|
|
|
Click on an image |
|
workers on a barrel cactus |
cholla flower bud |
On this page I will provide a brief species description and then discuss in turn:
Like other fire ants and unlike species in the subgenus Diplorhoptrum, S. aurea workers are relatively large and slightly polymorphic (see pictures below). This species is morphologically similar to the other fire ant found in the desert southwest S. maniosa , a species considered by some to be only a form of S. xyloni (Creighton, 1950). The primary character that distinguishes S. aurea from S. maniosa is the significantly smaller eyes of S. aurea's workers (Snelling and George, 1979). A less dependable character used to separate these two species is their color: S. aurea workers are generally light, golden yellow with pale brown stripes at the margins of their gastric segments while S. maniosa workers are light to dark red with a mostly black gaster. Snelling and George (1979) also argue that colonies of S. aurea are less populous than those of S. maniosa and that the workers are less aggressive when disturbed, two observations with which I concur. At any rate, while S. maniosa appeared to be fairly common in downtown Tucson all the Solenopsis workers that I keyed out at my field site appeared to be aurea.
|
|
|
Click on an image |
|
workers on a barrel cactus |
cholla flower bud |
Little has been reported about this species. Snelling and George (1979) mention that they found alates in December at Pushawalla Palms, CA. Additionally, there have been a couple of studies exploring temperature preference and heat tolerance in this species. Cokendolpher and Francke (1985) used a linear temperature gradient apparatus to compare the temperature preferences of four Solenopsis species --S. aurea, S. invicta, S. geminata, and S. xyloni -- at both 0 and 100% relative humidity. Since S. aurea is a desert species while the other three live in more mesic environments, the researchers expected that S. aurea would have the highest temperature preference. But, surprisingly, it turns out that this species has the lowest preferred temperature at both 0 and 100% relative humidity. Furthermore, by examining the same four species, Francke et al. (1985) showed that S. aurea is no more heat tolerant that the other three species.
Perhaps because they are just as sensitive to heat as their more mesic counterparts, S. aurea colonies at my field site led a largely subterranean existence. But beginning in the early summer and continuing into the fall, polymorphic S. aurea workers would arise from their underground abodes to collect extrafloral nectar and sometimes even pollen from EFN-bearing cacti. They would generally visit these plants for several months and then begin abandoning them so that by late winter/early spring very few cacti were visited by them. Below is a chart showing this cyclical visitation pattern.
|
The number of chainfruit cholla and barrel cacti occupied by S. aurea. For a general description of this survey see Introduction web page: Study #1. While numbers of plants occupied by S. aurea at at the survey time were generally reflective of the number of plants occupied at other parts of the day, there were some exceptions. For instance, daytime temperatures on 9/28/94 were unseasonably high and, as the chart above shows, I recorded S. aurea workers on only 18 plants. A survey later that night showed that S.aurea workers actually occupied 38 plants. The temperature was likewise unseasonably high on 10/3/93. Although I strongly suspect that this also led to a severe underestimate of the number of plants visited by S. aurea on this date, I did not conduct a nighttime survey to verify this suspicion. From these two situations it is interesting to note that S. aurea workers, like those of some other species, appear to have a maximal temperature at which they are active that changes with the season (Heatwole and Muir, 1989). In other words, workers regularly collected nectar in the summer at temperatures that caused inactivity during the fall. Finally, not all major deviations from the trend line are explainable by temperature. The unusual increase in the number of plants occupied by S. aurea on 7/20/94 was real and was the result of an unusually large number of temporary plant visitations that occurred for unknown reasons. |
There are a number of possible reasons for why S. aurea colonies visit chainfruit cholla and barrel cacti at certain times of the year and not others. Although they start visiting cacti prior to the summer monsoons much of their collecting is done during and just after it. Perhaps changes in humidity or soil moisture brought on by the monsoons favor surface activity in this species. It is also possible that either the composition or the amount of extrafloral nectar produced by chainfruit cholla and barrel cacti changes at this time of the year in a way that favors its collection. Finally, changes within the colonies themselves could could be responsible. Many if not most ant species undergo regular seasonal changes in the number, size, and condition of workers and sexuals (Jensen, 1978; Rissing, 1987; Tschinkel, 1993). If S. aurea undergoes such regular seasonal changes it could have effects on such pertinent variables as the nutritional requirements of colonies and the strength of the worker force available to take over cacti. Unfortunately, little is known about S. aurea 's reproductive biology or about seasonal changes in the demographics of its individual colony members.
Any of the above factors may also be sufficient for explaining why S. aurea colonies stop visiting EFN-bearing cacti in the later winter and early spring. But another possible explanation is that it is a response to intense predation. Two species of spiders in the genus Euryopis feed heavily upon these ants (see spider predation). These spiders almost certainly came to S. aurea occupied cacti to feed on these ants -- many of the invaded plants were previously controlled by Crematogaster opuntiae, a species not preyed upon by the spiders. And it was my subjective impression that the number of spiders on these plants increased over time as did overall predation. At the very least this hypothesis for why S. aurea colonies stop visiting EFN-bearing cacti seems readily testable.
S. aurea invasions have a number of repercussions for other ant species that use EFN-bearing cacti. Most conspicuously, S. aurea workers directly and often violently displace what is generally the most common ant on EFN-bearing cacti, C. opuntiae (see aggressive interactions between S. aurea and C. opuntiae). C. opuntiae workers cannot reoccupy EFN-bearing cacti until S. aurea workers abandon them the frequency of the two species on these plants varies in an inverse fashion. This is shown in the following graph.
|
Changes in C. opuntiae visitation of chainfruit cholla and barrel cacti in response to S. aurea visitations. See above figure caption for a description of this survey |
S. aurea workers also frequently displace Camponotus ocreatus workers from their plants. And like the case with C. opuntiae, these takeovers are often violent ( see aggressive interactions between S. aurea and C. ocreatus). But they differ in that C. ocreatus workers are often able to reuse at least some portion of their old plants shortly after a takeover. At least this seems to be the situation when chainfruit chollas are invaded. The reason for this stems from C. ocreatus 's preference for large chollas (see xxx). When S. aurea take over one of these cacti they rarely collect extrafloral nectar from all of its branches. Instead, they visit particular branches en masse and only now and then visit new ones. Within a couple of weeks C. ocreatus workers are often able to visit these unused branches by nimbly avoiding active S. aurea foraging trails -- something the much smaller and slower C. opuntiae workers apparently find much harder to do. But what happens when S. aurea colonies takes over barrel cacti being used by C. ocreatus? Since S. aurea workers usually visit all of the extrafloral nectaries on the crowns of barrel cacti it suggests that they would exclude C. ocreatus from using them. Very limited data, though, suggest that this may not be the case. Here, plant architecture may allow the two species to simultaneously visit the same plant (see aggressive interactions between S. aurea and C. ocreatus).
While S. aurea invasions and their displacement of C. opuntiae and C. ocreatus workers are dramatic, the effects of their invasions are not limited to these species. 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 |
|
|
S. aurea /C. opuntiae |
|
|
|
|
S. aurea / C. ocreatus |
|
|
|
|
S. aurea / Camponotus festinatus |
|
|
|
|
S. aurea / Forelius foetidus |
|
|
|
|
S. aurea / Forelius pruinosus |
|
|
|
|
Ant species occurrence in the presence of S. aurea. 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. |
First, the survey is in concordance with the above behavioral observations: C. opuntiae workers are unable to use plants once S. aurea workers displace them while C. ocreatus workers are able to rapidly reestablish use of their old plants. More interesting, though, is that Camponotus festinatus, Forelius foetidus, and Forelius pruinosus all seem to be attracted to S. aurea occupied plants in excess of that predicted by chance. This is in distinct contrast to C. opuntiae occupied plants where these three species and C. ocreatus are found at frequencies significantly less than that predicted by chance (see table on C. opuntiae web page). In essence, by displacing C. opuntiae colonies from plants, S. aurea may open them up for use by other ant species. This likely reflects differences in how the two species collect extrafloral nectar: C. opuntiae workers spread out over a cholla and are diligent in attending EFNs while S. aurea workers mass forage on particular branches leaving others unattended.
S. aurea collected not only nectar but also pollen at my field site. While workers from other ant species occasionally visited the flowers of EFN-bearing cacti on my field site, S. aurea workers were the only ones that regularly visited flowers and collected significant amounts of pollen from them. To collect this pollen S. aurea workers aren't particularly subtle. Using their sharp mandibles they cut the stamen filament and then carry away the anther. The effects of pollen theft are fairly inconspicuous on large, stamen-rich barrel cacti flowers. But this isn't the case when anthers are taken from chainfruit cholla flowers. As the photographs below show, S. aurea workers sometimes remove all of the anthers from these flowers. Typically they do this shortly after a flower opens.
|
|
|
Click on either image |
|
process of having its anthers removed |
all their anthers removed |
Just because barrel cactus pollen theft isn't conspicuous doesn't mean that it doesn't have fitness consequences for the plant. In fact, it may cost these plants far more than near total pollen removal does for chainfruit cholla. The reason for this is that while barrel cacti reproduce sexually chainfruit rarely if ever does (Johnson, 1918). Instead, chainfruit cholla reproduce vegetatively with branches and fruits serving as vegetative propagules. Unless successful cross-pollination is somehow required in order to produce fruits (fruits may or may not have viable seeds), the cost to chainfruit cholla of having its pollen stolen appears slight in comparison to the cost of having produced it in the first place.
But given that there may be costs to attracting S. aurea workers what are the possible benefits? To their credit, S. aurea workers are quite aggressive when disturbed. A pen raked across the crown of a barrel cactus or the top of chainfruit cholla would cause workers, especially smaller ones, to run up and down spines while releasing aerosolized poison from their sting gland ( a behavior known as flagging). Additional workers would even climb out onto the pen and attack it. Such behaviors suggest that these ants might be able to ward off vertebrate enemies of cacti. Also, I observed one instance where these ants drove off a large katydid.
Other factors are also probably important in evaluating what if any protection these ants may provide. First, their presence on cacti at certain times of the year and not others places limits on the potential protection they could provide. For instance, patterns of insect herbivory probably change throughout the year as does the vulnerability of these insects to ants as a function of their life stages. Second, because of the way S. aurea workers forage, their usefulness in plant defense may in part be a function of the hosting plant's size and architecture.
Snelling and George (1979) write of this ant: "Although dietary preferences are unknown this species is probably typical of the fire ants, a generally omnivorous scavenger-predator with preferences for oily or fatty meats and seeds." Since S. aurea spends a significant portion of its colony life cycle below ground it is difficult to determine with any certainly its entire dietary breadth. Nevertheless, in a rather extensive study of the ant community in and around Portal, Arizona, Andersen (1997) was unable to attract any of these ants to his tuna baits (he did capture a few in pitfall traps). Likewise, except for a single instance, I never observed S. aurea workers foraging for or feeding on anything other than nectar or pollen. Below is a picture of the one exception. I found the ground squirrel in this picture at the base of a chainfruit cholla (the base of the plant is in the upper left-hand corner) that was being used by a S. aurea colony. Workers from this colony already covered the animal when I arrived and these workers, especially the majors, seemed very proficient at carving it up. Given this and the other observations above, it suggests that these ants will scavenge but that they do not do it far from their nests -- at least on the desert surface.
|
Dead ground squirrel being scavenged |
|
Click on image to enlarge it |
In all of the above I have talked about S. aurea colonies as if they were readily identifiable entities; unfortunately they weren't. Most often an EFN-bearing cactus received workers from a single nest which was usually barely discernible and at the base of the plant or short distance from it. While I have not analyzed the data I have on this point, my subjective impression is that adjacent EFN-bearing cacti were often taken over at about the same time. This suggests that single colonies frequently visit more than one plant but multiple colonies responding to similar micoenvironmental cues could also explain this observation. Unfortunately, the easy although admittedly crude test of putting ants from different plants together and seeing how they behaved toward one another didn't work; even ants from very distant plants didn't attack each other. This obviously does not rule out the possibility of developing a more detailed behavioral assay. I have also collected individuals with the hopes of one day using DNA microsatellites to get at this question.
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.
Cokendolpher, J.C. and O.F. Francke. 1985. Temperature Preferences of four species of fire ants (Hymenoptera: Formicidae: Solenopsis). Psyche 92: 91-101.
Creighton, W.S. 1950. The ants of North America. Bulletin, Museum of Comparative Zoology. Harvard College, 104: 1-585. 57 pl.
Francke, O.F., L.R. Potts, and J.C. Cokendolpher. 1985. Heat tolerance of four species of fire ants (Hymenoptera: Formicidae: Solenopsis). Southwestern Naturalist 30: 59-68.
Heatwole, H. and R. Muir. 1989. Seasonal and daily activity of ants in the pre-Saharan steppe of Tunisia. Journal of Arid Environments 16: 49-67.
Jensen, T. F. 1978. An energy budget for a field population of Formica pratensis. Natura Jutlandica 20: 203-236.
Rissing, S.W. 1987. Annual cycles in worker size of the seed-harvester ant Veromessor pergandei. Behavioral Ecology and Sociobiology 20: 117-124.
Snelling, R.R. and C. George. 1979. The taxonomy, distribution, and ecology of California desert ants. Report to Bureau of Land Management, United States Department of Interior, Riverside California. 335 p.
Tschinkel, W. R. Sociometry and sociogenesis of colonies of the fire ant Solenopsis invicta during one annual cycle. 1993. Ecological Monographs 63: 425-457.
Copyright (c) 1998 Barry Sullender
Rice University
All rights reserved