Testing Central Place Foraging Theory in the Ant Genus Pogonomyrmex

1

Testing Central Place Foraging Theory in the Ant Genus Pogonomyrmex:

a review of the literature from 1977through 2008

A.B. Mayo

Animal Population Ecology

Dr. Miriam Benabib

December 9, 2009

INTRODUCTION

Foraging strategy is one of the most important and defining traits for any animal species because it relates obtaining energy to ecological situation. Thus, foraging behaviors have a direct impact on fitness and selection is expected to be fairly strong resulting in optimal patterns (Schoener 1971; Pyke 1984). Although morphological and physiological characteristics also play significant roles in foraging, optimal foraging theory focuses on behavioral components which are related to fitness through a currency and involve a time component (Pyke 1984). Optimal foraging theory attempts to predict the optimal diet based on maximizing benefit to cost of the currency. (See Schoener 1971; Pyke, Pulliam and Charnov 1977; Pyke 1984; Schoener 1983).

The originaloptimal foraging models predicted diet choice for animals that forage individually and eat basically where they capture or find their prey. Problems in using these models quickly become apparent when applied to social insects which forage from a central place to which they return with the forage and eat communally. Central place foraging theory was developed to account for costs associated with the distance to and from the food resource, additional handling of prey, and transporting the prey (Orians and Pearson 1979; Schoener 1979; Olsson, Brown, and Helf 2008). Optimal foraging models usually assume that, in the long-term, net energy gain is maximized, that animals have perfect knowledge of patches, and that animals have simultaneousdetection of available prey. However, some social insects maximize their net energy efficiency(Houston, Schmid-Hempel, Kacelnik 1988) and ants probably detect prey primarily by physical encounter, not necessarily taking the first item. Thus, ants make foraging decisions based on encounter rates through sequential assessment of forage (Davidson 1978).

Models of ant foraging are further complicated by eusociality, a variety of foraging strategies,possible variation at the individual forager as well as colony level,and that ants cache forage (Traniello 1989). Besides variation in behavior, some ant species exhibit worker polymorphism which may result in size matching between ant morphology and forage (Traniello 1989; Morehead and Feener 1998). Ant foraging strategies also display a continuum from individual forms (in which foragers leave the nest in random directions) to collective forms mediated by pheromones (for instance, ants that follow a trunk trail to a foraging area). Keeping these concerns in mind, this review will focus on the application of central place foraging theory to seed harvesting ants in the genus Pogonomyrmex.

METHODS

The general foraging literature was searched for reviews, explanations, and tests of optimal and central place foraging with an emphasis on foraging in ants for background on the theories in general. The literate included in this review was restricted specifically to field based tests of optimal and central place foraging in the seed harvesting ant genus Pogonomyrmex (Hymenoptera: Formicidae: Myrmicinae) and includes papers from 1977 through 2008.

LITERATURE REVIEW

Background

Seed harvesting ants of the genus Pogonomyrmex, are very similar in morphology and ecological niche. Workers are monomorphic and worker castes typically differ by age with foragers being oldest (Gordon 1999). Though Pogonomyrmex ants live in a diversity of habitats, many of these ants are prominent species in arid habitats which are severely seed limited(Bernstein 1974). For studied species, the diet is 60-100% seeds and when Pogonomyrmex species co-occur, there can be intense competition for food resources. These ants also forage on frass, vegetative and floral parts of plants, and animal parts. However, consideration of these non-seed items has been excluded from tests of optimal foraging probably because they are a small portion of the diet and animal parts may be foraged for proteins and not carbohydrates (energy), which may complicate the choice of currency in the model. A more detailed understanding of ant diet is needed (Kay 2002).

Despite the importance of foraging to colony structure and therefore, its impact on social organization and fitness, there are few studies that specifically test central place foraging in Pogonomyrmex ants. The studies done focus on different assumptions, use different techniques or variable ranges, and assume that all Pogonomyrmex species are similar enough that the findings from one study apply to other species. Of the 40+ species in the genus, those studied include P. barbatus, P. californicus, P. desertorum, P. maricopa, P. occidentalis, and P. rugosus, all of which are arid habitat dwellers. Despite morphological and niche similarities, these species show variability in foraging strategy (trunk trail foraging, individual foraging, or mixed strategies) which alters the basic concerns of the theory such as distance, energy and time costs on the individual forager level, among colonies, and among Pogonomyrmex species. Because several of these species co-occur, a consideration of differences in foraging strategies among speciesmay help explain how they co-exist (Whitford 1978; Morehead and Feener 1998).

For testing central place foraging models in Pogonomyrmex, a very basic model for optimal foraging is:

with e, the net energy gain from the forage item; B, the benefit (gross) of the item foraged; and C, the cost of the foraging the item; and the subscript i, designating the particular item. The costs of foraging include energy expended in travel to the patch (, in search within the patch (), in handling the forage (), and in the return travel carrying the forage () (based on Taylor 1977, 1978 and other articles in this review). This model assumes that energy (ei) is the currency. If the ants are foraging optimally, then they will select seeds that maximize the net energy intake such that, as the costs change, so do the seeds preferred. The tests which are reviewed here were field manipulations in which different variables were altered and the resulting choice of forage, rate change, or efficiencywere assessed.

Summary of Results

From the empirical tests of optimal and central place foraging in the Pogonomyrmex ants, certain patterns are shown across the species studied. While generally, the foraging patterns fit central place foraging models, some important differenceswere found (Bailey and Polis 1987). It is lamentable that, as Taylor pointed out, many of the parameters that are needed to fully test central place foraging models in ants are very difficult to obtain (Taylor 1977, 1978). These missing measurements were assumed constant or assumed controlled for across colonies. The primary concern in these studies has been to establish the currency, the important time costs, how forage items are ranked, and how these concerns vary among species and might influence co-existence.

Although foraging theories focus on behavioral traits, behaviors are shaped and often constrained by morphological features. Since Pogonomyrmex ants carry seeds in their mandibles and these species vary in size, several morphological features of the head, thorax, and leg were tested for their importance in reducing the cost of foraging (Morehead and Feener 1998). Noneof the morphological traits had a significant impact on seed selection. This test also refutes the idea that there might be seed to body size matching.

The currency for foraging in these ants is the net energy gain (Taylor 1977; Bailey and Polis 1987; Fewell 1988; Weier and Feener 1995; Morehead and Feener 1998). Taylor (1977) was able to rule out minimization models and Fewell (1988) found that P. occidentalis preferred low plant cover trails which yielded the same net energy gain as more dense cover trails, even though the use of low cover trails was less energy efficient. The cost of foraging is nearly negligible (Weier and Feener 1995) with an estimate forP. occidentalis of 0.1% of the energy from the forage (Fewell 1988). Weier and Feener (1995)found that, despite the employment of different foraging strategies by P. rugosus and P. maricopa, energy gain per trip, about 100 times the cost of these trips, did not differ significantly. This result led Weier and Feener tosuggest that the cost of foraging could be “safely ignored in future studies.”

In earlier studies, handling and return time had been assumed constant but somePogonomyrmex ants do remove various extra-floral and other seed parts (Whitford 1978). Contrary to these studies, Bailey and Polis (1987) found that both handling time, including travel time back to the nest, and success rate of securing a seed increased with seed size.

From Pyke’s review (1977), Pogonomyrmex ants are expected to rank seed types based on expectations of energy intake for each seed type (Davidson 1978). The ants did optimize net energy gain in terms of seed size and ranked seeds consistently according to size with respect to changes in distance from the nest, abundance of seeds, and size of the seed patch (Taylor 1977). While Taylor (1977) and Davidson (1978) found evidence to support this ranking of seeds by size, the ants did not narrow the size range of seeds as distance increased in Taylor’s work but did in Davidson’s study, though the ants never took the smallest seeds offered. Davidson suggested that Taylor did not use a great enough distance to show this effect. However, Davidson did find partial preferences which also challenges optimal foraging. The partial preferences probably reflect individual forager variation as well as the need for foragers to continually encounter seeds in order to consistently rank them. Continual encounter cannot be assumed for any forager since individual foragers may make few trips in any one day. This brings us to a particular concern of ants: recruitment.

Recruitment to a patch may be achieved by an increase in the total number of ants recruited to that area or an increase in the rate of a particular forager’s return to that patch. The combination of these two forms of recruitment across patchesdetermine the colony foraging rate. Taylor (1977) was able to demonstrate several aspects of recruitment including that the effect of an increase in temperature on an increase in running speed allows these ants to maintain recruitment (multiple trips per patch) by a decrease of the total ants recruited to a patch. He also demonstrated the important effect of interference among recruits to a patch, that the net rate of recruitment increased with increasing seed density in a patch. Since recruitment can mean more ants in an area, the increased presence of foragers can physically interfere with an individual ant’s search and return time. Thus, the optimum recruitment to any given patch is an intermediate level (Taylor 1978). Taylor’s model explains the different levels of response to variable changes as adjustments to maintain net energy gain rate and net recruitment rate.

Finally, because hunger state and temperature impact physiology and indirectly affect behavior, these parameters have been neglected in foraging models. However, they are important determinants of foraging because ants are social and cache seeds. In these studies, no attempt was made to assess the hunger state of the colonies studied. Temperature ranges for activity periods was measured but not considered as an effect on foraging decisions. Temperature, however, was found to have an important impact on metabolism and running speed with increases in temperature increasing both metabolism and running speed (Fewell 1988; Morehead and Feener 1998). Temperature is also an important constraint on when and how much time is available for foraging which also affects the ranking of seeds and recruitment rates (Taylor 1977; Morehead and Feener 1998). Despite these known affects, most studies ignore temperature in producing optimal foraging models.

CONCLUSIONS

The overall picture resulting from this empirical work is that Pogonomyrmex ants are maximizing their net energy gain and generally fit the assumptions for central place foraging models but also that these models need to be modified to accommodate details of time parameters such as handling time, differences in foraging strategy, and the variation among individual foragers. In addition, temperature and hunger state effects need to be incorporated into the model and a fuller account of diet needs to be made. The main departures from central place foraging models appear to be primarily related to the low cost of foraging and the complex pattern of recruitment.

Differences in foraging strategy and seed selection may have more to do with competition with other seed harvesting ants than with maximizing foraging (Acosta et al. 1995). While Morehead and Feener (1998) suggested that the variation they observed due to differences in foraging strategy needs to be accommodated in central place foraging, in an earlier paper Weier and Feener (1995)found no significant differences in net energy gain rates between P. rugosus and P. maricopadespite differences in their foraging strategies. Whitford (1978) found seed selection differences related to foraging strategy differences among co-occurring Pogonomyrmex species which allowed these species to reduce competitive encounters. Other studies demonstrate that some Pogonomyrmex species modify their preferred seeds in response to presence of other Pogonomyrmex ants (Hanson 1978; Davidson 1985; Cook and Mayo in prep.). Thus, Pogonomyrmex ants appear to be able to employ several behavioral modifications to maintain their net energy intake gain. This ability probably plays a significant role in their dominance in arid habitats. What this might mean for other species in other habitats remains to be investigated.

Implications for Pogonomyrmex comanche

Pogonomyrmex comanche like other Pogonomyrmex ants has a diet primarily of seeds. My data for P. comanche suggest that their diet is 70-85% seeds. However, this species lives in very sandy soils in prairie habitats which are probably less severely seed limited than more arid habitats. It is rare for other Pogonomyrmex ants to co-occur with P. comanche because of soil preferences among the species. Little is known about P. comanche and to my knowledge nothing has been published. In order to test central place foraging models on P. comanche, the diet and foraging strategy needs to be established, including foraging rates, ranking of seeds, distances foraged, colony interactions in defining forage areas, and temperature effects. So far, I have met with mixed results in trying to establish any of these parameters. I suspect that P. comanche is an opportunist in forage selection and that the habitat they are in has much to do with foraging strategies and the apparent differences from the species so far studied.

P. comanche with a Froelichia floridana seed (the fluffy white; commonly called snake cotton), Fort Worth Wildlife Refuge and Nature Center, Fort Worth, Texas

LITERATURE CITED

Acosta, FJ, Lopez F, Serrano JM. 1995. Dispersed versus central-place foraging: intra- and intercolonial competition in the strategy of trunk trail arrangement of a harvester ant. The American Naturalist 145(3): 389-411.

Bailey KH, Polis GA. 1987. Optimal and central-place foraging theory applied to a desert harvester ant, Pogonomyrmex californicus. Oecologia 72: 440-448.

Bernstein RA. 1974. Seasonal food abundance and foraging activity in some desert ants. The American Naturalist 108(962): 490-498.

Cook JL, Mayo AB. (in prep.) Resource partitioning between sympatric seed-harvester ants in the genus Pogonomyrmex (Hymenoptera: Formicidae).

Davidson DW. 1978. Experimental tests of the optimal diet in two social insects. Behavioral Ecology and Sociobiology 4: 35-41.

-- 1985. An experimental study of diffuse competition in harvester ants. The American Naturalist 125: 500-506.

Fewell JH. 1988. Energetic and time costs of foraging in harvester ants, Pogonomyrmex occidentalis. Behavioral Ecology and Sociobiology 22: 401-408.

Gordon DM. 1999.Ants at Work. New York: The Free Press. 182 pp.

Hanson SR. 1978. Resource utilization and coexistence of three species of Pogonomyrmex ants in an upper Sonoran grassland community. Oecologia 35: 109-117.

Houston A, Schmidt-Hempel P, Kacelnik A. 1988. Foraging strategy, worker mortality, and the growth of the colony in social insects. The American Naturalist 113(10): 107-114.

Kay A. 2002. Applying optimal foraging theory to assess nutrient availability ratios in ants. Ecology 83(7): 1935-1944.

Morehead SA, Feener DH. 1998. Foraging behavior and morphology: seed selection in the harvester ant genus Pogonomyrmex. Oecologia 114: 548-555.

Olsson O, Brown JS, Helf KL. 2008. A guide to central place effects in foraging. Theoretical Population Biology. 74: 22-33.

Orians GH, Pearson NE. 1979. Chapter 6: On the theory of central place foraging, p. 157-177. In: Horn DJ, Stairs GR and Mitchell RD, eds. 1979. eds. Analysis of Ecological Systems. Columbus, OH: Ohio State University Press. 321 pp.

Pyke GH. 1984. Optimal foraging theory: a critical review. Annual Review of Ecology and Systematics 15: 523 – 575.

Pyke GH, Pulliam, Charnov EL. 1977. Optimal foraging: a selective review of theory and tests. The Quarterly Review of Biology 52(2): 137-154.

Schoener TW. 1971. Theory of feeding strategies. Annual Review of Ecology and Systematics 2: 369-404.

--1979. Generality of the size-distance relation in models of optimal feeding. The American Naturalist 114(6): 902-914.

--1983. Simple models of optimal feeding-territory size: a reconciliation. The American Naturalist 121(5): 608-629.

Taylor F. 1977. Foraging behavior in ants: experiments with two species of Myrmecine ants. Behavioral Ecology and Sociobiology 2: 147-167.

--1978. Foraging behavior of ants: theoretical considerations. Journal of Theoretical Biology 71: 541-565.

Traniello JFA. 1989. Foraging strategies in ants. Annual Review of Entomology 34: 191-210.

Weier JA, Feener DH. 1995. Foraging in the seed-harvest ant genus Pogonomyrmex: are energy costs important? Behavioral Ecology and Sociobiology 36:291-300.

Whitford WG. 1978. Foraging in seed-harvester ants Pogonomyrmex spp. Ecology 59(1): 185- 189.