Presented March 23, 1996

Proceedings of a Second Conference on Quail Management*

 Lawrence E. Gilbert Director,

Brackenridge Field Laboratory University of Texas, Austin, Texas

Abstract:

Because imported fire ants, Solenopsis invicta, arrived in the U.S. free of the specific parasites and pathogens which infect them in their native South America, they have enjoyed a dramatic competitive advantage over indigenous ants as they spread through southern North America.

In Brazil these ants exist as low-density single-queen colonies and are not generally viewed as pests. Up to five species of Pseudacteon phorid fly parasitoids can be observed attacking S. invicta in any given location in their homeland. A similar situation holds for native fire ants, S. geminata, in Texas, where two host-specific Pseudacteon attack them but not their invading cousin. Like S. invicta, S. geminata becomes a pest when introduced to other regions free of biological controls.

Experiments on the interaction of Pseudacteon and their Solenopsis hosts suggest that these flies have the capacity to reduce the dominance and ecological impact of introduced fire ants. Several species of Pseudacteon demonstrate specificity to the imported fire ant and are now permitted for release in Texas.

Introduction:

Pest insects and weeds are typically organisms introduced between continents free of their co-evolved predators, pathogens and parasites. Without natural biological controls, such species often dominate the native communities which they have invaded. Any disturbance of the native biotic system may promote these introduced pests since they tend to be opportunistic colonists. In the case of imported fire ants in the United States (IFAs), topsoil disruption and pesticides constitute disturbances to native ant communities which help spread the IFA.

In 1981 the IFA invasion reached Brackenridge Field Laboratory (BFL) in Austin, Texas. Fortunately, Feener (1978) had completed a biological survey of ants for the area and therefore we had a baseline condition against which to measure change. Starting with undergraduate honors student Bill Van Eimeren and finishing with Postdoctoral Associate Sanford Porter, we mapped the course of the invasion from 1983 to 1987 (Porter et al. 1988).

We were surprised by the pattern of the invasion. In most ants, including fire ants, winged virgin queens leave their colony, mate high in the air and land to establish a new colony alone. At first, such airborne invasions would be predicted to lead to new colonies patchily distributed across a landscape still dominated by dozens of species of native ants. This is not what we found. We found a continuous carpet of S. invicta spreading like a large amoeba across the area (Figure 1 87K). At the edge of the amoeba we found native ants on one side and dense IFAs on the other. For every mound of native fire ant lost, seven mounds of the imported species were gained. All native ants declined dramatically, and arthropod diversity was reduced by over 40% (Porter and Savignano 1990).

As the study proceeded at BFL, we noted that all S. invicta colonies in the IFA amoeba were of the multiple-queen or polygyne variety. This form of the fire ant first appeared in the 1970s and is now the predominant form in much of Texas (Figure 2 83K). The more typical single-queen or monogyne form does not reach the extreme colony densities seen in Texas polygyne populations because monogyne colonies defend intraspecific territories, attacking and killing workers from other colonies. Conversely, the lack of territorial defense in polygynes allows them to build to vast numbers with densities of over 1000 colonies per acre not uncommon in parts of Texas (Porter et al. 1991).

What accounts for the pattern of spread observed at BFL? We noted two aspects which required explanation. First, in the zone not yet occupied, there were no S. invicta colonies, large or small, in spite of thousands of mated queens raining down over the area. Second, large colonies appeared along the boundary as it spread. Ed Vargo's work showed how both of these observations are related to polygyny. First, he found that polygyne queens are smaller, lay fewer eggs and are less capable of initiating a colony alone than are their monogyne counterparts (Vargo and Fletcher 1989). Apparently, such queens do not stand a chance of surviving and reproducing in areas occupied by native ants. Teaming up with Porter, Vargo demonstrated that polygyne colonies reproduce by budding, i.e. queens may leave with a group of workers and establish new colonies nearby (Vargo and Porter 1989). This explains the pattern of spread observed at BFL.

Exceptions to the pattern of spread and distribution I have described further support the role of polygyny and interaction with native ants. The one patch of S. invicta separate from the continuous front appeared in a cultivated area treated with pesticides (Figure 1, top 87K). We assume that lone polygyne queens were able to colonize in the absence of native ant resistance. Conversely, the only two isolated colonies of S. invicta discovered in the survey were both monogyne. It is relevant to note that the last open habitat to succumb to the invasion was a patch of little bluestem which dates back to the 1920's as an open grassland. Do the native ants occupying undisturbed native grassland provide better resistance to the invasion? Everything we know about fire ant ecology would support this assumption.

Solenopsis invicta and Relatives in South America

Surprisingly, until Porter and co-workers (Porter et al. 1992) surveyed fire ant densities in Brazil, no substantial studies had been carried out to test whether imported Solenopsis species are relatively less dominant in their home regions. Comparable transects between Texas and Florida and in Mato Grosso revealed striking differences (Figure 3 90K). Brazilian roadsides have 10% of the fire ant density seen in the U.S., and comparisons with multiple-queen populations in Texas are even more extreme (Figure 4 47K). While there are occasional reports of local fire ant problems in Brazil, these are probably related to severe habitat disruption. Our native S. geminata, while generally sparse like S. invicta in South America, can be locally abundant to the point of being a nuisance.

In my own experience in Brazil, mounds are not so easy to find while driving cross country. Even along fields and roadsides in suitable habitat, it is not uncommon to walk 100 - 200 yards between adjacent mounds. In urban lawns in Campinas, Saõ Paulo state, S. invicta densities may rival those seen here in Texas. Remarkably, these mounds are virtually ignored by Brazilians, and dozen of other ant species--large and small--coexist with the fire ants. It is possible to have a picnic next to a fire ant mound in Brazil. Furthermore, if one disturbs a mound in Brazil, workers pour out in defense, as we see here in Texas. But in Brazil, workers run back underground to avoid the attack of parasitoid flies in the genus Pseudacteon (Phoridae). The same can be said for S. geminata in Texas, if you can find them.

Phorid Flies as a Biological Control of Fire Ants

From a large family (Phoridae) of typically scavenging flies, a few genera of ant parasitoids have evolved, including species which specialize on fire ants (see Disney, 1994). Some of the first natural history studies of ant specialist phorid flies were conducted by C.T. Brues (1901, 1907) around Austin, Texas, early this century. Taxonomic descriptions and host associations for ant-attacking phorid genera appeared in the literature during that time. In the early 1970s during a general search for biological control agents, USDA researchers noted that Pseudacteon species frequently associated with fire ants in South America (Williams et al. 1973, Jouvenaz 1983). At that time (as now) the focus was on biotic agents that would inflict high rates of direct mortality on hosts. Interest in phorid flies diminished when it became obvious that only a tiny fraction of ants are infected and killed (1- 3%) by these parasitoids. A conceptual breakthrough with other species of ants and phorid flies led us to reassess the potential of phorids in controlling ants. As part of his dissertation in the (pre-invasion) ant community at BFL in the late 1970s, Feener (1978) conducted experiments to determine species dominance at food baits. He noted that the outcome of competition between Pheidole dentata and Solenopsis texana was always victory by P. dentata (which mobilizes large-headed soldiers specifically against ants of the genus Solenopsis) unless phorid flies arrived to attack the P. dentata soldiers. At that point, the latter retreated to hide, and the food was taken by S. texana (Feener 1981). Feener and I saw immediately that his work revealed the importance of indirect effects (through behavior rather than mortality) and indicated that the presence or absence of host-specific phorids could explain why imported fire ants (without phorids) dominate not only native fire ants (with phorids) but other native ants as well (many with phorids or other biological controls).

From the early 1980s to the early 1990s, I talked and wrote proposals about the prospects for phorids in IFA biocontrol, but we got nowhere with funding for appropriate research. Porter went to the USDA in Gainesville in 1990 and later began to focus on the phorid question there. In February 1994, I travelled to Campinas, Brazil, on an NSF travel grant obtained in collaboration with BFL postdoc Michael Kaspari to establish a project with colleague W.W. Benson at the State University of Campinas (UNICAMP). Porter, meanwhile, was just starting work nearby at Rio Claro with Harry Fowler. Dr. Matt Orr joined the BFL group in March 1994 and went to Brazil May through June to conduct (with the help of Sergio Seike) our first experiments on the impact of phorids on fire ants in Brazil. I should say that the $10,000 required for this work was arranged by an avid quail hunter, rancher and conservationist, who helped "prime the pump" for everything to follow as far as our work on the phorid question is concerned.

The results of our initial observations and experiments indicated a dramatic impact of phorid flies on foraging by fire ant workers during the day in Brazil (Orr, et al. 1995). We found that by removing flies, we could induce a rapid increase in foraging S. invicta at any time of day (Figure 5 78K). However, with flies present, other ants always manage to win competition for food. Independently, Porter and colleagues obtained similar results with different species (S. saevissima) and somewhat different techniques (Porter et al. 1995a). Recently, Orr, Seike and I have expanded our field study to examine the ways that different species of Pseudacteon might impact host ants. We have observed nine of the fourteen species known to be associated with Solenopsis in Brazil and Argentina. We find that some Pseudacteon species are primarily found around disturbed mounds with thousands of workers milling about releasing alarm pheromones. Others are primarily associated with quiet foraging trails, such as supply lines connecting a dead grasshopper with the mound (Figure 6 92K). Some approach victims from the head first; some approach from the rear of the victim. Some pursue a single ant for over a minute before attempting to oviposit while others make frequent but possibly inefficient attempts. Some are more likely to attack larger workers, others less likely, and sizes of phorid species differ substantially. Our current goal is to determine which species or combination of Pseudacteon species will be the most effective in reducing the competitive dominance of S. invicta in its introduced range. Further basic studies on interactions between S. invicta, ant competitors and Pseudacteon in Brazil will be initiated in collaboration with Don Feener (Now at the University of Utah), Matt Orr and our Brazilian colleagues as soon as the Washington gridlock frees up our three-year NSF grant.

A parallel line of research, spearheaded by Lloyd Morrison at BFL, investigates the basis for S. invicta's dominance over S. geminata. We are studying the interactions of these species on the edges of small inholdings of the native ant system in Central Texas. S. geminata contends with two species of Pseudacteon which are absolutely species-specific. Surprisingly, when competing directly for food with S. invicta, S. geminata is a more than worthy adversary in the presence or absence of its parasitoid flies--this holds in the laboratory or in the field. However, when S. geminata forages alone, the arrival of phorids causes defensive posturing and a reduction of foraging as noted by Feener and Brown (1992). Thus, in the parlance of community ecology, Pseudacteon influence exploitative, but not interference, competition between these species (Morrison and Gilbert in preparation). Our studies of S. geminata Ð S. invicta interaction zones is also intended to provide baseline data for monitoring the impact of releasing S. geminata-specific Pseudacteon from South America in Texas. We anticipate being able to detect an expansion of native ant "bubbles" in a "sea" of imported fire ant as phorids which attack the latter reduce their relative advantage and return parity to the ant community.

Porter et al. (1995a) in tests involving 13 genera of ants have determined that South American Pseudacteon are highly specific to the genus Solenopsis as collecting records suggest (Disney 1994). Therefore, a third line of our research addresses the issue of whether South American Pseudacteon species exist which are as host restricted on the S. invicta species group as Texas Pseudacteon are to S. geminata and its species group. The answer, based on work completed over the last year at BFL (Gilbert et al. in preparation), is that several species are highly restricted to S. invicta while others are less discriminating. We used a conservative no-choice test in which individual test flies are first placed in a glass-covered tray with a few hundred S. invicta workers. If the fly shows motivation to oviposit on its known host, we then transfer it to a tray of S. geminata for a set period and carefully observe any indications that it might show interest in our native species. Finally, we move the fly back to the tray of S. invicta to make sure it was indeed interested in ovipositing while confined with the non-host. The "+ - +" pattern of response repeated over an adequate sample of individual females of a species was the criterion for adequate specificity for use in biocontrol. We obtained USDA APHIS permits to release three species in May 1995. At least one other species failed this test by attacking both S. invicta and S. geminata.

Further lines of research are related to the initiation of introduced populations in Texas. If captive flies are indicative (they may not be!), Pseudacteon species have a brief adult life span of several days in nature. At any given instant, most of a population will be at some stage of development inside fire ant hosts. Thus, while harsh weather might kill a cohort of adults above ground, the larval population persists in an environment buffered by the behavior of the ants. Even so, the phorids of tropical Brazil may not be suited to survival in the more extreme temperature zone climates now occupied by IFA in the United States. Therefore, with the help of Patricia Folgarait in Buenos Aires, Argentina, we have been learning details of the ecology and activity cycles of Pseudacteon in a climate much more similar to that of Texas and the southeastern United States than that of Brazil. The goal of the work is to identify species of Pseudacteon that may thrive outside of subtropical south Texas and Florida, and to screen more widely for suitable biological control potential within Pseudacteon.

Although the BFL group (Morrison et al. in preparation) and Porter et al. (1995b) have reared a few Pseudacteon from egg to adult in the laboratory, maintaining a breeding population has not been accomplished. Thus, to attempt introduction, we have collected fire ant workers from outdoor colonies, exposed them to appropriate species of wild caught Pseudacteon (usually from 1 to 48 hours after the phorids have arrived from Brazil). Attacked ants are removed by observers, held in the laboratory for a brief period, then returned to source mounds.

Unfortunately, each of the first three attempts coincided with extremely harsh (hot or cold) weather during the brief period when adult flies were expected to emerge, mate, and oviposit. While Brazil also may have episodes of harsh weather which might kill adult populations, any losses will soon be replaced by adults eclosing from pupae. The presence of eggs, larvae and pupae in the natural population acts as a buffer against short-term disaster. The problem we face for biocontrol attempts is how to introduce a Pseudacteon population with appropriate age structure so that the success or failure of a particular age cohort is not so critical. One possibility is to control the conditions where introductions are attempted. Thus, we are initiating the construction of a large tropical greenhouse designed to allow Pseudacteon to build up populations in a semi-natural state. Even so, it may take repeated collecting trips at weekly intervals to initiate an age-structured population (one with all developmental stages coexisting). A goal for this year is to develop the capacity to grow Pseudacteon populations in a greenhouse or laboratory here in Texas. This has proven to be a challenge but should not stall things too long.

Finally, we know more about the larval life (e.g., Porter et al. 1995b) of Pseudacteon than the adult life. We are only able to keep adults alive 2 - 5 days at normal room temperature. They do feed on sugar solution, but adult diets in nature are not known. When we disturb mounds, phorids seemingly appear out of nowhere, even in some rather harsh and bleak situations (dry, little lush vegetation). Where are these flies while waiting for the opportunity to attack ants? Can they live longer than we think by resting in special microsites? This is crucial natural history information that we do not possess and that is difficult to obtain.

Summary and Prospects:

By their mere presence, phorid flies alter the capacity of host ants to forage efficiently for food or defend nests. Their most significant impact, therefore, is not to reduce populations through direct mortality, but rather to reduce the capacity of target species to compete with other ants for food or (presumably) nest sites. Such indirect effects must translate into lower equilibrium population densities through reduction in the rates that resources can be provided for raising brood. The expected result is that a balanced and diverse phorid community will prevent any ant species from achieving overwhelming dominance. Our goal is to achieve such a balance.

In much of Texas, high-density polygyne fire ant populations are vulnerable to a phorid epidemic simply because flies won't have to search long for hosts. Moreover, if the imported ants have lost some of their anti-phorid vigilance over the last seven decades, phorids may be more effective in rates of oviposition, and this may initially impose a high direct mortality. In this scenario, mortality-caused natural selection by phorids would, through time, restore the avoidance behavior seen in South American S. invicta. At that point indirect effects would again become relatively more important to biocontrol. If S. invicta anti-fly behavior has not been lost since the ants' introduction, we expect indirect effects to be important from the outset of introducing South American Pseudacteon to Texas.

Either way, however, the presence of intact remnants of the native ant community are critical to the success of this approach. Without native ants to take advantage of phorids harassing S. invicta during the day, the rate at which phorids reduce the impact of S. invicta will be much reduced. In anticipation of using Pseudacteon to control S. invicta, it will be of critical importance to minimize harm to native ants, including native fire ants (and, indeed, any components of the faunal system which compete for food or space with S. invicta). In particular, education of citizens on the distinctions between native and imported fire ants (Figure 7 48K) is of critical importance if phorid flies are to be given a fair chance as agents of biological control. Our minimal goal with current funding is to learn enough in three years to be able to "fish or cut bait" concerning the use of phorids in the control of imported fire ants. As far as I know, there has never been a successful biocontrol of a pest ant. My colleagues and I would like to change that dismal fact.

Acknowledgments:

Thanks to all the colleagues and coworkers mentioned above for sharing information and ideas. I thank Debbie Miller, Ed Vargo, and Lloyd Morrison for reading and commenting on the manuscript. John Crutchfield at BFL and my assistant Sharon Bramblett have helped this project in innumerable ways. Research described in this paper was supported in part by a contract provided by the Texas Department of Agriculture, a National Science Foundation travel grant, The James R. Dougherty Jr. Foundation, The Fondren Foundation, and the Houston Livestock Show and Rodeo Education Committee. Special thanks to Ann Sorenson, Ben Vaughan III and Mark Forgason.

References Cited

Brues, C.T. 1901. Two new myrmecophileus genera of aberrant phoridae from Texas. Amer. Nat. 35:337-356.

Brues, C.T. 1907. On the phorid general Plastophora and Pseudacteon. Ent. News 18:430.

Disney. 1994. Scuttleflies: The Phoridae. Chapman and Hall. 467 pp.

Feener, D.H.J. 1978. Structure and organization in a litter-foraging ant community: roles of interference competition and parasitism. Ph.D. Dissertation. University of Texas, Austin.

Feener, D.H.J. 1981. Competition between ant species: outcome controlled by parasitic flies. Science 214:815-817.

Feener, D.H., Jr. and B.V. Brown. 1992. Reduced foraging of Solenopsis geminata in the presence of parasitic Pseudacteon spp. Ann. Ent. Soc. Am. 85:80-84.

Jouvenaz, D.P. 1983. Natural enemies of fire ants. Fla. Entomol. 56:259-262.

Morrison and Gilbert in preparation [see p. 4 of draft]

Morrison et al. in preparation [see p. 5 of draft]

Orr, M.R., S.H. Seike, W.W. Benson, and L.E. Gilbert. 1995. Flies suppress fire ants. Nature 373:292-293.

orter, S.D., B.V. Eimeren, and L.E. Gilbert. 1988. Invasion of red imported fire ants (Hymenoptera: Formicidae): Microgeography of competitive replacement. Ann. Entomol. Soc. Am. 81:913-918.

Porter, S.D., A. Bhatkar, R. Mulder, S.B. Vinson, and D. Clair. 1991. Distribution and density of polygyne fire ants in Texas (Hymenoptera: Formicidae). J. Econ. Entomol. 84:866-874.

Porter, S.D., H.G. Fowler, and W.P. MacKay. 1992. A comparison of fire ant population densities in North and South America (Hymenoptera: Formicidae). J. Economic Entomol. 85:1154-1161.

Porter, S.D., H.G. Fowler, S. Campiolo, and M. Pesquero. 1995a. Host specificity of several Pseudacteon parasites of fire ants in South America. Fla. Entomol. 78:70-75.

Porter, S.D., M.A. Pesquero, S. Campiolo, and H.G. Fowler. 1995b. Growth and development of Pseudacteon phorid fly maggots in the heads of Solenopsis fire ant workers. Env. Entomol. 24:475-479.

Porter, S.D. and D. Savignano. 1990. An invasion of polygyne fire ants decimates native ants and disrupts arthropod community. Ecology 71:2095-2106.

Vargo, E.C. and D.J.C. Fletcher. 1989. On the relationship between queen number and fecundity in polygyne colonies of the fire ant Solenopsis invicta. Physiol. Entomol. 14:223-232.

Vargo, E.L. and S.D. Porter. 1989. Colony reproduction by budding in the polygyne form of Solenopsis invicta (Hymenoptera: Formicidae). Ann. Entomol. Soc. Am. 82:307-313.

Williams, R.N., J.R. Panaia, D. Gallo, and W.H. Whitcomb. 1973. Fire ants attacked by phorid flies. Fla. Entomol. 56:259-262.

*Sponsored by the:

• Texas Agricultural Extension Service
• South Texas Chapter of Quail Unlimited
• Animal and Wildlife Sciences Dept.,Texas A&M Univ.-Kingsville
• Wildlife & Fisheries Sciences Dept., Texas A&M Univ.-College Station

Cooperating Groups:

• Texas Parks and Wildlife Dept.
• USDA Natural Resource Conservation Service, Rob and Bessie Welder Wildlife Foundation March 23, 1996
• Texas A&M University-Kingsville, Kingsville, TX

Edited by:

Will E. Cohen, Ph.D. Assistant Professor and Extension Wildlife Specialist, Dept. of Wildlife and Fisheries Sciences (College Station), Texas Agricultural Extension Service, TAMU