE 1. The relationship between bridge dime nsions and food transport efficiency in Eciton burchellii. Relationships are shown between a) ant velocity and length, b) traffic density and length, c) traffic density and width, and d) collision rate and width. All correlations shown are statistically si gnificant.
FIGURE 2. Comparison of bridge and non bridge traffic efficiency in Eciton burchellii . Means +/ SE are shown for bridge and non bridge measurements of a) collision rate and b) velocity. Discussion B ridge S tructure and E fficiency Bridge l ength positively correlates with increased velocity and traffic flow, both of which increase the rate of food return ing to the bivouac and therefore foraging efficiency. A longer bridge means that more uneven ground is made of ant bodies. T his result supp orts the theory that ants build bridges in order to reduce time spent climbing over other obstacles. Increased velocity can also account for the increased traffic density because a faster bridge should allow ants to cross more often. Increased bridge widt h was correlated with increased traffic density and increased collision rates. At first this might sound paradoxical, because a higher traffic flow indicates a more efficient bridge while more collisions signi fy a less efficient structure. However , width is 1.7 times more strongly correlated with traffic density than collisions ( R 2 = 0.337 versus 0.197) and the slope of the regression for traffic density and width is 3.35 compared to 1.41 for collision rate and width . In other words, traffic density incr eases faster than collision rate, and efficiency is conserved. If bridge width increases efficiency, why not make bridges excessively wider than the main column, which is generally about 3 cm wide ( Gotwald 1995)? The bridges must be built efficiently as we ll; ants in bridges are not contributing to food transportation , and if too many and Franks (2007) calcula ted that a 23 percent daily net loss in prey would result if ten percent of potential foragers were in bridges, compared to a 27 percent daily net gain in prey wi th only one percent of potential foragers in bridges , assuming each worker in a bridge could have brought back one extra prey item. Bridge width is thus most efficient when conforming to the size of the main column.
The correlation between the length and width of the bridges is not surprising: as the bridge length increases, it makes sense that a corresponding increase in width might help stabilize the bridge by giving support to the minims that construct it. Bridge and Control Sites Ants were much more efficient were bridge s were not necessary. Eciton burchellii moved at both a faster velocity and there were fewer collisions. Because bridges can be placed anywhere there is uneven terrain, the bridge may constitute a choke point that leaves it more susceptible to collisions. This may lead one to question why there are bridges in the first place if they do not contribute more efficient food transport compared to uncovered terrain, but it is important to remember the underlying theory that efficiency dictates bridge building, but bridge building does not dictate efficiency. Compared to a column with no obstacles in its way, a nts crossing a bridge in une ven terrain may take longer to cross; compared to the uneven terrain with no bridge, however, ants may be less efficient than if there were a bridge. Bridges are built whenever an area of the column is slowed down by the terrain and suffers from increased traffic congestion. A dditional O bservations Ants that stopped momentarily before a bridge may have needed to figure out the placement of the bridge; bridges may have to cover unconventional terrain that the ants crossing the bridge need to asses because a temporary halt along the path is far less detrimental to the survival of the colony than an ant completely losing the path of the colony . Ants moving in clusters have been shown to increase overall transport efficiency in Atta c olomb ic a (Dussutour et al. 2009). Because ants leaving the nest will give way to ants heading back with food , including in E. burchellii ( Couzin and Franks 2003 ) , unladen ants heading to the bivouac will often choose to follow a slower, burdened ant than try to go around it. Mathem atical models have shown that the decrease in speed required to follow an ant carrying prey is less wasteful than the average number of collisions an ant would su stain otherwise (Dussutour et al. 2009). Moving in clusters is beneficial for teams of ants th at are carrying food, for teams of ants that have dropped off food at the bivouac and are returning to the swarm, and for reducing overall collision s . Bridges are a well studied but poorly understood method of efficient prey transport in Eciton burchellii . Increasing the length and width of a bridge can augment traffic density and velocity at the expense of a higher collision rate. Bridges are implemented when there is uneven terrain that may hinder individual ants from crossing quickly. O therwise , bridges are not built because they render the colony slower, with more collisions, and are overall less efficient, in the sense of transportation efficiency . Even within a bridge there are still many other efficiency oriented behaviors such as spatial distributio n of traffic flow. Couzin and Franks (2003) show that local traffic rules and pheromone trails collectively partition lanes of inbound and outgoing lanes of movement which also reduce the number of collisions on any particular section of trail. This is a p rime example of how even the smallest caste , the minim, cannot bring food to the bivouac very fast, but it still fulfills an essential role in helping sustain the entire group. Bec ause the smallest caste is slowest and cannot carry as much food ( Franks 198 5), they are less
efficient at food transport. However, minims constitute 51 percent of all ants in bridges and 40 percent of minim behavior is spent in bridges (Nell 2010). Ants fill gaps based on their size and will work together to fill larger holes (Powell and Franks 2007). Minims can fill smaller holes without wasting a larger (and therefore faster) caste, increasing colony efficiency. Further study of this partic ular mechanism should focus on the benefit of a given ant bridge compared to the same l ocation without a brid ge. This may shed light onto strategies by which ants choose to make a bridge in the first place, how they decide to place it, and when building bridges can be more of a hindrance than a benefit. On a broad spectrum this investigation shows that efficiency mechanisms are often implemented on an individual or miniscule level that has repercussions on a much larger scale that can not only affect individual survival but also long ter m fitness of an entire species. Acknowledgements This project would not have been possible witho ut the support of the staff of the CIEE Tropic al Ecology and Conservation prog ram for their advice and equipment . I would like to thank my advisor, Anjali Kumar, for her expansive expertise on army ants and on tropical biology in general , as well as for helping me find research on army ant bridges and helping me analyze my data. I would also like to thank Nathan Sellers for acting as my research assistant a nd for locating the ants both on and off trail even though he had far more work to do than me . Finally I would like to thank the University of Georgia Research Station for giving me access to the forest on their property where the army ants lived. Liter ature Cited A NDERSON , C. AND D.W. M C S HEA . 2001. Individual versus social complexity, with particular reference to ant colonies. Biol. Rev. 76: 211 237. C OUZIN , I. D. AND N.R. F RANKS . 2003 . Self organized lane formation and optimized traffic flow in army ants. Proceedings: Biological Sciences , B. 270: 139 146. D ENEUBOURG , J.L. AND S. G OSS . 1989. Collective patterns and decision making. Ethology Ecology and Evolution 1: 295 311. D USSUTOUR , A., S. B ESHERS , J.L. D ENEUBOURG , AND V. F OURCASSIÃ‰ . 2009. Priority rules govern the organization of traffic on foraging trails under crowding conditions in the leaf cutting ant Atta colombica. Journal of Experimental Biology 212:499 505. F RANKS , N.R. 1982. New Method for Censusing Animal Populations: The Number of Eciton burchelli Army Ant Colonies on Barro Colorado Island, Panama. Oecologia 52: 266 268. FRANKS , N.R. 1985. Reproduction, foraging efficiency and worker polymorphism in army ants. In: Experimental Behavioral Ecology a nd Sociobiology: in Memoriam Karl von Frisch , 1886 1982 (Ed. by B. Holldobler & M. Lindauer), p. 91 107. Sunderland, Massachusetts: Sindauer. F RANKS , N.R. AND F LETCHER , C.R. 1983. Spatial Patterns and Army Ant Foraging and Migration: Eciton b urchelli on Barro Colorado Island, Panama. Behavioral Ecology and Sociobiology 12: 261 270. H URLBERT , A.H., F. B ALLANTYNE IV , AND S. P OWELL . 2008. Shaking a leg and hot to trot: the effects of body size and temperature on running speed in ants. Ecological Entomology 33: 144 154.
G OTWALD , W.H. 1995. Army Ants: The Biology of Social Predation. Cornell University Press, Ithaca, NY. N ELL , C. 2010. Polyethism in Eciton burchellii army ants: a look at task specializations of worker castes. CIEE Monteverde Tr opical Ecology and Conservation, Spring 2010: 165 173. P OWELL , S., AND N.R. F RANKS . 2005. Caste evolution and ecology: a special worker for novel prey. P roceedings: Biological Sciences 272: 2173 2180 P OWELL , S. AND N.R. F RANKS . 2006 . Ecology and the evolution of worker morphological diversity: a comparative analysis with Eciton army ants. Functional Ecology 20: 1105 1114. P OWELL , S. AND N.R. F RANKS . 2007 . How a few help all: living pothole plugs speed prey delivery in the army ant Ecit on Burchellii . Animal Behaviour 73: 1067 1076. P YKE , G.H. AND H.L. P ULLIAM AND E.R. C HARNOV . 1977. Optimal Foraging: A Selective Review of Theory and Tests. The Quarterly Review of Biology 52: 137 154.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00233
Construccin de puentes y la eficiencia de forrajeo en la hormiga arriera Eciton burchellii
Bridge building and foraging efficiency in the army ant Eciton burchellii
Efficiency in food transport is vital for all organisms. Eusocial insects are efficient because of decentralized colony control, caste specialization, and unique behaviors. Bridge building in the army ant Eciton burchellii has been widely studied as a mechanism that augments prey transport efficiency. Little data exist on why bridges confer an advantage and how their dimensions facilitate rapid food delivery. It is also less well studied how efficiency on a bridge compares to the average efficiency of ant traffic on transport pathways that do not have ant bridges. Bridge lengths and widths were recorded and patterns were looked for between bridge dimensions and ant velocity, traffic density, and collision rate as proxies of food transport efficiency. Bridges were also compared to non-bridge sites and velocity, traffic density, and collision rates were compared between sites. Bridge dimensions were found to correlate with an increase in all three parameters. Non-bridge sites had significantly higher velocities and fewer collisions, and thus were more efficient. Larger bridges confer more efficiency up to the point where too many ants in bridges reduces the potential number of foragers. Areas without bridges were far more efficient than sites with bridges. This is possibly explained by the use of bridges only in uneven terrain; when compared to uneven terrain with no bridge, efficiency may be increased.
La eficiencia en el transporte de alimentos es vital para todos los organismos. Los insectos eusociales son eficientes debido al control descentralizado de la colonia, la especializacin de las castas, y los comportamientos nicos. Se ha estudiado extensamente la construccin de puentes en la hormiga arriera Eciton burchellii como un mecanismo que aumenta la eficiencia del transporte de la presa. Existen pocos datos sobre porque los puentes confieren una ventaja y como sus dimensiones facilitan la entrega rpida de alimentos. Tampoco se ha estudiado como la eficiencia en un puente compara la eficiencia en una columna normal sin puente. Se midi el largo y el ancho de los puentes y se mir los patrones entre las dimensiones del puente y la velocidad de las hormigas, la densidad del trfico de las hormigas, y las colisiones entre las hormigas como ejemplos de la eficiencia de transporte de alimentos. Tambin, los puentes estaban comparados a sitios sin puentes, comparando la velocidad, la densidad del trfico y las colisiones. Las dimensiones de los puentes mostraron una correlacin positiva con los tres parmetros. Los sitios sin puentes tenan velocidades ms altas con menos colisiones, y por estas razones eran ms eficientes. Los puentes ms largos son ms eficientes hasta el punto donde hay bastantes hormigas en el puente que reduzca el nmero potencial de hormigas forrajeando. Las zonas sin puentes eran ms eficientes que los sitios con puentes. Una explicacin es que los puentes estn usados solamente en terrenos desiguales; si hacemos una comparacin con el terreno desigual sin puentes, es posible que aumente la eficiencia.
Text in English.
Costa Rica--Puntarenas--Monteverde Zone--San Luis
Costa Rica--Puntarenas--Zona de Monteverde-San Luis
Tropical Ecology Spring 2011
Foraging behavior in animals
Ecologa Tropical Primavera 2011
Comportamiento de forrajeo en animales
t Monteverde Institute : Tropical Ecology