|
|
|
Click an image |
|
|
|
On this page I will provide a brief species description and then discuss in turn:
Crematogasteor opuntiae inhabits the hot deserts of Arizona, southern Nevada, and southeastern California. Wheeler and Wheeler (1986) describe the worker coloration as ranging from head and thorax yellow, gaster dark reddish brown to head dark reddish brown, thorax brownish yellow, gaster dark reddish brown with darker infuscation. In all cases the head and gaster are shiny while the thorax is only feebly so.
Beyond Pickett and Clark's (1977) study of the possible protection this ant provides Opuntia acanthocarpa and Wheeler's (1994) examination of the ultrastructure of its spermatheca and its associated gland, there is little detailed information on this ant.
It is worth noting, though, that other ants in this genus play an important role in various ecosystems. In both African and Asian wet tropical forests Crematogaster species are frequently competitive dominants in arboreal habitats (Davidson and McKey 1993). Many also partake in symbiotic relationships with plants ranging from the use of domatia to construction of ant gardens (Davidson 1988).
The ability of these ants to behaviorally dominate other ants species stems from several unusual evolutionary developments in the genus. Among Crematogaster species the sting is no longer functional but is instead flattened in order to bear droplets of viscous liquid poison (Forel 1928). The delivery of this poison is facilitated by the attachment of the postpetiole to the dorsal surface (instead of the anterior end) of the gaster and by the shape of the gaster (flattened on top and rounded at the bottom). These two morphological features allow workers to raise their gaster in a scorpion-like way with the tip pointed forward (Wheeler and Wheeler 1986). Still, such a defensive system might seem inferior to one where a functional sting is present. Davidson and McKey (1993) point out, however, that while stinging is a very effective strategy against vertebrate predators, it is often less effective than chemical warfare when the enemy is another ant species.
The scientific name of this ant implies a close association between it and the EFN-bearing cacti in the genus Opuntia. But, at my field site, C. opuntiae showed no special preference for Opuntias over other extrafloral nectar-producing cacti. What seemed to be most important to C. opuntiae workers -- and all extrafloral nectar gathering ants in my study for that matter -- was how much extrafloral nectar a particular plant was producing.
|
|
|
Click an image |
|
|
|
On the whole, C. opuntiae was the most common EFN visiting ant at my site. Most healthy barrel cacti and cholla attracted dozens of C. opuntiae workers and, in the case of very large chollas, even thousands of them. If C. opuntiae workers were absent on a plant it was either because the plant didn't produce enough nectar to attract the ants or because some other ant species was excluding them: in no case could their absence be attributed to the lack of a nearby colony.
Two species of ants restricted C. opuntiae visitation. In the late summer and fall, Solenopsis aurea workers actively displaced C. opuntiae workers from many of their plants (see aggressive interactions between Crematogaster opuntiae and Solenopsis aurea). This, and the return of C. opuntiae workers to plants after S. aurea workers abandoned them, gave rise to an annual cyclical pattern of plant visitation as shown below.
|
Changes in C. opuntiae visitation of chainfruit cholla and barrel cacti in response to S. aurea visitations. For a general description of this survey see Introduction web page: Study #1. |
What the above graph doesn't show is that C. opuntiae colonies were also occasionally displaced from their plants after dark by nocturnal Camponotus ocreatus workers. Unlike the situation with S. aurea, though, C. opuntiae colonies were often able to visit contested plants during the day and were even able to fend off C. ocreatus incursions with some frequency (see aggressive interactions between Crematogaster opuntiae and Camponotus ocreatus).
On my field site, C. opuntiae workers generally collected nectar from midmorning to well after midnight with the highest visitation to EFN-bearing plants typically occurring from shortly after dark until midnight. This was the broadest daily nectar collecting schedule of any ant species on my site and it varied only slightly with season. Along with F. foetidus, C. opuntiae workers were the only ants to regularly collect nectar during the heat of the day in summer. While there were probably benefits in collecting nectar at times when other ants wouldn't or couldn't, there were clearly costs to this behavior. On June 29, 1994 temperatures soared to 46 degrees Celsius (116 degrees F). By late afternoon on that day, workers like those shown below were on chollas all over my field site: workers who had climbed out onto spines and clamped down with their mandibles before dying.
|
Dead C. opuntiae |
|
Click on image |
While extrafloral nectar might make up the majority of C. opuntiae's diet, the ants do feed on other things. In addition to getting plant juices from EFNs, the ants also gather it at times from aphids. Although it might surprise some, spring time in the desert is occasionally cool and damp enough to support local outbreaks of these small plant-sucking insects. Like many ant species in more temperate climates, C. opuntiae workers collect their excrement, a substance known as honeydew.
|
C. opuntiae workers |
|
Click on image |
C. opuntiae's diet also includes meat. In addition to some scavenging, workers regularly capture termites. How they catch these insects is interesting. On my field site termites typically came above ground at night to collect plant debris. To protect themselves from various predators, they first covered anything they planned to collect with a plaster made from soil, saliva, and perhaps some of their feces. The photograph below shows an erect plant covered by plaster; most of the time, though, termites plaster over small areas of the desert and work underneath these coverings.
|
C. opuntiae workers |
|
Click on image |
C. opuntiae workers cannot break through the plaster once it dries and hardens. What they do instead is hang around the periphery of these plaster sheds and wait for new construction to take place. When it occurs the ants move to these areas of moist plaster and try to grab the termites who are doing the construction. Most of the time they are unsuccessful, but occasionally a worker will manage to stick its head up under the plaster roof and grab a termite. This worker then waves its gaster (rear-end) up and down, releasing a pheromone that attracts other ants to it. These other ants come and help dislodge the termite from beneath its protective cover.
C. opuntiae colonies are highly territorial. Unfortunately I was unable to estimate average territory size or average number of EFN-bearing plants per colony; there were a large number of C. opuntiae colonies on my field site and I only mapped out the territories of a few colonies whose workers visited barrel cacti and chainfruit chollas. Below is a territorial map for one of these colonies.
 |
|
Map of a C. opuntiae colony. Nests and in some cases multiple entrances to the same nest are indicated by triangles. Chainfruit chollas visited by this colony are indicated by stars. Connecting these nests and chollas are a series of well-defined trails. This colony also had a large refuse site under a rock (cross enclosed in a circle) near its major nest site. During the summer of 1994 this colony fought a war with another to the south and took from it the chainfruit cholla shown in red. Nest site and trails in red were established after this war. |
While this C. opuntiae colony was extraordinary in size, it shared many features with other colonies that I mapped. First, all of these colonies were polydomic, having at least one major nest site and a number of smaller ones. The major nest sites were always in the ground while minor ones were either in the ground or inside chollas. Second, an extensive system of regular trails connected nest sites to other nest sites and nest sites to EFN-bearing cacti. Finally, all colonies had a refuse site that was generally located under a rock near the major nest site.
Whether C. opuntiae colonies patrol and defend absolute territories isn't entirely clear. But they certainly attempted to defend their EFN-bearing cacti against S. aurea invasions or C. ocreatus incursions (see Aggressive interactions). As shown in the photograph below, they also occasionally managed to grab other species when they wandered into one of their trails.
|
C. opuntiae workers |
|
Click on image |
But C. opuntiae workers behaved most violently when same-species workers from other colonies strayed into one of their trails. When this happened, resident workers quickly found the intruder and locked onto it with their mandibles. Holding the intruder in place, these resident workers then recruited others who, with their dull mandibles, joined in the laborious process of drawing and quartering the captured ant.
Much less common than executions but far more spectacular were the wars that C. opuntiae colonies occasionally waged against each other. These affairs involved hundreds or even thousands of ants. During two-and-a-half years of observation, I only witnessed two of these confrontations. The first occurred at the very beginning of my study and I wasn't able to determine either the outcome or even which two colonies fought. I was, however, able to follow the second war from beginning to end. It lasted nearly a week and consisted of a series of battles. While both colonies survived, the invading one ended up losing one of its chainfruit chollas (the red cholla in the above map). Pictured below, first from a distance and then closer up, is the first and largest battle of this war.
|
|
|
Click on an image |
Unlike the situation with Solenopsis aurea, the presence of C. opuntiae on EFN-bearing cacti tended to decrease the chance that other ant species would visit the plant. Below are some results of a survey of 230 cholla that I conducted on the nights of September 10, 11, 13, 20, and 24, 1994.
|
|
of Co-occurrences |
of Co-occurrences |
|
|
C. opuntiae / S. aurea |
|
|
|
|
C. opuntiae / Camponotus ocreatus |
|
|
|
|
C. opuntiae / Camponotus festinatus |
|
|
|
|
C. opuntiae / Forelius foetidus |
|
|
|
|
C. opuntiae / Forelius pruinosus |
|
|
|
|
Ant species occurrence in the presence of C. opuntiae. Observed number of co-occurrences were corrected for continuity using the Yate's correction and then compared to the expected number using the log-likelihood ratio test (G-test).For a brief description of this survey see Introduction web page: study #4. |
As discussed above and on the Aggressive interactions web page, C. opuntiae workers frequently interact violently with workers of S. aurea and C. ocreatus. Because of this, it isn't surprising that these species co-occur with C. opuntiae less frequently than expected by chance. But, as shown above, C. opuntiae and S. aurea do co-occur occasionally on the same EFN-bearing cacti. In almost all cases, though, the co-inhabited chollas were very large ones. In fact, chainfruit cholla with both C. opuntiae and S. aurea were significantly larger than plants with either species alone (Mann-Whitney U test, p<.0001 in both cases). This survey supports observations at my field site that C. opuntiae colonies can re-establish some use of plants taken over by S. aurea if they can avoid S. aurea foraging columns. While this is accomplished most easily on chainfruit chollas with large bases, several C. opuntiae colonies at my field site also gained access to plants by using branches that touched the ground.
Although not applicable to the above survey, I also observed that as nights became progressively colder the likelihood that both species could be found on the same plant also increased. Shifting their foraging schedule to the warmer part of the day, many S. aurea colonies essentially vacated their plants shortly after nightfall. This made it possible for C. opuntiae to use some of these plants. Sometimes they even visited the cacti before all the S. aurea workers had left and they often even collected nectar only a few centimeters away from them. For whatever reason, neither species behaved aggressively under these circumstances.
While it is fairly apparent why C. ocreatus and S. aurea often don't share plants with C. opuntiae, why the two Forelius species don't is less clear: both species are fast and nimble and can easily avoid C. opuntiae attacks. The explanation, though, may lie in the nature of the C. opuntiae-occupied plants at the time of the year when the survey was done and the mass foraging behavior of the two Forelius species. Because the survey was done in the fall and at night, C. opuntiae workers remained on only the most heavily defended plants. The high worker density on many of these plants would likely mean that there was little or no extrafloral nectar available to elicit the mass foraging of the two Forelius species. This may also explain why C. festinatus workers do not avoid C. opuntiae-occupied plants. Unlike the two Forelius species, C. festinatus forages alone or in small groups and individual workers wander widely over plants.
Several facts suggest that these ants could provide valuable assistance to plants. More than those of any other ant species at my field site, C. opuntiae workers consistently visited the plants they occupied. Also, with regard to chainfruit chollas, they didn't tend to leave certain branches unprotected like S. aurea workers frequently did. Finally, my observations and those of others clearly indicate that these ants often behave aggressively toward various cactus herbivores (Pickett and Clark (1979); also see photograph).
Something to be considered, though, is the apparent change in behavior of these ants with density. When worker density is high these ants often become very excited in response to disturbance -- they would circle around the EFNs they were feeding on with their mandibles open and many also ran out to the ends of spines. At night I often didn't even have to physically disturb a cactus to elicit this response; the ants seemed to respond to an odor cue when I approached them from upwind. Also, the workers would occasionally grab the parchment-like sheath found on new cactus spines (visible on some spines in the above photograph) with their mandibles and chew on it. Whether this was a tactile signal (these ants already have a stridulatory organ -- Hölldobler and Wilson (1990)) or served some other purpose, the noise created by it was clearly audible.
Conversely, for all their activity on densely occupied plants, the workers are often shy and retiring on lightly occupied ones. Rather than behaving aggressively when I disturbed their plants, these ants often became very still. If I touched them with a pencil they often fell from the plant.
Andersen, A.N. 1997. Functional groups and patterns of organization of North American ant communities: a comparison with Australia. Journal of Biogeography 24: 433-460.
Davidson, D.W. 1988. Ecological studies of neotropical ant-gardens. Ecology 69: 1138-1152.
Davidson, D.W. and D. McKey. 1993. The evolutionary ecology of symbiotic ant-plant relationships. Journal of Hymenopteran Research 2: 13-83.
Forel, A. 1928. The Social World of Ants in Relation with that of Man, 2 volumes. Translated by C.K. Ogden. Putnam's Sons, London.
Hölldobler, B. and E.O. Wilson. 1990. The Ants. The Belknap Press of Harvard University Press, Cambridge, Massachusetts.
Pickett, C.H. and W.D. Clark. 1979. The function of extrafloral nectaries in Opuntia acanthocarpa (Cactaceae). American Journal of Botany 66: 618-625.
Wheeler, D.E. and P.H. Krutzsch. 1994. Ultrastructure of the spermatheca and its associated gland in the ant Crematogaster opuntiae (Hymenoptera, Formicidae). Zoomorphology 114: 203-212.
Wheeler, G.C. and J. Wheeler. 1986. The ants of Nevada. Los Angeles County Museum of Natural History, Los Angeles. 138 pp.
Copyright (c) 1998 Barry Sullender
Rice University
All rights reserved