Excerpt: Interactions between Pheidole megacephala and Linepithema humile in Oahu, Hawaii by Sam S. Fluker

Excerpt from the paper by Sam S. Fluker, "Sympatric associations among selected ant species and some effects of ants on sugarcane mealybugs in Hawaii" (1969)

In tests conducted in the laboratory, it was found that when colonies of Argentine ants (Linepithema humile) or Pheidole megacephala were placed in contact with a colony of Anoplolepis gracilipes, the workers and queens of A. gracilipes were always destroyed and the brood was taken as food by the other ant. A. gracilipes seemed to make no effort to defend itself or the brood. Instead the workers would run about the enciosure in panic and most would be killed by running into the "Tanglefoot" barrier rather than in fights with the other ants.

Colonies of Argentine ants and Pheidole which had been maintained in the laboratory for three weeks were put together so that there was continuous contact along one side of the nest platforms. Within ten minutes, the whole colony of Argentine ants had crossed over to the Pheidole platform. This invasion included not only workers but the brood and many queens as well. There seemed to be no concentrated attack by the Argentine ants; however, there were chance encounters caused by the Argentine ants scurrying about. An occasional Argentine ant would go into the area where the Pheidole colony was concentrated. It would be immediately attacked by one or more Pheidole workers and usually would be killed. The only unusual excitement in the Pheidole nest was caused by the occasional invasion by an Argentine ant. However, there seemed to be a response to the invasion throughout the Pheidole colony whereby the Pheidole began to seal the entrances to the nest with trash as a protective measure.

In less than 90 minutes, the Argentine ants had moved all their brood and queens back into their original colonies with only an occasional Argentine ant venturing onto the Pheidole platform. when an Argentine ant would cross over to the other platform, it would only go to the outer perimeter.

After four hours, there was no sign that either ant would attack the other in sufficient numbers to eliminate the colony. The Argentine ants still moved about their platform in the usual foraging manner, occasionally crossing over to the Pheidole piatform. The Pheidoe were still in the process of sealing the entrances to their nest. There were a few soldiers ani workers of PheidoZe on the Argentine platform.

At five hours after beginning the experiment, the Argentine ants began moving their brood from one nest to another. In a short time the colony became very excited and started moving the brood out of both nests and eventually setitied down at the opposite end of the platform from the nests. This excitation of the colony seemed to have been preceded by the invasion of the Argentine ant colony by four or five Phetdote soldiers, The Argentine ants seemed to go out of their way to avoid contact with the soldiers, : even when the soldiers were injured. Pheidole soldiers are much slower in their movements than the Argentine ant workerss; but they snap their mandibles at any moving object whichi near They will do this to even their own species when they are very excited.

After a 12 hour period had elapsed, the Argentine ant colony was still grouped with their queens and brood at the corner of the platform farthest from the nests. After the lights in the laboratory had been on for five to ten minutes, the Argentine ants began to move back into the two nests. Within a period of ten minutes, they had moved al1 of their brood back into the nests. At this time there was very little disturbance of the Argentine ant colony by Pheidole.

At 24 hours from the beginning of the experiment, the complete Argentine ant colony had moved again. This time they moved from the nests into the corner farthest from the Pheidole and their own nests. At this point in time, the former Argentine ant nests were occupied by six to ten Pheidole soldiers and approximately fifteen workers in each nest. The Argentine ants appeared to be very sluggish in their movements and were making no attempt to re-establish themselves in their nests. Even when a 75-watt 1ight was turned on directly above them, no attempt was made to protect themselves or their brood from the light. The colony of Pheidole seemed to be engaged in its normal activities without undue excitement,

Late in the morning of this second day of observations, another artificial nest was put on the platform containing Argentine ants. The colony immediately began moving into the nest. In less than 20 minutes the entire colony of Argentine ants was inside the nest.

Approximately four hours later, the colony of Argentine ants was still in the new nest with the Pheidole occupying the original Argentine ant nests.

At eight o clock in the morning of the following day, it was observed that the Argentine ants had moved out of the new artificial nest and were on top and to one corner of the nest. Shining bright lights on the colony did not force them to move. Only one Pheidole was seen in the nest which the Argentine ants had vacated.

The experiment explained above was repeated in the laboratory four times. Each experiment ended with Pheidole having possession of the nests and the Argentine ants huddled in a corner as far away from Pheidole as they could possibly get.

A later experiment consisted of placing a small colony of Argentine ants in an artificial nest with sand completely covering the floor of the platform to a depth of one-half inch. This platform was connected to another platform containing a very large colony of Pheidole. The Argentine ants immediately began to cross over to the Pheidole platform. There were so many Pheidole at the base of the bridge that the Argentine ants were unsuccessful in crossing over to the other platform. When one would try to make the crossing, it would be immediately attacked by Pheidole soldiers and workers. This situation was observed for one hour. The following morning, it was noted that the complete colony of Argentine ants had been killed by the Pheidole ants. The nest which had been originally occupied by the Argentine ants was inhabited by Pheidole. Upon close examination, it was observed that numerous Pheidole workers were carrying brood to their original nests. It was determined that this was the Argentine ant eggs, larvae and pupae which were probably used as food by the Pheidole colony.

In efforts to establish Argentine ants in the ant infested plots of sugarcane at the Waimanalo Experiment Station, two large colonies of Argentine ants were brought from Wailua Agriculture Company Field Opaeula Number Nine to the Waimanalo plots. The ants were transported from Wailua to Waimanalo in heavy plastic bags The two plastic bags containing the ants were placed at the edge of the plot and a hole was punched in each bag. This hole was made to allow the insertion of a sugarcane leaf into the bags. The leaf was positioned so that the ants would move along the leaf to a horizontal cane stalk which was lying on the ground. Some loose soil had been previously placed over a portion of the stalk about three feet from the plastic bags as a possible nest. The ants immediately began moving their brood from the plastic bags to the place where the soil was covering the stalk. By close observation, it was noted that the workers would take the brood to the soil covering the stalk and return to the plastic bags for more brood.

During six hours of almost constant observation, there was a steady stream of workers with brood going to the new nest from the plastic bags and returning empty to the bags after more brood. At the end of six hours, it was noticed that some Pheidole soldiers and workers were at the base of the sugarcane stalk approximately two feet from the nest. At the same time large numbers of Pheidole soldiers and workers were observed on the ground about one foot from the Argentine nest. Within four hours after the first Pheidole was seen in the vicinity of the Argentine ant nest, the Pheidote ants had moved into the Argentine ant nest with the result that the Argentine ants had moved their brood out of the nest and through an area which had been treated with 2.58 Dieldrin granules. The Argentine ants moved even though they outnumbered the Pheidole tremendously. There were close to ten thousand Argentine workers in the nest and less than one hundred Pheidole near the area.

Argentine ants were introduced to the Waimanalo Experiment Station plots on several occasions prior to this using similar methods. Each time they disappeared, so this particular attenpt was made to determine their actions and movements.

The behavior exhibited by Pheidole and the Argentine ants suggests that glandular secretions are involved and play a large role in the behavior on one ant toward the other. Crowell (1968) theorized that the Argentine ant released some type of glandular secretion which was very repugnant to Pheidole, causing Pheidole to move away from concentrations of Argentine ants. If indeed there is a chemical repellent released by the Argentine ants, my studies seem to indicate that only under certain conditions do the Argentine ants release this glandular secretion. This appears to be during the time immediately preceding an invasion of new territory by the Argentine ants. If a glandular secretion is produced by the Argentine ants, this could account for the "Buffer Zone" that is commonly found separating the Argentine ants and Pheidole during the former's "invasion cycle". At other times, Argentine ant and Pheidole colonies are found very close together.

The studies conducted in the laboratory and sugarcane plots at Waimanalo Experiment Station appear to indicate that the Pheidole soldier may release a glandular secretion that is offensive to the Argentine ants. The action of the Argentine ant in the presence of the Pheidole soldier suggests this possibility. Although the Argentine ants will attack and kill Pheidole workers and will continue to mutilate their bodies after death, the Argentine ants appears to go to great extremes to avoid a Pheidole soldier. Even if the Argentine ants far outnumber the Pheidole soldiers, they will make no attempt to attack. No attempt was made to determine if the Pheidole soldier has a glandular secretion that was responsible for the actions of the Argentine ants. However, I find it difficult to believe that the mere physical presence of a few Pheidole soldiers is all that is needed to cause a colony of thousands of Argentine ants to abandon otherwise suitable nesting sites.

How P. megacephala repels Eciton army ant raids

Eciton burchelli army ant.
Image by Alex Wild (from Wikipedia)

There is a certain morbid fascination associated with watching the depredations of various army and driver ants on their hapless prey. The huge colonies of these ants and their mass raids make them formidable predators against most ant colonies. Even the very populous Atta leafcutter colonies, which are normally immune to most army ant species, can be overwhelmed by army ants specialized against them, such as Nomamyrmex esenbecki. (Swartz M, 1994)

Some invasive ant species too are prey to army ants. Colonies of the little fire ant Wasmannia auropunctata for example, are devastated by the army ant Neivamyrmex compressinodis and perhaps some other Neivamyrnex species (J. Le Breton et al, 2007).

Wasmannia auropunctata.
Image from Hawaii Invasive Species Council (HISC).

Solenopsis geminata colonies also fall prey to Eciton and Labidus army ant raids, although most times the fire ants manage to abscond from their nest ahead of the invasion (Perfecto 1992).

Pheidole ants as a rule tend to be favored prey not only for the larger Eciton army ants, but for the smaller Neivamyrmex ants as well, probably due to their smaller colony populations and ubiquity in tropical and subtropical environments. This has resulted in some very interesting social behavior, including the classic discovery of enemy specification by E.O. Wilson (Wilson E.O, 1976).

However, unlike other Pheidole ants, P. megacephala with its huge unicolonial societies and extremely aggressive workers might pose a problem for even the most voracious army ants.

Indeed, P. megacephala in Africa actually preys on the formidable Dorylus driver ants, and a study done by Dejean et al (2014) reveals that this invasive ant is able to handle the New World Eciton army ants as well.

P. megacephala minors prey on Dorylus worker.
Image courtesy of Alex Wild. Click to go to Myrmecos.net

The researchers observed interactions between P. megacephala and two Eciton species in Puerto Morelos, Mexico. They analyzed 24 raids by Eciton burchelli on a P. megacephala colony, and 11 raids by Eciton hamatum.

According to the researchers, the incoming raids were met by fierce resistance from the Pheidole, with many army ants spread-eagled by the defenders as the raiders pushed inward through the colony entrances. In addition to those being killed in the melee by the P. megacephala, outgoing army ants after seizing booty or escaping from the Pheidole nest were in turn attacked by their own nestmates!

During each raid, an average of 1,869 E. burchelli workers were killed by their own colony members, while an average of 1,380 E. hamatum army ants were killed by their fellow nest mates. These casualties make up a not insubstantial 0.4 % and 0.55% of the total estimated population of typical E. burchelli and E. hamatum colonies, and the loss of outgoing raiders stifled the invasion of the P. megacephala nests.

The researchers noted that it seems as if an aqueous compound(s) was transferred to the attacking army ants as they battled P. megacephala, which caused their colony mates to attack them instead, and thus indirectly abort the raid.

Given the high mortality suffered by the army ants, and since P. megacephala is not native to the neotropics and thus a "novelty" to the native ants there, it would be interesting to speculate on whether the army ants might later evolve mechanisms to avoid raiding such nests, just as they avoid other heavily protected species like Atta leafcutters.

For more information on this interesting study, consult the paper:

Dejean Alain, Azémar Frédéric, Roux Olivier. (2014) "An invasive ant species able to counterattack marabunta raids." Comptes rendus biologies 337 (7-8): 474-9.

Abstract:

In the Neotropics where it was introduced, the invasive ant Pheidole megacephala counterattacked raids by the army ants Eciton burchellii or E. hamatum. The Eciton workers that returned to their bivouac were attacked and spread-eagled and most of them killed by their outgoing colony mates. Little by little the zone where returning and outgoing Eciton workers encountered one another moved away from the Pheidole nest which was no longer
attacked, so that most of the colony was spared. Using a water-based technique rounded out by bioassays, we show that Pheidole compounds were transferred onto the Eciton cuticle during the counterattacks, so that outgoing workers do not recognize returning colony mates, likely perceived as potential prey. Because P. megacephala is an introduced African species, this kind of protection, which cannot be the result of coevolutive processes,
corresponds to a kind of by-product due to its aggressiveness during colony defence.

Dejean Alain, Azémar Frédéric, Roux Olivier. (2014) "An invasive ant species able to counterattack marabunta raids." Comptes rendus biologies 337 (7-8): 474-9.

J. Le Breton, A. Dejean, G. Snelling, J. Orivel.  (2007) Specialized predation on Wasmannia auropunctata by the army ant species Neivamyrmex compressinodis. J Appl Entomol, 131, pp. 740–743

PERFECTO I. Observations of a Labidus coecus (Latreille) underground raid in the central highlands of Costa Rica. Psyche. 1992;99:214-22

Swartz, M. (1994) Predation on an Atta cephalotes colony by an army ant, Nomamyrmex esenbeckii. Biotropica 30(4):682-684

Wilson EO. (1976) The organization of colony defenses in the ant Pheidole dentata Mayr (Hymenoptera: Formicidae). Behavioral Ecology and Sociobiology  1: 63–81.

Butterfly effects: Acacias, Elephants, and the Ants that guide their relationship

Acacia drepanolobium and Crematogaster nigriceps.
Image courtesy of Pharaoh han

It is sometimes difficult for people to imagine how the activities of minute creatures like ants can have major effects at the meter scale world, but a paper I read recently demonstrates such a process clearly.

In Kenya there is a tree called the whistling thorn acacia (Vachellia drepanolobium – formerly Acacia drepanolobium), an iconic shrub of the East African savanna that grows to 6 meters tall and is covered with intimidating thorns, some of which have bulbous bases and are called domatia.

The bulbous bases are hollow, and house symbiotic ants that protect the tree from browsing by herbivores. The tree also provides the resident ants nutrition in the form of sugary secretions from glands at the base of their leaves.

The tree houses one of four different species of ants: Crematogaster sjostedti,  C. mimosae, C. nigriceps, and Tetraponera penzigi, all of which swarm out to attack browsing animals with varying degrees of effectiveness, though the overall efficacy of the ants as guardians is quite high. As shown in the youtube video above, acacias with no ants tend to do very badly when large herbivores are present.

This relationship between the acacias, herbivores like elephants, and the ants that guard the acacias has recently been disrupted by the appearance of Pheidole megacephala (BHA) in the Laikipia region of Kenya within the last 10-15 years. In a research paper in the journal Ecology, Corinna Riginos and her colleagues moved complete trees along with their resident ants from uninfested locations to areas infested with BHA.

Almost immediately, BHA discovered the new trees and started moving up the trunk, only to be met in force by the aggressive Crematogaster species, who streamed down to meet and repel the invaders. In all such cases, however, the resident ants were fairly quickly pushed up the trunk and soon vanquished, their nests looted and any remaining adult defenders killed if they could not escape. The only exception was in trees with the Tetraponera ants, who instead of meeting the invaders head on, instead retreated into their domatia where the BHA could not get to them.

Acacia seedpods. Image courtesy of Chr. Kooyman

The researchers found that even 30 days after the invasion, with BHA patrolling throughout the entire tree, the Tetraponera ants managed to eke out an existence. Foragers who came out lay still and did not fight BHA ants who came to them, and the BHA ants did not seem to consider them to be hostile. In fact, the density of Tetraponera colonies significantly increased in invaded areas, due to the removal of their Crematogaster competitors.

Unfortunately for the acacia trees, the Tetraponera is the least effective defender against encroaching herbivores, and BHA itself does not attack large mammals who decide to browse on the plant. This resulted in significant damage to the host acacias, with the researchers finding up to a seven fold increase in the number of trees catastrophically damaged by elephants in invaded areas versus those trees in uninvaded locations. This damage to the trees due to the disruption of the mutualism between plants and ants has the potential to cause significant changes in the dynamics of the savanna ecosystems where this acacia is dominant.

The relevant paper is:

Riginos, C., M.A, Karande, D.I. Rubenstein & T.M. Palmer. 2015. Disruption of a protective ant–plant mutualism by an invasive ant increases elephant damage to savanna trees. Ecology 96:654-661.3

The abstract of the paper:

Invasive species can indirectly affect ecosystem processes via the disruption of mutualisms. The mutualism between the whistling thorn acacia (Acacia drepanolobium) and four species of symbiotic ants is an ecologically important one; ants strongly defend trees against elephants, which can otherwise have dramatic impacts on tree cover. In Laikipia, Kenya, the invasive big headed ant (Pheidole megacephala) has established itself at numerous locations within the last 10-15 years. In invaded areas on five properties, we found that three species of symbiotic Crematogaster ants were virtually extirpated, whereas Tetraponera penzigi co-occurred with P. megacephala. Tetraponera penzigi appears to persist because of its non-aggressive behavior; in a whole-tree translocation experiment, Crematogaster defended host trees against P. megacephala but were extirpated from trees within hours. In contrast, T. penzigi retreated into domatia and withstood invading ants for >30 days. In the field, the loss of defensive Crematogaster ants in invaded areas led to a five- to seven-fold increase in the number of trees catastrophically damaged by elephants compared to un-invaded areas. In savannas, tree cover drives many ecosystem processes and provides essential forage for many large mammal species; thus, the invasion of big-headed ants may strongly alter the dynamics and diversity of East Africa’s whistling thorn savannas by disrupting this system’s keystone acacia-ant mutualism.