CEC Research | https://doi.org/10.21973/N3PW9W Fall 2019 1/5

Optimalforagingbehaviorindesertharvesterants(Veromessorpergandei)

CarleeBowen1,GeorgeEskander2,BrianLai3,MariamSamy2

UniversityofCalifornia,SantaCruz1,UniversityofCalifornia,Riverside2,UniversityofCalifornia,Merced3

Optimal foraging behavior is exhibited when individuals in a species maximize theirnutritional intake, while minimizing their energy expenditure during the foragingprocess. In this study, we explored the foraging behavior of Black Harvester Ants(Veromessorpergandei)toassessiftheirforaginghabitscoincidedwithoptimalforagingtheory.Weobserved35VeromessorcoloniesinAnzaBorregoStateParkinacontrolledexperimentwhere piles of large and small pumpkin and oat seedswere placed on aforaging path both near and far from the nest.We found thatVeromessorpreferredsmall seedsover large,nearseedsover far,andpumpkinseedsoveroatseeds.TheseresultssupportedourpredictionsthatVeromessorwouldcollecthighervalueandlowercost foods, aligning with patterns predicted by optimal foraging theory. Investigatingthese foraging behaviors allows us to better estimate future patterns of not onlyVeromessor,butotherforagersaswell.Keywords: O.F.T (optimal foraging theory),Veromessor pergandei, foraging cost, foodpreference,fitness

INTRODUCTION

Optimal foraging theory predicts thatanimals are adapted to minimize theirenergy expenditure while simultaneouslymaximizingtheirnutrientintake(OriansandPearson 1979).For example, the beespecies Apis mellifera will select flowerswithhighersugarconcentrationratherthanlower sugar concentration in order tomaximize honey production efficiency(Roubik and Buchmann 1984). Optimalforaging patterns can also be seen inLepomismacrochirus, a speciesof fish thataimstooptimizeforagingtimebyselectingprey that are of higher abundance andvisibility (Werner and Hall 1974). Optimalforaging has been naturally selected forbecause of the increased fitness of

individuals that maximize foragingefficiency (Pyke et al. 1977). Optimalforagingbehaviormayalsobeseen inantsastheyforageforfoodtobringbacktothecolony.Thisevolutionarybehaviorincreasesthe overall fitness of the colony, allowingforagers to retrieve a wide range of foodsvarying from plant seeds to other animalswhenresourcesarelimited(Cerda1997).Althoughantsaresmallinsize,theyhave

arelatively largeecological impactontheirsurrounding plant communities. They canshapethestructureandcompositionofsoiland plant communities through aeration,seed dispersal, and increased nutrientcontentinthesoilleadingtogreaterfertility(MacMahon et al. 2000). Ants are capableof performing these ecological functionsthrough their unique foraging behaviors.

CEC Research | https://doi.org/10.21973/N3PW9W Fall 2019 2/5

Many species of ants communicate whileforaging by using pheromones notifyingotherantsofa foodsource,enablingthemto move copious amounts of food fromsurroundingareastotheirnest.Black harvester ants (Veromessor

pergandei, hereafter referred to asVeromessor), commonly found throughoutthe Sonoran Desert, is a eusocialpolymorphic ant species. In other words,theyliveincoloniesandhaveindividualsofdrastically differentmorph sizes. However,despite this advantageous morphology,there are still questions regarding howtheseforagingantsarestrategicallyworkingtoobtaintheir food inordertoprovidefortheir colony (Traniello 1989). The forager’spurpose in a Veromessor colony is toprovide food to the queen, larvae, andother ants, making foraging efficiencycrucial. Food size, distance from the nest,and nutritional content is importantregarding optimal foraging theory becausetheremay be a preference of smaller andcloserfoodsduetolowerforagingcostsandhigher nutrient foods due to greatermetabolicrewards.In this study, we conducted an

experiment to test the effects of foragingcost and foodquality on foraging behaviorof Veromessor. We manipulated food sizeand distance from the ant nest as a proxyfor foraging cost and food type as a proxyfor nutritional quality most favorablenutritional content. In this case, foodscontaining higher fat and protein contentarereferredtoashighernutrientfoods.WehypothesizedthatVeromessorwouldprefersmaller food sizes, food closer to the antcolony, and food with the highestnutritionalcontent.

METHODS

2.1SamplingDesign.

ThisstudywasperformedintheSonoranDesert in San Diego County at theSteele/Burnand Anza-Borrego Reserve(33.2559oN,116.375oW) locatedinBorregoSprings, California. This reserve isdominatedbycreosotebush.Weobserved35 ant colonies in total. Veromessor arehighly abundant throughout the area. Thestudy was conducted from November 2-5,2019from9:30-11:30and14:30-16:30,themost active times of day for Veromessorcolonies.Food type treatments includedsmall and

large oat and pumpkin seed particles(hereafter referred to as seeds). Oatscontain1.4%fatand2.4%proteinwhereaspumpkin seeds contain 19% fat and 19%protein. For food size manipulation, a soilsievewasusedtoseparatesmalland largeseeds of both food treatments aftermanually crushing them into various sizes.Large oat seeds were greater than 2 mmandsmalloatseedswerebetween1mm-2mm. Large pumpkin seeds were greaterthan1mmwhilesmallpumpkinseedsweresmaller than .5 mm. Treatments weremeasured this way in order to control forforagingcostdue topumpkinseedshavingahigherdensitythanoatseeds.

2.2ExperimentalDesign

To manipulate distance, seeds wereplaced 1 meter (near) and 2 meters (far)awayfromtheantnest.Ateachcolony,halfa teaspoon of each treatment wasmeasuredandplacedinarandomizedorderat both distances, perpendicular to theforaging path (Figure 1). We placed the

CEC Research | https://doi.org/10.21973/N3PW9W Fall 2019 3/5

food treatments approximately threecentimetersawayfromeachothertoavoidany cross-contamination between antsgathering food. Each colony was observedfor 10 minutes immediately after all theseedswereplaced.Ahitwasrecordedeachtimeanantpickedupaseedtobringbacktothenest.

Figure 1. Placement of food treatments. Weconductedthisexperiment intheSonoranDesert inSan Diego County at the Steele/Burnand Anza-BorregoReservefromNovember2–5.Fourdifferenttreatments of food were placed at one meter andtwometers away from the colony perpendicular tothe Veromessor foraging path. Bolded circlesindicate the larger sized seeds. Shaded circlesindicatepumpkinseeds.Non-shadedcirclesindicateoatseeds.

Statistical analyses were conducted inJMP Statistical Software v14. A factorialANOVA model was used to determinewhich of the three factors (food size,distanceandtype)wasthemost importanton Veromessor foraging behaviors andwhetherornottherewereanyinteractionsbetweenthesefactors.Welog-transformedournumberofhitstonormalizeourdata.

RESULTS

OverallVeromessorantspreferredsmallerseedsoverlargerseedsforbothfoodtypes(N=35,t=6.26,p<0.0001,Figure2).Antscollected more seeds that were near thecoloniesthanfromfar(N=35,t=2.92,p=0.003, Figure 3). Ants preferred pumpkinseedsoveroatseeds (N=35, t=7.93,p<0.0001, Figure 4). There were nointeractions between the effects of foodsize,distance,ortype.

Figure 2. Number of seeds collected vs. seed size.Bars represent the average number of small andlarge seeds collected. For both food types, antscollectedagreaternumberofsmallseedsthanlargeseeds. Error bars represent + 1 SEM. (N = 35, t =6.26, p < 0.0001). Numbers of hits were log-transformedtonormalizedata.

CEC Research | https://doi.org/10.21973/N3PW9W Fall 2019 4/5

Figure 3. Number of seeds collected vs. distancefromcolony.Barsrepresenttheaveragenumberofnearand far seeds collected.Ants tookmore seedsfrom near treatments than from far treatments.Errorbars represent+1SEM. (N=35, t=2.92,p=0.0038). Numbers of hits were log-transformed tonormalizedata.

Figure4.Numberof seeds collectedvs. seed type.Barsrepresenttheaveragenumberofpumpkinandoatseedscollected.Ants tookmorepumpkinseedsthan oat seeds. Error bars represent + 1 SEM. (N =35, t=7.93,p<0.0001).Numbersofhitswere log-transformedtonormalizedata.

DISCUSSION

Ourstudysupportedthephenomenathatants exhibit optimal foraging behavior. Allvariablesmeasuredportrayed independenteffects on ant foraging behavior.Veromessor selected more small seedsoverall compared to large seeds, indicatingtheir preference for lower foraging cost.Carryingasmallerseedrequireslessenergyexpenditure than bringing a largerseedback to the colony (Crist andMacMahon 1992, Tomer 2012). Our seedsize results were consistent with paststudies in which a different species ofharvester ants, Pogonomyrmex, generallyfavoredsmallerseedsoverlargerseedsalsodue to their lower foraging cost (Pirk andCasenave2009).Veromessorantsalsoexemplifiedforaging

behavior through the distance and foodtype manipulations. There were steadypatterns seen with the selection of seedscollected closer to their nest as well agreater selection of pumpkin seeds.Pumpkin seeds have almost twenty timesmore protein and fat content than oatseeds, making them higher in nutrientsultimately leading to thehigherpreferenceshownbyVeromessor.A relatedstudywasconducted on Veromessor and found thatthey also preferred seeds with higherproteinandfatcontent(Gordon1980).Thispatternwas also seen in previous patternsobservedinPogonomyrmexsp.wheretherewere greater preferences for seeds withhigher nutritional content (Crist andMacMahon,1992).Throughout all trials, Veromessor

consistently exhibited strategies aligningwith the optimal foraging theory.Veromessoroptimized theirnutrient intakewhile also minimizing foraging costs by

CEC Research | https://doi.org/10.21973/N3PW9W Fall 2019 5/5

collecting smaller sized pumpkin seedscloser to their colony compared to anyother seed treatment. These optimalforagingpatternsofVeromessormaybeanindication of their level of fitness.Maintaining an abundance of substantialnutrients,suchasproteinorfats,providesalonger future for the colony. To furtherunderstandhowadaptationsofVeromessormightevolveovertime,afuturestudymaybe conducted using obstacles in theirforagingpath.Thismanipulationwillassesstheir learning curve according to thepredictions of foragers consistently aimingto optimize their efficiency. Investigatingthese foraging behaviors allows us toimproveourcomprehensionofVeromessorfitness and strengthen predictions of theoptimalforagingtheory.

ACKNOWLEDGMENTS

Special thanks to Tim Miller, KrikorAndonian, Reina Heinz, and Jim Dice. Thiswork was performed at the University ofCalifornia’s Steele/Burnand Anza BorregoDesert Research Center, doi:10.21973/N3Q94F.

REFERENCES

Cerda, X., Retana, J., & Cros, S. 1997. Thermaldisruption of transitive hierarchies inMediterranean ant communities. Journal ofAnimalEcology66(3):363–374.

Flanagan, T.P., Letendre, K., Burnside,W.R., Fricke,G.M.&Moses,M.E. 2012.Quantifying the effectof colony size and food distribution on harvesterant foraging. PLoS ONE 7(7):e39427.https://doi.org/10.1371/journal.pone.0039427

Gordon, S. 1980.Analysisof twelveSonoranDesertseed species preferred by the desert harvesterant.Madroño27:68–78.

Grüter, C., Czaczkes, T.J. & Ratnieks, F.L.W. 2011.Decision making in ant foragers (Lasius niger)facing conflicting private and social information.BehaviorEcologySociobiology65:141.

Heinrich,B. 1983.Dobumblebees forageoptimally,and does itmatter? Integrative and ComparativeBiology23(2):273–281.

Holder, K. & Polis, G.A. 1987. Optimal and central-placeforagingtheoryappliedtoadesertharvesterant, Pogonomyrmex californicus. Oecologia72:440–448.

Kretzer, J. & Cully, J. 2001. Prairie dog effects onharvester ant species diversity and density.JournalofRangeManagement54(1):11–14.

Morehead,S.A.&Jr.,D.H.F.1998.Foragingbehaviorand morphology: seed selection in the harvesterant genus, Pogonomyrmex. Oecologia 114:548–555.

MacMahon, J.,Mull, J.&Crist,T.O.2000.Harvesterants (Pogonomyrmex spp.): their community andecosystem influences. Annual Review of EcologyandSystematics31(1):265–291.

Pirk, J. andCasenave, J.L.D.2010. Influenceof seedsizeon feedingpreferencesanddiet compositionof three sympatric harvester ants in the centralMonte Desert, Argentina. Ecol. Res. 25(2):439–445.

Roubik,D.,Buchmann,S.,1984.NectarselectionbyMelipona and Apis mellifera (Hymenoptera:Apidae) and the ecology of nectar intake by beecoloniesinatropicalforest.Oecologia61:1–10.

Traniello, J.F.A. 1989. Foraging strategies of ants.Annu.Rev.Entomol.34:191–210.

Vittori, K., Talbot, G., Gautrais, J., Fourcassié, V.,Araujo, A.F.R., & Theraulaz, G. 2006. Pathefficiency of ant foraging trails in an artificialnetwork. Journal of Theoretical Biology 239:507–515.