Ken R. Helms.
Natural history of the ant Pheidole desertorum Wheeler in a desert grassland habitat.
Psyche 102:35-47, 1995.

Full text (searchable PDF, 976K)
Durable link: http://psyche.entclub.org/102/102-035.html

The following unprocessed text is extracted from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.

NATURAL HISTORY OF THE ANT PHEIDOLE DESERTORUM WHEELER IN A DESERT GRASSLAND HABITAT
Department of Zoology
Arizona State University
Tempe, AZ 85287-1501, U.S.A.*
Pheidole desertom colonies are large for the genus; the num- ber of adult workers in established colonies ranges from 2,460-24,814. They are nocturnal during summer and are predators and scavengers on arthropods. Both minor and major workers for- age, although foraging by majors appears to occur primarily when food is abundant. Young major workers may function as repletes. The adult majorlminor worker ratio varies greatly among colonies; much of that variance appears explained by colony size and matu- rity of adult colony reproductive broods. Most colonies produce reproductives each year and colony sex ratios are extremely sex- biased. Mating season begins following summer rainfall. Males and gynes fly prior to sunrise; males form aerial swarms which gynes enter. Mating occurs on the ground, then gynes fly away, presumably to suitable colony founding sites. Colony foundation is normally haplometrotic, although pleometrotic queen associations with workers are found.
Ants in the genus Pheidole occur nearly world-wide and are common in such diverse habitats as deserts and tropical forests, There may be as many species in Pheidole as occur in any other ant genus (Wilson, 1976; Holldobler and Wilson, 1990). However, little is known about the natural history of most species. *Correspondence: E-Mail: agkrh @ ASUVM.inre.asu.edu Manuscript received 21 February 1995.




================================================================================

36 Psyche [vo~. 102
Pheidole desertorum occurs in northern Mexico and the south- western United States (Arizona, Nevada, New Mexico, Oklahoma, and Texas [Krombein 1979, Wheeler and Wheeler 19861). There are notes on the natural history of P. desertorum (Cole, 1937; Creighton, 1950; Krombein, 1979; Wheeler and Wheeler, 1986) and the species has been included in the desert community ecology studies of Davidson (1977a,b). Their nesting biology and impor- tance as prey for (and defense against) Neivamyrmex army ants have been studied extensively (Mirenda, Eakins, Gravelle, and Topoff, 1980; Droual and Topoff, 198 1 ; Droual, 1984). I studied a Chihuahua desert population of P. desertorum over a five-year period and have developed an account of major aspects of their natural history.
The study was conducted from 1988 to 1992, primarily on a 0.8 km2 site in the San Simon Valley, near Rodeo, New Mexico. The area is desert grassland (Mirenda, et al. 1980), although shrubs are common, primarily Mormon tea (Ephedra trifurca), snakeweed (Gutierrezia sp.), and burroweed (Haplopappus sp.). Pheidole desertorum colonies are nocturnal during summer and were located at night with peanut butter baits set at 3.5m intervals along randomly selected transects and at 3.5m equidistant intervals within 105 X 105m grids (100 baits per grid) (Helms, 1994). Most colonies were left intact and monitored in the field. Colony repro- ductive status was assessed as described by Helms and Rissing (1990) and Helms (1994). A subset of colonies was collected with traps that exploit this species' nest evacuation response to attack by Neivamyrmex army ants (Helms and Rissing, 1990). After col- lection, female alates (referred to here as gynes) and males were removed and counted. The remaining workers and brood of each colony were divided into 20 groups of equal weight. The workers and brood in each group were counted until the cumulative mean number of major and minor workers and brood changed by 5% or less over two consecutive groups (i.e., until the mean numbers of individuals per group stabilized). Total number of majors, minors, and brood within each colony was then estimated by multiplying the cumulative mean numbers by 20. Samples of adults and pupae of each caste were dried at 50å¡ and body mass measured (å 0.01



================================================================================

199.51 Helms 37
mg). The headwidths of additional samples of major and minor workers were measured above the eyes (å 0.01 mm). RESULTS AND DISCUSSION
General Description
As in most members of the genus, P. desertorum has a dimor- phic worker caste consisting of large (major) workers (or soldiers) and smaller (minor) workers (Tables 1, 2). However, the appear- ance of workers is atypical for the genus. As noted by Wheeler and Wheeler (1986), major workers of P. desertorurn are large com- pared to those of most species in the genus, but their heads are not as greatly enlarged relative to the body. Minor workers are also large for Pheidole and have unusually long legs and antennae. Wheeler and Wheeler (1986) described major workers as appearing superficially similar to Pogonornyrmex and minor workers as resembling Aphaenogaster.
Colony Locations, Size, and Density
Pheidole desertorurn nests in the soil; nests were most often found at the base of shrubs and clumps of grasses. However, nest site location varies according to habitat. Not far from the study site, in the foothills and valleys of the Chiricahua Mountains of Arizona, nests are often located under large stones, which are uncommon on the grassland site. Nests under stones are also reported from other Arizona locations (Cole, 1937) and from Nevada (Wheeler and Wheeler, 1986).
Colonies have multiple nests (Droual and Topoff, 1981), up to seven in my study. In general, only one nest is occupied at any one time, although emigrations between nests are common, particularly when the soil is moist (Droual and Topoff, 1981). Maintenance of multiple nests appears important in providing refuge when the colony is attacked by Neivamyrmex army ants (primarily N. nigrescens) (Mirenda, et al. 1980; Droual and Topoff, 1981; Droual, 1984).
Consistent with earlier work, N. nigrescens appeared to be an important predator on P. desertorum during this study. Raids on P. desertorum (and other Pheidole) were observed in 1991 and 1992;



================================================================================

Table 1. Worker body size in Pheidole desertorurn measured by headwidth above the eyes (H.W.) in millimeters. Colony Size
In Total No. Mean H.W. Mean H.W.
of Adult of Major of Minor
Colony Workers Workers (S.D.)^ Limits Nc Workers (S.D.)". Limits Nc - - - - - - -- - - -
a S.D. = standard deviation.
Major and minor worker headwidths were tested for deviation from normal distributions in each of the five colonies and found not significantly different (P > 0.2 in all cases, Kolmogorov-Smirnov goodness of fit tests). Number of individuals measured per colony. Reproductively immature colony.




================================================================================

Table 2. Pupal and adult dry weight (mg) and standard deviation (S.D.) in Pheidole desertorurn. Values are means across colonies, i.e., the sum of the mean of each colony divided by the number of colonies. Adult (post- pupal) individuals are a representative sample from colonies; they include younger (callow) as well as older indi- viduals in the approximate frequency of their occurrence. Pupa Dry Adult Dry Mean % Adult Increase Weight (S.D.)" Nb Weight (S.D.)" Nb T v d Over Pupal Weighte Gynes
(N=9^ 2.81 (0.52) 23.9 7.04 (1.73) 27.2 0 <0.005 150.5 Males
(N=9') 1.12(0.18) 34.4 1.16 (0.15) 40.0 9 NS 3.6 Major Workers
(N= 1 5f) 0.97 (0.36) 38.1 2.1 1 (0.68) 64.9 1 <0.005 117.5 Minor Workers
(N=20f) 0.23 (0.04) 93.4 0.27 (0.05) 97.5 34 <0.005 17.4 a Standard deviation of the mean values among colonies. Mean number of individuals measured per colony. Test statistic from Wilcoxon's signed-ranks test for paired comparisons (Sokal and Rohlf, 1981). Probability that adult dry weight is greater than pupa dry weight (tests are one-tailed). NS = P > 0.05. [(mean adult weight - mean pupal weight) 1 mean pupal weight] X 100. Number of colonies from which individuals were measured.



================================================================================

40 Psyche [vo~. 102
N. nigrescens were sometimes found bivouacked within P. deserto- rum nests.
Colony density is high on the grassland site, and colonies can be much larger than proposed by Creighton (1950). Fifty-six and 92 P. desertorurn colonies were found within the two 105 X 105m grids. Mean number of adult workers per colony censused was 9,881 (Table 3).
Territoriality
Of those colonies located along transects, I assessed 61 for internest movement at least once per week over a one-month period (June 8-July 8, 1989); 37 emigrated between nests at least once. Those colonies moved an average of 2.90m (S.D. = 2.10m) from their original nest location. Average distance from the origi- nal nest location to that of their nearest known conspecific neigh- bor along transects was 14.59m (S.D. = 3.34m). Of those colonies which emigrated more than once, the last known move was toward or back into the original nest in 12 of 21 cases (also see Droual and Topoff, 198 1).
Nests of colonies are clumped away from their neighbors, con- sistent with the hypothesis that colonies actively avoid one another or with the hypothesis that foundress nests rarely survive if they are located near established nests.
Workers from neighboring colonies fight when they encounter one-another at baits. Fights often result in death, and fighting per- sists until one colony dominates the bait. Fighting at baits was also observed between P. desertorurn and other ant species, primarily Pheidole hyatti, Pheidole xerophila, and Solenopsis sp. Pheidole desertorurn nearly always displaced other Pheidole; however, P. desertorurn were often displaced when Solenopsis sp. was present. Foraging Behavior and Diet
Only minor workers leave the nest at the start of foraging activ- ity (shortly after sunset). However, major workers leave the nest and come to baits; their foraging role may be associated with exploiting clumped or otherwise abundant food sources. Workers forage to baits in diffuse columns, with majors and minors return- ing to the nest with food. However, not all majors are moving in the column or collecting food; nonforaging majors may be defend-



================================================================================

19951 Helms 41
Table 3. Census data from Pheidole desertorum colonies. Numbers in parenthe- ses are the number of pupae, other numbers are the number of adults. Number of Number of Number of Number of Collection Colony Minor Workers Major Workers Gynes Males Date 'Pupae not censused.
ing the column and the food source. These different foraging roles of P. desertorum majors and minors are similar to those found for Pheidole pallidula and Pheidole teneriffana (Szlep-Fessel, 1970). Pheidole desertorum are predators and scavengers at the grass- land site. Workers return to nests with a variety of arthropods (and parts). Those recognized were grasshoppers, beetles, termites, and bugs primarily. With the arrival of the first rains of summer, alate termites are captured in great numbers and stored in nests with the wings removed. Alates (primarily males) of other ants are scav- enged during the rainy season. During spring and late fall (dry, apparently food-limited periods), it is common to find exoskeleton parts of workers of other ants (often recognized as Pogonornyrmex) within nests.
Pheidole desertorurn from the same general location as this study were identified as seed-harvesters by Davidson (1 977a,b). However, I never observed workers carrying seeds, and seeds were found rarely in nests and only in small numbers. In June 1989, I presented four colonies with cracked wheat, rolled oats, and whole seeds of Bermuda grass and Kentucky blue grass; such seeds are harvested readily by desert seed-eating ants and are small enough to be transported by P. desertorurn workers (Taylor, 1977; Rissing, 198 1 ; Rissing, personal communication). Pheidole desertorum fre- quently encountered the seeds but never collected them.



================================================================================

42 Psyche [VOI. 102
Food Storage
Long-term food storage in P. desertorum may be accomplished by replete workers. Those workers are majors with distended (approximately 1.5 to twice normal size) gasters, whose behavior is sluggish. The gaster of these workers holds a golden-colored, oily substance enclosed in a structure separate from the crop and the rest of the alimentary tract (Cover and Helms, unpublished). In Pheidole pallidula, major workers with distended gasters store lipids whose energy is somehow transferred to nestmates (Lachaud, Passera, Grimal, Detrain, and Beugnon, 1992). In P. desertorum, major workers with distended gasters are young and not pigmented fully. Since major workers more than double their dry body weight following pupation (Table 2), it is possible that young major workers serve in energy storage, with that role diminishing as energy is transferred to nestmates later in the majors' lifespan. Young minor workers can also have distended gasters, and the dry body weight of minor workers also increases following pupation, but to a lesser extent than majors (Table 2). The Frequency of Major and Minor Workers The percent of adult workers that were majors varied greatly among colonies (9.0 to 23.6%; Table 3). Two colonies were small and presumably young; one produced no reproductives and the other produced only three gynes (Colonies A9, A10; Table 3). The percent of adult workers that were majors was less in these colonies (9.0 and 12.6%) than in the remaining mature colonies. However, there was also substantial variation among mature colonies; the percent of adult workers that were majors ranged from 13.1-23.6% (Table 3). This variation is associated with dif- ference among colonies in age of their reproductive broods. Increasing maturity of reproductive broods is correlated with decreasing adult niajor/minor ratio (Fig. 1A) and increasing major brood production (Fig. 1B).
It appears that production of sexual broods may limit production of major workers, and the adult majorlminor ratio decreases with increasing duration of reproductive effort. Major worker mortality likely contributes to this decline. However, as a reproductive brood becomes increasingly mature, the production of majors increases, and the majodminor ratio is presumably restored. Similar effects



================================================================================

19951 Helms
10 - o-\-*ÌÔiÌÔÌÔiÌÔÌ
0 25 50 75 100 0 25 50 75 100
Percent of Reproductive Brood (Alate + Pupae) That Are Alate Fig. 1: Percent of adult and pupal workers that are major workers as a function of the maturity of Pheidole desertorum reproductive broods. A. Kendall's coeffi- cient of rank correlation = -1.0, P = 0.05. B. Kendall's coefficient of rank correla- tion = 0.8, P = 0.10. Data are from colonies A4, Xl, X3, A8, and A6 in Table 3. of reproduction on the majorlminor ratio have been found for Phei- dole dentata (Johnston and Wilson, 1985). Reproduction
Most colonies in the population produced sexuals each year. In 1991, 255 of 287 colonies had alates. The time of year at which gyne and male eggs are produced is not known; gyne and male pupae can be found in nests by mid-May. Most sexual forms are pigmented fully and mature by early to mid-July. Colonies can also produce a second reproductive brood. Reproduction was assessed in late July and early August 1991 in 25 colonies with a mature reproductive brood in early July; 10 of these colonies had an addi- tional reproductive brood consisting primarily of individuals not pigmented fully (callows) and/or pupae. Worker pupae were observed throughout the period when reproductive brood occurred. Pheidole desertorum gynes are much larger than males (Table 2). Gynes and males also differ greatly in temporal changes in body mass. Body mass of males remains essentially unchanged fol- lowing pupation, while gynes experience a dramatic increase (Table 2). Mean dry weight of 60 gynes (30 from each of two colonies) collected as they emerged on the surface to fly was 8.75mg; given a mean gyne pupa dry weight of 2.8 1mg (Table 2), gynes more than triple their pupal dry weight before flight. This



================================================================================

44 Psyche [vo~. 102
substantial temporal increase in gyne body weight may be the rule in ant species with independent colony foundation (see, e.g., Keller and Passera, 1989; Tschinkel, 1993).
The Sex Ratio
The sex ratio of reproductive broods is extremely biased; 145 of 146 colonies whose sex ratio was assessed where at least 50 repro- ductive~ were observed produced either only gynes, only males, or a sex ratio extremely male-biased (on average, > 97% males) (Helms and Rissing, 1990; Helms, 1994; also see Droual, 1982). Such dimorphism among colonies in the sex ratio also occurs in Pheidole xerophila tucsonica (Helms and Rissing, 1990), Pheidole hyatti (Helms, unpublished data), and perhaps Pheidole dentata (Johnston and Wilson, 1985).
The frequency of gyne-specialist and male-specialist colonies found within the P. desertorurn population was very near equal (171 gyne-specialist and 177 male-specialist colonies; Helms, 1994). In addition, 144 of 151 colonies specialized in the same sex over each consecutive year during a two to five year period (Helms, 1994).
Mating Flights
Mating season begins following the first substantial rainfall in July. Alates begin emerging on the surface one and one-half to two hours before sunrise. They begin to fly from the nest perimeter approximately 45 minutes prior to sunrise until just before sunrise. Alates often climb blades of grass, stones, or other objects before taking flight. Flights generally occur when there is substantial soil moisture, it is warm, and there is no breeze. Conditions in which mating flights were observed are summarized in Table 4. After leaving the nest, males form aerial groups (swarms) which hover approximately 2-5m above the ground; swarms occur simul- taneously at various locations throughout the site. No distinguish- able topographic feature was associated with swarm location. Observed swarms were not large; five to approximately 50 males were observed. Swarms did not persist beyond one-half hour past sunrise. Gynes fly into swarms and hover with males; they often left those swarms without mating. No more than two gynes occurred in a swarm at any one time. In three cases observed, there



================================================================================

1995 J Helms 45
Table 4. Environmental conditions at one hour prior to sunrise when Pheidole desertorum mating flights were monitored in July 1991. Date
Precipitation
In 24 hours
Prior (mm)
Depth of Soil
Moisture (cm)
Temperature
(OC)
Estimated
Wind Speeda
Mating
Flights
Observed
No
Yes
No
No
Yes
Yes
No
Yes
Yes
a In miles per hour.
was physical contact between a gyne and male, and they dropped immediately to the ground where mating occurred. One mating lasted five minutes, 23 seconds. In two of three matings observed, the gyne appeared to terminate mating by biting the male on the gaster; all three gynes flew away after mating. One gyne momen- tarily reentered the same swarm prior to departing. The flight of gynes following mating could be common in desert Pheidole. While it does not occur in Pheidole sitarches (Wilson,
1957), it occurs in P. desertorum, P. hyatti (Helms, unpublished data), and P. xerophila tucsonica (Johnson, personal communication). Pheidole gynes could mate with multiple males. Allozyme data show that multiple mating might occur in P. deser- torum, but if so, it appears rare (Helms, 1994). Therefore, flight after mating may occur primarily for dispersal to colony founding sites.
Colony Foundation
Pheidole desertorum foundress nests were observed after initia- tion of the mating season and were discovered only under large piles of old cow dung (approximately 25cm or more in diameter). However, incipient colonies were rarely found, and given the loca- tions of adult nests, most queens on the site may attempt colony



================================================================================

46 Psyche [vo~. 102
foundation at the base of shrubs and clumps of grasses. Foundress nests under cow dung may indicate that colony foundation will occur under stones in areas where they are present. Foundress queens of P. desertorurn sometimes join others within a single founding nest; of 12 foundress nests discovered in 1991, nine had one queen, two had two queens, and one had three queens. Minim workers were present in two of the pleometrotic groups. Pleometrotic groups were collected and kept in the labora- tory; only single queens remained alive after two months following collection. However, allozyme data suggest multiple reproductive queens might sometimes occur within a single adult nest in the field (Helms, 1994). Pleometrotic groups in nature may sometimes persist into the colony reproductive stage. Thanks to J. Alcock, J.P. Collins, S.H. Faeth, J.H. Fewell, M.C. Moore, S.W. Rissing, and the Social Insect Research Group at Ari- zona State University for discussion and constructive comments on earlier versions of the manuscript. S.W. Rissing suggested the method of colony census and encouraged me to write-up the nat- ural history. I also thank the staff and volunteers of the Southwest- ern Research Station of the American Museum of Natural History for hospitality and assistance, S. Cover for locating foundress nests and sharing his knowledge on Pheidole and their mating flights, and H. Topoff for sharing his knowledge on army ants. The research was supported financially by the Department of Zoology and the Graduate Student Association Research Development Pro- gram (Arizona State University), the Theodore Roosevelt Memor- ial and Southwestern Research Station Student Funds (American Museum of Natural History), and the National Science Foundation (Dissertation Improvement Grant BSR 9001309 and Grant BSR 89063 19, S.W. Rissing, Principal Investigator). Cole, A.C., Jr. 1937. An annotated list of the ants of Arizona. Ent. News 48:97-101.
Creighton, W.S. 1950. The ants of North America. Bull. Mus. Comp. Zool. 104: 1-585.
Davidson, D.W. 1977a. Species diversity and community organization in desert seed-eating ants. Ecology 58:711-724.




================================================================================

19951 Helms 47
. 1977b. Foraging ecology and community organization in desert seed-eat- ing ants. Ecology 58:725-737.
Droual, R. 1982. Sex ratios in the ant Pheidole desertorurn. Am. Zool. 22:971. . 1984. Anti-predator behaviour in the ant Pheidole desertorum: the impor- tance of multiple nests. Anim. Behav. 32: 1054-1058. Droual, R., and H. Topoff. 1981. The emigration behavior of two species of the genus Pheidole (Formicidae: Myrmicinae). Psyche 88: 135-150. Helms, K.R. 1994. Colony and population sex ratio evolution in the ant Pheidole desertorum Wheeler. Ph.D. Diss., Arizona State University, Tempe. Helms, K.R., and S.W. Rissing. 1990. Single sex alate production by colonies of Pheidole desertorurn and Pheidole xerophila tucsonica (Hymenoptera: Formici- dae). Psyche 97:213-216.
Holldobler. B., and E.O. Wilson. 1990. The Ants. Harvard University Press, Cam- bridge, MA.
Johnston, A.B., and E.O. Wilson. 1985. Correlates of variation in the majorlminor ratio of the ant Pheidole dentata (Hymenoptera: Formicidae). Ann. Entomol. Soc. Am. 78:8-11.
Keller, L., and L. Passera. 1989. Size and fat content of gynes in relation to the mode of colony founding in ants (Hymenoptera; Formicidae). Oecologia 80:236-240.
Krombein, K.V. 1979. Catalog of the Hymenoptera in America North of Mexico. Smithsonian Institution Press, Washington, D.C. Lachaud, J-P., L. Passera, A. Grimal, C. Detrain, and G. Beugnon. 1992. Lipid stor- age by major workers and starvation resistance in the ant Pheidole pallidula (Hymenoptera, Formicidae). p. 153-160 In J. Billen, ed., Biology and Evolution