|
The beginning of an |
|
Click on image |
On this page I will discuss in turn:
Because C. opuntiae is the primary user of extrafloral nectaries on my field site, they are also the species that suffers most when S. aurea colonies begin invading EFN-bearing cacti in the late summer and early fall.
A typical S. aurea invasion begins when a few subterranean workers make their way to the surface. They usually do this just after sunset and generally near the base of the contested plant. On plants with few C. opuntiae workers, the host colony may simply abandon the EFN-bearing cactus when S. aurea workers begin invading. But, when there are larger numbers of workers, the host colony normally resists the initial invasion. In fact, the first indication I often had of an invasion was the presence of a large number C. opuntiae workers running around beneath the contested plant. Close examination invariably revealed soil-covered S. aurea workers being assaulted by frenzied workers from the resident C. opuntiae colony.
|
The beginning of an |
|
Click on image |
But almost always this initial advantage in numbers waned as more and more S. aurea workers came to the surface; in over a dozen cases, I only observed possibly two instances where an invasion was repelled. Most of the time the invaders had established trails to the EFN-bearing cactus they were after by daybreak. The following evening normally found S. aurea workers in complete control of the contested plant and visiting its extrafloral nectaries. Typically the only C. opuntiae workers around were a few individuals in the trail leading to the plant. These workers would stop at some distance from the plant and would go no farther. Once, though, a C. opuntiae colony continued the battle for a second night and below is a photograph of two isolated ants struggling in this contest.
|
The beginning of an |
|
Click on image |
There are probably a couple of reasons for why C. opuntiae workers lose in fights with S. aurea workers. First, while C. opuntiae workers have dull mandibles and lack a functional sting, S. aurea workers are well-equipped in both of these departments. Any advantages C. opuntiae has in the use of chemical warfare (Davidson and McKey 1993) may be offset by the fact that S. aurea also engages in it by flagging poison from their sting gland. They accomplish this by turning their gasters up and shaking them violently. In the process small droplets of poison are turned into a noxious aerosol.
A second reason why C. opuntiae colonies usually fail in their resistance is that while the workers of a single C. opuntiae colony may be dispersed among a number of EFN-bearing cacti in their territory, S. aurea workers generally invade a single plant en force. It is interesting to note in this regard that one of the instances where a C. opuntiae colony repelled an S. aurea invasion involved a chainfruit cholla that regularly had the highest number of C. opuntiae workers per plant on my site. Not only this, but the main nest site of this C. opuntiae colony was very close to the cholla.
The above reasons for why C. opuntiae workers generally lose in fights are also possible reasons for why they generally avoid continued fighting. As if these weren't enough reasons there are two additional ones. First, as shown in the photograph below, S. aurea workers can and will kill large numbers of C. opuntiae workers given the opportunity. On the left side of this photograph is a clump of ants consisting primarily of dead C. opuntiae workers.
|
The beginning of an |
|
Click on image |
A second reason for avoiding fights is that active C. opuntiae recruitment trails may serve as conduits for further S. aurea invasions. As an example, theC. opuntiae colony mentioned above that continued fighting for a second night ended up losing not only the contested cholla but also two of its nests. Apparently invadingS. aurea workers used incoming C. opuntiea workers or the trails laid down by them as a guide to the nests. The invaders only stayed in the nests for about a week and then abandoned them. Whether they were in search of other EFN-bearing plants, C. opuntiae brood, or just trying to eliminate an overly combative enemy once and for all isn't clear. It is interesting, though, that their forward progress stopped when the C. opuntiae colony broke off contact with them and kept their distance. Perhaps this explains why most C. opuntiae workers don't get very close to plants that have been taken over recently by S. aurea colonies. It may also explain why C. opuntiae workers sometimes remove brood from nearby nests.
C. opuntiae is not the only ant species that suffers from S. aurea invasions. When invasions are on-going nocturnal Camponotus ocreatus may arrive at their cacti in the evening to find S. aurea workers already gathering nectar from them. Fights between workers of the two colonies invariably ensue. At first glance the contest seems unequal. C. ocreatus workers are much larger that those of S. aureus and their mandibles, especially those of the majors, are impressive. Further, while they lack a functional stinger they can use the modified organ to spray formic acid at their enemies.
With their impressive arsenal, C. ocreatus workers kill many invading S. aurea workers. They are most effective away from the primary S. aurea foraging trail where they can combat their smaller enemies in one-on-one encounters. Here, the larger ants grab their smaller foes with crushing mandibles and then turn their gasters under to spray them with formic acid. Discarding their victims, they immediately go after others. But other S. aurea workers soon respond to the alarm pheremone released by their fallen sisters and, in a short time, the area swarms with them. Some of the responding ants attack the C. ocreatus workers while others flag poison. This last strategy is particularly effective: most C. ocreatus workers who come in contact with the poison are repulsed by it and move some distance away where they vigorously groom themselves in order to remove the obnoxious substance. Some C. ocreatus workers, though, make the mistake of moving directly into the swarming mass of S. aurea workers. These ants are soon attacked by multiple S. aurea workers who grab onto them with their mandibles and continue holding on even when dismembered. Physically hampered in their movements by an accumulation of attached corpses and sickened by continouous poison gas attacks, such C. ocreatus workers are eventually overwhelmed and killed.
As a result of all this, several hours into the first night of an invasion it wasn't unusual to find a half-dozen or more dead C. ocreatus workers at the base of their cholla. Below is a photograph of a victim from one of these battles.
|
The beginning of an |
|
Click on image |
For a few night following a usurpation, C. ocreatus workers will continue to fight withS. aurea workers around the base of their old cactus. But eventually they stop. Interestingly, unlike the situation with C. opuntiae, C. ocreatus workers are generally able to use their old cacti within a few weeks of an S. aurea take-over. As pointed out earlier, S. aurea colonies frequently don't visit all the branches on a cholla they control and so many of these are available to any ant that can reach them. Since large C. ocreatus workers are far more nimble than their C. opuntiae counterparts, they can more easily avoid active S. aurea foraging trails and reach unused branches. That C. ocreatus workers are so readily able to do this brings to question why they fight in the first place. While resource defense is still a reasonable explanation --even if they are able to collect nectar after an invasion they still presumably collect less of it than before --the explanation may also be a function of C. ocreatus colonies taking some time to "unlearn" -- perhaps at the cost of certain workers -- old trails to their cactus.
Unfortunately, I never witnessed S. aurea workers taking over a barrel cactus heavily used by C. ocreatus (while C. ocreatus visited a number of these plants sporadically and in small numbers, they regularly visited in large numbers only one barrel cactus, and this one was never invaded by S. aurea) so I can't even say whether displacement occurs, much less whether it is brief or more long-term. I did, however, observe both species simultaneously visiting a barrel cactus that was just outside the boundaries of my field site. In this instance, C. ocreatus workers would regularly come to the plant in the evening and attack and kill already present S. aurea workers. In the midst of these nightly melees, the C. ocreatus workers were able to collect some extrafloral nectar. But they were always eventually repulsed by the accumulated effects of S. aurea's flagged poison. The repulsed workers would then climb out onto the spines of the barrel cactus and spend long periods of time grooming themselves.
As a curious aside, S. aurea workers often partially bury their victims and I've seen them do this to both C. opuntiae and C. ocreatus workers. While I am not aware of instances in the literature where ants of one species bury individuals of another (although some do drop rocks on interspecific competitors -- Möglich and Alpert 1979), ants of many species use debris to cover certain objects in their environment. The usual explanation is that they cover things they find obnoxious, although this explanation begs for a deeper one.
Interactions between these two species were relatively rare, especially in comparison to the frequencies of the their interactions with S. aurea. Only three cacti on my plot, one a barrel cactus and the other two chainfruit chollas, were regularly visited by both C. opuntiae and C. ocreatus workers. One of these chainfruit chollas was very large and also frequently visited by S. aurea. At any rate, interactions between C. opuntiae and C. ocreatus workers were not common on this plant. In the other two instances, C. opuntiae would use the cacti during the day and C. ocreatus would replace them shortly after sunset. While some C. opuntiae workers would hurriedly abandon these plants before C. ocreatus workers arrived, others would not. Apparently, the job of the first C. ocreatus worker to arrive at these cacti was to go around and attack any C. opuntiae holdovers. While many of the holdovers escaped unharmed, a significant number were crushed or had their bodies severed. Below is a sequence of three photographs showing the fate of a C. opuntiae worker that stayed too long.
|
|
|
|
Click on an image |
From the above photographs it might seem like C. opuntiae workers are unable to stop C. ocreatus workers from coming on to a cactus; that the best a C. opuntiae colony can hope for is to use a shared plant during the day. But there were a few chainfruit chollas quite near C. ocreatus nests that were never visited by these ants. These chainfruit chollas, unlike the two cacti where C. ocreatus workers took over at night, had large numbers of C. opuntiae workers on them suggesting that a very large worker force may be able to repel C. ocreatus workers. However, I never observed C. opuntiae workers chasing off C. ocreatus workers from the heavily occupied plants -- C. ocreatus workers simply didn't visit them. But at the very beginning of my study and at the base of a chainfruit cholla slighly off my main field site, I did see C. opuntiae workers kill a C. ocreatus worker. The cholla was heavily occupied by C. opuntiae workers and, like in the case of incipient S. aurea invasions, a large number of C. ocreatus workers swarmed around the base of the plant.
On the C. opuntiae species web page I reported that C. opuntiae response to my presence changed as a function of their density on a plant: at low densities they were retiring while at high densities they became very active. I conjectured that this change in behavior as a function of density might affect the kind of protection these ants provide cacti. Given the above, it would be also interesting to see if this change in behavior as a function of density predicts whether C. opuntiae workers confront or abandon plants in response to C. ocreatus intrusions.
Davidson, D.W. and D. McKey. 1993. The evolutionary ecology of symbiotic ant-plant relationships. Journal of Hymenopteran Research 2: 13-83.
Möglich, Michael H.J. and G.D. Alpert. 1979. Stone dropping by Conomyrma bicolor (Hymenoptera: Formicidae): a new technique of interference competition. Behavioral Ecology and Sociobiology 6: 105-113.
Copyright (c) 1998 Barry Sullender
Rice University
All rights reserved