Behavioral Ecology Vol. 10 No. 4: 428-435
© 1999 International Society for Behavioral Ecology
Queen recruitment in a multiple-queen population of the fire ant Solenopsis invicta
a Department of Genetics, University of Georgia, Athens, GA 30602-7223, USA b Department of Entomology, University of Georgia, Athens, GA 30602-7223, USA
Address correspondence to M. A. D. Goodisman, who is now at the Department of Genetics, La Trobe University, Bundoora, VIC, 3083, Australia. E-mail: madgood{at}gen.latrobe.edu.au
Received 11 May 1998; revised 16 December 1998; accepted 6 January 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We assessed patterns of new queen recruitment in a polygyne (multiple queens per nest) population of the fire ant Solenopsis invicta in its introduced range. Newly recruited queens were identified using four physiological markers, and genotypic data from nuclear and mitochondrial markers were used to estimate relatedness of new nest mate queens to each other and to the older nest mate queens. In total, 1.7% of the queens collected in late spring and early summer were deemed to be new recruits. The relatedness of these queens to each other and to the older queens within nests was not significantly different from zero, suggesting that newly recruited queens represent a random sample of potential reproductive queens in the population. Moreover, the number of new queens recruited within nests was not correlated with the number of older queens present and did not differ significantly from a Poisson distribution. Thus, queen recruitment in this population of S. invicta appears to occur at random with respect to the number of older queens present within nests.
Key words: fire ants, polygyny, relatedness, social insects, Solenopsis invicta.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Multiple reproductive queens frequently coexist within single social insect colonies (Bourke and Franks, 1995
; Crozier and Pamilo,
1996;
Hölldobler
and Wilson, 1990; Keller,
1993; Ross and Carpenter,
1991). These multiple-queen (polygyne) colonies differ
substantially from conspecific colonies headed by a single reproductive queen
(monogyne colonies). In ants, for example, queen fecundity, dispersal, and
longevity, as well as nest mate relatedness, generally are reduced in polygyne
colonies (Bourke and Franks,
1995; Crozier and Pamilo,
1996;
Hölldobler
and Wilson, 1990). Knowledge of the proximate mechanisms by which
multiple queens come to inhabit a single nest is integral to understanding the
factors that lead to these important fitness-related differences between the
two types of colonies.
Most studies investigating new queen recruitment have used queen
introductions under laboratory conditions (e.g.,
Fletcher and Blum, 1983;
Keller and Ross, 1993a;
Stuart et al., 1993;
Sundström,
1997). These studies revealed that factors such as queen mating
status and nest of origin can be important in determining queen acceptance.
However, laboratory conditions do not necessarily mimic those found in the
field and thus may not always accurately represent natural processes. Field
analyses of queen recruitment, by circumventing this problem, offer a rarely
explored but valuable avenue of investigation into the dynamics of polygyne
colony sociogenesis.
The polygyne form of the introduced fire ant Solenopsis invicta is
an excellent candidate for assessing patterns of queen recruitment under
natural conditions. The social biology of S. invicta has been well
studied
(Hölldobler
and Wilson, 1990; Ross,
1993; Ross and Keller,
1995a; Vinson,
1997), facilitating the design and interpretation of meaningful
field studies. Moreover, the high colony queen number observed in introduced
polygyne populations (Goodisman and Ross,
1997; Ross, 1993;
Vargo and Fletcher, 1987)
implies that queen recruitment occurs at a fairly high rate and thus newly
recruited queens may be detectable under natural conditions (e.g.,
Glancey and Lofgren,
1988).
This study had three main goals. The first was to estimate the frequency of
new queen recruitment in an introduced, polygyne population of S.
invicta. This information is critical to understanding the patterns of
polygyne colony growth and may be used to estimate the life span of polygyne
queens. This second variable is particularly important to the study of queen
number in ants because the recruitment of multiple queens is often associated
with a decrease in queen longevity (Bourke
and Franks, 1995; Keller and
Genoud, 1997;
Hölldobler
and Wilson, 1990). Estimates of queen longevity in monogyne S.
invicta populations already exist
(Tschinkel, 1987), and a
corresponding estimate for polygyne queens would be of considerable
interest.
The second goal of this study was to estimate the relatedness of newly
recruited queens to each other and to the older queens within nests. Previous
studies of nest mate relatedness in introduced polygyne populations of S.
invicta revealed that queen relatedness was statistically
indistinguishable from zero (Goodisman and
Ross, 1997, 1998;
Ross, 1993;
Ross and Fletcher, 1985).
Nevertheless, the relatedness of newly recruited queens may differ from that
of older queens. For instance, elevated relatedness of newly recruited queens
would signal that they were not drawn at random from the pool of potential
reproductives in the population but were the offspring of one or a few older
queens in the colony. Moreover, this result would suggest that differential
queen mortality based on relatedness occurs during recruitment. This study
attempts to detect such features by comparing the relatedness of newly
recruited queens to that of older queens within the same colonies.
The final objective of this investigation was to describe colony-level
patterns of new queen recruitment. The relationship between the number of new
queens and the number of older queens within nests, as well as the
distribution of newly accepted queens across nests, may yield insights into
the mechanisms governing the recruitment process
(Crozier and Pamilo, 1996;
Nonacs, 1988;
Pamilo, 1991). This study
explicitly examines these important patterns under natural conditions.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Development of markers
To detect newly recruited queens (NRQs) within nests, it was first necessary to identify markers that distinguish NRQs from older queens. To find such markers, we compared 114 alate (winged), prereproductive queens with 45 dealate (wingless), egg-laying queens (wing shedding is associated with the onset of reproduction; Vargo and Laurel, 1994
). These ants originated from seven previously collected
polygyne nests that had been kept in the laboratory for at least 1 week and
were not used in other parts of this study. Because queens undergo substantial
physiological changes as they become reproductively active
(Hölldobler
and Wilson, 1990; Toom et al.,
1976a, b,
c;
Vander Meer et al., 1982), we
reasoned that NRQs would have characteristics intermediate between alate and
dealate queens. In total, we identified four markers useful in identifying
NRQs.
Activity at glycerol-3-phosphate dehydrogenase-1 locus
The enzyme product of the locus glycerol-3-phosphate
dehydrogenase-1 (G3pdh-1) supports flight in S.
invicta, and its activity diminishes after alate queens shed their wings
(Hung et al., 1979). We
assayed the activity of this enzyme visually, using protein electrophoresis
and standard staining techniques on samples of thoracic tissue from queens
(Shoemaker et al., 1992). The
banding patterns of alates became discernible approximately 30 s after adding
the requisite stain ingredients, and genotypes could be scored readily 5-10
min later. The enzyme activity of dealate queens was not nearly as strong.
Bands could first be seen 15 min after staining and were just strong enough to
score reliably after 45 min. The intensity of staining for G3pdh-1 in
dealates never reached the level of that in alates, even after the gels were
allowed to stain for more than 2 h. Thus, we expected G3pdh-1 band
intensity in NRQs to be substantially stronger than in older, dealate
queens.
Banding pattern from general protein stain
Protein electrophoresis on thoracic tissue revealed another marker that
differentiated alate and dealate queens. Using a nonspecific general protein
stain (Shoemaker et al.,
1992), we identified a single band, hereafter referred to as the
"alate protein band," that consistently appeared in the banding
profiles of alate queens but never in those of older dealate queens. We
expected the banding profile of NRQs to contain this band, although its
intensity may be reduced relative to that seen in alate queens that have not
flown or initiated oogenesis.
Condition of wing muscles
Ant queens histolyze their flight muscles to provide a source of energy
when initiating reproduction
(Hölldobler
and Wilson, 1990). The result of this process can be seen in
S. invicta by inspecting the thoracic cavity of queens
(Glancey et al., 1980;
Markin et al., 1972). The wing
muscles of alate queens comprise visible red fibrous strands. In contrast, the
thoraces of dealate queens contain a puffy, whitish substance, with little
evidence of muscle fibers. We expected NRQs to show slightly histolyzed wing
muscles.
Fat content of gasters
Ant queens accumulate fat after they eclose. This fat serves as an energy
source and is metabolized by the queens as they initiate reproduction
(Keller and Passera, 1989). We
used standard methods (Keller and Passera,
1988, 1989;
Keller and Ross, 1993b) to
measure the mass of stored fat in the gasters (abdomen minus propodeum) of
alate and dealate S. invicta queens. Gasters were dried for 24 h and
weighed. Fat was then extracted from the gasters with 4-5 ml of petroleum
ether (boiling point 40-60°C), and the gasters were dried and weighed
again. The difference between the weights before and after ether extraction
was taken to be the mass of stored fat in the gaster.
Alate queen gasters contained 1.20 ± 0.05 (&xmacr;
± SEM) mg of fat. [This value is lower than previously reported for
S. invicta alate queens (Keller
and Ross, 1993b), perhaps because the individuals used here had
not matured fully.] In contrast, fat accounted for only 0.29±0.03 mg in
dealate queen gasters. We found that a weight-standardized measure of gaster
fat content, defined as (mass of gaster before extraction — mass of
gaster after extraction)/(mass of gaster before extraction), yielded
substantial power in distinguishing the two classes of queens. After
standardization, the means for alate and dealate queen gaster fat content
transform to 0.49 ± 0.01 and 0.16 ± 0.01, respectively, and
differ significantly from one another (F1,158 = 295.61,
p <.0001). We anticipated that NRQs would show transformed fat
content values slightly below those of alate queens but still substantially
greater than older, dealate queens.
Time course of change in NRQ markers
We wished to assess the time span over which NRQs could be identified
within nests. That is, how long would these individuals bear the above
markings and be distinguishable from the older queens? To do this, we first
collected 155 polygyne dealate queens after mating flights
(DeHeer et al., 1999) and put
them into culture in the laboratory (as in
Bernasconi and Keller, 1996).
Queens were isolated in groups of about 10, and provided with food, water, and
approximately 30 worker pupae. We recognize that this scenario may differ from
the context in which polygyne queens normally initiate reproduction
(Glancey and Lofgren, 1988;
Porter, 1991;
Vargo and Porter, 1989),
however, introduction of queens directly into polygyne nests was not an option
because the majority of queens are eliminated by workers under laboratory
conditions (Keller and Ross,
1995). Queens were harvested over the course of the next 15 days
and scored for the 4 characteristics described above to monitor the change of
the markers over time. Two queens died over the course of the study and were
not included in the analysis.
Collection of samples
We collected 92 S. invicta nests on 5 different days from 6 sites
separated by no more than 2 km in Walton County, Georgia, USA. To maximize the
probability of discovering NRQs, nests were collected in the late spring and
early summer, which apparently represents the major period of mating activity
and queen recruitment in introduced S. invicta
(Glancey and Lofgren, 1988;
Vinson and Greenberg, 1986).
To capture the majority of queens within nests, collecting took place on warm,
sunny days that were preceded by rainfall. Nests were placed in buckets, and
the inhabitants were separated from the soil by flooding
(Jouvenaz et al., 1977) and
placed into large trays. Within 48 h, all dealate queens were collected by
searching through the ants in the trays. Nests were considered to be polygyne
if they contained two or more dealate queens.
Analysis of queens
A large proportion of queens remains permanently unmated in introduced
polygyne S. invicta populations
(Ross, 1993;
Ross and Keller, 1995b).
Therefore, we determined the mating status of queens by examining queen
spermathecae. Sperm in mated queens was evident as an opaque, whitish mass.
All queens were scored for activity at G3pdh-1, presence/absence of
the alate protein band, evidence of flight muscles, and standardized fat
content, as described above. Queens that matched expectations derived from the
alate/dealate comparisons and time course experiments for these markers were
considered to be NRQs.
We determined the multilocus genotypes of queens electrophoretically at
three polymorphic nuclear loci (Aat-2, Gp-9, and Pgm-1) in
addition to G3pdh-1 (Ross,
1997; Ross and Shoemaker,
1997; Shoemaker et al.,
1992). Whereas the loci Aat-2, Pgm-1, and
G3pdh-1 presumably are neutral
(Shoemaker et al., 1992),
Gp-9 (or a locus closely linked to it) apparently is under strong
selection in the polygyne form of S. invicta
(Ross, 1997). Alate queens and
workers display all three possible genotypes at this diallelic locus, but
reproductive queens are virtually always (99.9%) heterozygous
(Ross, 1997). Therefore, any
putative NRQs that were homozygous at Gp-9 were excluded from
analysis.
Queen genotypes also were assayed at a 4-kb region of the mitochondrial DNA
(mtDNA) (Ross and Shoemaker,
1997). The mtDNA haplotype was obtained by amplification with
polymerase chain reaction followed by digestion with the enzyme
HinfI, which differentiates among the three haplotypes (two common
and one rare) found in this polygyne population
(Ross and Shoemaker, 1997;
Shoemaker and Ross, 1996).
Genotypic data from the three neutral nuclear loci (Aat-2, Pgm-1,
and G3pdh-1) and the mtDNA were used in conjunction with the program
Relatedness 4.2 (Queller and Goodnight,
1989) to calculate the relatedness of nest mate queens. Genetic
differentiation among sites was accounted for by using the "deme"
function of the program. We used all captured queens to calculate the
population allele frequencies (P*), and groups (nests)
were always weighted equally. Standard errors for the relatedness estimates
were obtained by jackknifing over nests; a relatedness value was considered to
be statistically different from zero if the 95% confidence intervals of the
estimate (r ± 1.96 x SEM) did not overlap zero.
Within-nest correlations between the number of NRQs and the number of older queens were calculated using Spearman's correlation coefficient (rS). Significance of differences in number of NRQs among sites and among days was assessed with a G test of independence. A Kolmogorov-Smirnov goodness-of-fit test was used to determine if the numbers of new queens accepted into nests differed from a Poisson distribution.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Time course of change in NRQ markers
Activity of the product of G3pdh-1 in our experimental queens was noticeably greater than that of egg-laying, dealate controls for the first 7 days after mating. These results are consistent with previous studies of this enzyme in S. invicta (Hung et al., 1979
). The alate protein band showed a similar decay pattern
and was visible in queens through day 6 of isolation. The wing muscles of
experimental queens were notably different from those of mature, dealate
queens for approximately 5 days after the mating flight. Subsequently, the
contents of the thoracic cavity varied and became difficult to consistently
distinguish from dealate controls. The rate of histolysis in this study
apparently was slightly greater than that reported in earlier studies
(Glancey et al., 1980;
Markin et al., 1972). This
difference may be attributable to the fact that polygyne queens were used in
this experiment, whereas monogyne queens were used in the earlier
analyses. Standardized fat content decreased linearly during the time-course experiment (r2 =.20, F1,151 = 38.56, p <.0001), with the relationship between fat content and time described by the equation: fat content = -0.013 x day + 0.507. An average experimental queen reached the upper 95% limit (&xmacr; + 1.96 x SD = 0.32) for the approximately normal distribution of fat content in older dealate queens in about 14 days. However, if the linear trend continued, the mean of the experimental queens would not equal that of the dealate controls (0.16) for approximately 27 days.
We found no noticeable differences in the rate of decay of the four markers between mated and unmated queens (data not shown). When considering all the markers together, we conclude that NRQs can be differentiated from mature queens for roughly 1 week after shedding their wings and initiating reproduction.
Distribution and characteristics of nests and queens
Of the 92 nests collected, 85 were considered to be polygyne, whereas the
remaining 7 contained either 1 or no queens and thus were not included in
other analyses. Table 1 shows
the sampling dates, number of nests, and number of queens collected from each
site and the entire population. Queen number per nest varied widely (mean =
30.55, SD = 41.46, range 2-257), although it did not differ significantly
among sites (F5,79 = 2.08, p >.05). Overall,
the frequency of mated queens (0.488) was relatively low, but close to that
observed at a nearby site in a previous study
(Ross and Keller, 1995b).
|
Distribution and characteristics of NRQs
In total, 45 queens from 25 of the nests showed characteristics consistent
with being NRQs. However, two of these individuals were homozygous at the
locus Gp-9 and thus would not have become permanent reproductives
(Ross, 1997); therefore, they
were eliminated from further study, leaving a total of 43 NRQs from 24
nests.
NRQs accounted for 1.7% of all queens collected. Only 5 of the 43 NRQs (11.6%) could be confirmed as having mated, a proportion that is significantly lower than for the older queens in the population (z = 63.49, p <.0001). Another notable point is that NRQs were significantly lighter than the mature queens in our sample (two-way ANOVA: colony, F84,2487 = 16.39, p <.0001; queen status, F1,2487 = 36.79, p <.0001; interaction, F23,2487 = 1.36, p >.1). This result suggests that the fecundity of mature, reproductive queens increases after the recruitment period or that lighter NRQs undergo higher rates of mortality after initial acceptance.
Relatedness of nest mate queens
In accord with previous results
(Goodisman and Ross, 1997;
Ross, 1993;
Ross and Fletcher, 1985), the
relatedness of the older nest mate queens was indistinguishable from zero (95%
confidence intervals overlap with zero) when either nuclear or mitochondrial
markers are considered (Figure
1). Of specific interest for this study, however, are the
relatedness estimates of the NRQs to each other and to the older queens.
Except for the marginally significant, negative mitochondrial estimate for
NRQs, these values do not differ significantly from zero.
|
Patterns of new queen recruitment
The proportions of NRQs differed among the six sites (G test of
independence, G5 = 24.53, p <.001) and among
the 5 days on which samples were collected (G test of independence,
G4 = 23.82, p <.001). Because nests were
collected from most of the sites on only 1 day, the effects of site and day
are confounded and cannot be distinguished from one another. However, there is
no obvious trend between proportion of NRQs and day collected
(Table 1).
Spearman's correlation coefficient was used to investigate the relationship between the number of older queens and number of NRQs within nests (Figure 2). The correlation between the two variables is not significant if all nests are included in the analysis (rS =.004, n = 85, p >.5) nor if only nests that recruited new queens are considered (rS = -.125, n = 24, p >.5). Thus, our data cannot reject the hypothesis that NRQs are recruited into nests at random with respect to the number of older queens present. We also tested whether the distribution of NRQs per colony differed from that expected under a Poisson distribution. A total of 61, 11, 10, 2, 0, 0, and 1 nests accepted 0, 1, 2, 3, 4, 5, and 6 NRQs, respectively. This distribution does not differ significantly from a Poisson distribution with a mean of 0.51 (Kolmogorov-Smirnov goodness of fit test, z = 1.06, p >.2).
|
Confirmation of NRQ status
The markers developed to identify NRQs would likely misidentify as
legitimate NRQs prereproductive queens that had lost their wings accidentally
due to the trauma of the collection process. These "traumatic
dealates," if present in substantial numbers, could substantially alter
our conclusions regarding queen recruitment. Three tests can be conducted to
evaluate the importance of this potential problem.
Standardized fat content of gasters
If the sample of NRQs includes a large proportion of traumatic dealates,
then the standardized fat content of the sample should approach that of alate
controls. However, if the putative NRQs actually are new reproductive queens,
then the standardized fat content of their gasters should be significantly
less than that of the alate controls. We found that the mean standardized fat
content of the NRQs (0.38±0.01) is significantly below that of alate
controls (0.49±0.01, t155 = 5.69, p
<.0001) and significantly above that of dealate controls (0.16±0.01,
t87 = 55.88, p <.0001). This intermediate fat
content is consistent with the NRQ designation.
Relatedness of NRQs
Nest mate alates in introduced polygyne S. invicta populations
exhibit significant positive relatedness to one another
(Ross, 1993). Therefore, if a
large fraction of putative NRQs in our sample were traumatic dealates,
estimates of relatedness between NRQs would be expected to be greater than
zero. Moreover, significant positive relatedness of traumatic dealates to the
older, nest mate queens that would have produced them might also be expected.
In contrast, if the individuals we identified as NRQs are indeed new queens,
then their relatedness to each other and to the older queens would be expected
to be indistinguishable from zero, as has repeatedly been found to be the case
for reproductive queens in this population
(Goodisman and Ross, 1997,
1998;
Ross, 1993;
Ross and Fletcher, 1985). From
Figure 1 we see that none of
the nestmate queen relatedness estimates involving NRQs are significantly
greater than zero. Thus, these relatedness estimates are consistent with the
NRQ designation.
Frequency of wing breakage
To assess the probability of traumatic dealation during the collecting
process, 583 alate queens that were incidentally collected from 8 of the nests
for this study were examined to see if one or more of their 4 wings were
missing. If traumatic dealation occurred frequently, then many of these alates
would have fewer than four wings.
Inspection of the alates revealed that 571, 10, 2, and 0 had lost 0, 1, 2, and 3 of their wings, respectively. Using maximum likelihood methods, we found that the distribution of wing breakage matches a binomial distribution (exact test, p >.3) where the probability of losing any single wing is 0.006. Thus, the probability of a queen losing all four of her wings through traumatic dealation is (0.006)4 = 1.3x10-9. It therefore is unlikely that any of the queens designated as NRQs were traumatic dealates.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We studied the process of queen recruitment in a wild population of introduced polygyne S. invicta. Newly recruited queens constituted 1.7% of the total number of queens sampled during late spring and early summer. Moreover, the relatedness of the NRQs to one another and to the older queens in their nests was statistically indistinguishable from zero when both nuclear and mitochondrial markers were considered. We found no significant correlation between the number of NRQs and the number of older queens within nests, and the distribution of new queens within nests fit a Poisson distribution. Thus, we conclude that NRQs are accepted into nests at random with respect to existing queen number.
Frequency of recruitment and queen longevity
The frequency of NRQs observed in this study can be used to estimate queen
longevity within natural polygyne populations. Such calculations require
simplifying assumptions and therefore should be taken as providing only rough
estimates. We first suppose that our study population is at equilibrium with
respect to recruitment rate and queen number. We next assume that all NRQs
replace queens that died and that mortality follows an exponential
distribution (Lee, 1992). From
our laboratory studies, NRQs could be identified for approximately 1 week, and
so 1.7% of the queens within this population die per week. If this rate is
constant throughout the year, then the mean life span of a polygyne S.
invicta queen is 58.8 (1/0.017) weeks, or approximately 1 year. However,
if recruitment only occurs during the period of substantial alate production,
which lasts from May through September in this population
(Vargo and Fletcher, 1987),
then the expected life span increases to almost 3 years. Our results are
consistent with a previous laboratory study of polygyne queens in this
species. Ross (1988) found
that approximately 40% of polygyne queens established in laboratory colonies
died in less than 1 year. If mortality occurred at a constant rate, then the
life expectancy of polygyne S. invicta queens from this study is
1/0.4 = 2.5 years.
Tschinkel (1987) estimated
the typical life span of successful colony-founding monogyne S.
invicta queens as ranging from 5 to 7 years. The lower estimated life
span of polygyne queens in this species conforms to previous results from
other species (Keller and Genoud,
1997; Keller and Passera,
1990) and expectations under life-history theory; polygyne queens
likely produce sexual offspring at an earlier age than monogyne queens, and a
shorter life span is generally associated with earlier reproduction
(Bourke and Franks, 1995,
Keller and Genoud, 1997).
Relatedness of NRQs
Our relatedness estimates for NRQs indicate that they are unrelated to each
other and to the older queens in the nest. These results are consistent with
previous studies in polygyne S. invicta in the introduced range that
showed that nest mate queens are no more closely related to one another than
they are to queens in other nests
(Goodisman and Ross, 1997,
1998;
Ross, 1993;
Ross and Fletcher, 1985;
Ross et al., 1996). Thus, the
NRQs appear to represent a random sample of the potential reproductive queens
in the population.
A laboratory study of Formica truncorum
(Sundström,
1997) and a field study of F. lugubris
(Fortelius et al., 1993)
reported that polygyne workers do not discriminate between nest mate and
non—nest-mate females that are introduced into colonies. However,
polygyne queens of F. truncorum are related
(Sundström,
1993), indicating that such discrimination may occur under natural
conditions or that queen movement between nests is substantially restricted. A
similar experiment in polygyne Leptothorax curvispinosus found that
queens were more likely to be adopted if they were introduced into their natal
nest (Stuart et al., 1993).
Like F. truncorum, nest mate queens in this species tend to be
related (Stuart et al., 1993),
and thus the recruitment process likely displays substantial differences from
that in the introduced fire ant.
Patterns of recruitment
The patterns of queen recruitment observed in this study may be used to
test hypotheses concerning the development of polygyne nests. For instance, if
queen number within each nest remains approximately the same over time, then
we would expect a negative relationship between the number of NRQs and number
of older queens within nests. This correlation should arise because all nests
would only be replacing queens that had died, and nests with many queens would
need to replace more queens than those with fewer queens. However, our data do
not reflect this pattern. Rather, our results are consistent with nests
accepting NRQs at random with respect to the number of older queens present.
Thus, queen number within nests appears to be subject to variation over
time.
It is somewhat surprising that the number of older queens within nests does
not influence new queen recruitment. Levels of queen pheromones within
polygyne S. invicta nests, which appear to be related to queen
number, play otherwise important roles in the regulation of reproduction. For
example, queen-derived pheromones suppress the production of sexuals, the
initiation of reproduction by virgin queens, and queen fecundity
(Vargo, 1992;
Vargo and Fletcher, 1986a,
b,
1987,
1989;
Vargo and Laurel, 1994).
Therefore, it might be expected that the high levels of pheromone associated
with many queens would induce workers to reject additional queens. If
recruitment in introduced polygyne S. invicta nests occurs at random,
then queen number within nests potentially can increase without bound,
explaining the high queen numbers observed in some nests in older polygyne
populations (Goodisman and Ross,
1997, this study; Ross,
1993; Ross et al.,
1996).
The fraction of NRQs within our study nests did not show any consistent
change over the course of the reproductive season
(Table 1). This finding differs
from that of Fortelius et al.
(1993), who noted that queen
recruitment in F. lugubris decreased through the flight season. It is
possible that the recruitment biology of S. invicta differs from
F. lugubris. Also, this study may have detected too few NRQs or
extended over an insufficient period of time to discern any seasonal
pattern.
Matedness in NRQs
A notable feature of our data set is that older queens were mated
significantly more often than were NRQs. One possible explanation for this
result is that unmated queens are more successful at entering polygyne nests
than mated queens. Indeed, two studies of polygyne Formica ants
reported such a result (Fortelius et al.,
1993;
Sundström,
1997). However, if polygyne S. invicta nests accepted and
retained unmated and mated queens at the ratio observed in the new recruits,
then the proportion of older mated queens in the population should be
substantially lower than is observed.
It is possible that unmated NRQs mate intranidally after entering nests. However, the proportion of mated queens collected on the ground immediately after mating flights in this study population is similar to the proportion of mated, older queens within nests, suggesting that most mating occurs during the flight and not after queens enter an established colony (Goodisman M, DeHeer C, Ross K, unpublished data).
Another explanation proposes that workers eliminate mated and unmated
queens at different rates (most queens attempting to enter polygyne S.
invicta nests are killed; Keller and
Ross, 1993a). That is, many of the mated queens that attempt to
enter nests are rapidly eliminated, and many of the unmated queens initially
observed also would be killed over time. A related hypothesis is that the
markers we used to detect NRQs decay more slowly in unmated queens than in
mated queens. Thus, newly recruited mated queens become indistinguishable from
older queens sooner than their unmated counterparts. Such hypotheses are
supported by studies that show that mating helps trigger physiological and
behavioral changes that mark the onset of oogenesis
(Cupp et al., 1973;
Toom et al., 1976b). However,
our experimental assessment of the rate of decay of markers did not reveal any
notable differences between mated and unmated queens (data not shown).
Comparison of recruitment patterns to those found in a previous
study
The results of this study differ substantially from those of a related
study in a different introduced population of polygyne S. invicta.
Glancey and Lofgren (1988)
found that 10 of 16 confirmed polygyne nests recruited new queens, and 51.0%
of the total queens showed characteristics consistent with being NRQs. This
proportion of NRQs is significantly greater than that found in our study
(1.7%, z = 30.19, p <.0001). The contrasting results are
not likely due to variation in the reliability of markers for new queens in
the two studies because a common marker was used, nor are they likely due to
differences in season of sampling because collections were made in the late
spring in both cases. Rather, the contrasting proportions of NRQs detected may
be due to differences in such demographic features as the density or age of
nests in the two populations.
Despite the difference in the proportion of NRQs observed, this previous study displays important qualitative consistencies with some results from the present investigation. For instance, the correlation between the number of older queens present in polygyne nests and the number of NRQs does not differ significantly from zero (rs = -.022, n = 16, p >.9) for the Glancey and Lofgren data. Moreover, although the distribution of NRQs among nests in the Glancey and Lofgren study differs significantly from a Poisson distribution (Kolmogorov-Smirnov z = 1.96, p <.001), it does not differ significantly from a normal distribution (z = 1.27, p >.05) with estimated mean and standard deviation of 6.19 and 12.76, respectively. The fit of number of NRQs in nests to random distributions (the Poisson or the normal) in the two studies is consistent with the hypothesis that queens in both populations may be entering nests at random and without regard to the number of older queens present.
Conclusions
Three hypotheses have been put forward to explain the acceptance of
multiple queens into established social insect colonies: mutualism, kin
selection, and parasitism (Crozier and
Pamilo, 1996; Nonacs,
1988; Pamilo,
1991). Under the mutualistic hypothesis, cooperation leads to an
increase in the personal fitness of the NRQs and the older queens within
nests. Although our data cannot formally reject this hypothesis, mutualism
probably cannot explain queen recruitment in polygyne colonies in introduced
S. invicta because increasing queen number decreases both queen
reproduction (Vargo, 1992;
Vargo and Fletcher, 1986b,
1987,
1989) and queen longevity
(Bourke and Franks, 1995;
Keller and Genoud, 1997;
Tschinkel, 1987; this study).
Kin selection hypotheses also are unlikely to explain the acceptance of
multiple queens in introduced populations. In contrast to native South
American populations (Ross et al.,
1996), queens within introduced polygyne S. invicta nests
are statistically unrelated to one another
(Goodisman and Ross, 1997,
1998;
Ross and Fletcher, 1985;
Ross, 1993), and so existing
queens likely gain no fitness benefits by allowing new queens to enter. In
addition, workers in introduced populations do not appear to favor related
sexuals in their nest (DeHeer and Ross,
1997), so that nepotism does not seem to rescue inclusive fitness
benefits for polygyne workers.
Parasitism of preexisting nests by unrelated queens appears to be the best
explanation for the acceptance of multiple queens in introduced polygyne
populations of S. invicta. The density of nests in North American
populations greatly exceeds that in native South American populations
(Porter et al., 1992,
1997). This change may have
led to the occupation of most available S. invicta nesting sites,
which would in turn create selective pressure on prereproductive queens to
enter established nests (Crozier and
Pamilo, 1996; Elmes,
1973; Herbers,
1986, 1993;
Nonacs, 1988;
Pamilo, 1991;
Rosengren et al., 1993;
Ross et al., 1996). The number
of queens attempting to enter nests may be so high that recognition cues
become confused and workers are unable to prevent many queens from joining the
colony. Therefore, in this species, a proximate environmental change appears
to have dramatically altered the causes for the acceptance of multiple
reproductive queens.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We thank M. Chapuisat, C. DeHeer, J. Evans, M. Mescher, D. Promislow, and two anonymous reviewers for helpful discussion and comments on an earlier draft, and D. Promislow for use of the analytical balance. This research was supported in part by Environmental Protection Agency Graduate Fellowship U-914967-01-0 (to M.A.D.G.), the Georgia Agricultural Experiment Stations of the University of Georgia, and the National Geographic Society.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Bernasconi G, Keller L, 1996. Reproductive conflicts in cooperative associations of fire ant queens (Solenopsis invicta).Proc R Soc Lond B 263:509-513.
Bourke AFG, Franks NR, 1995. Social evolution in ants. Princeton, New Jersey: Princeton University Press.
Crozier RH, Pamilo P, 1996. Evolution of social insect colonies: sex allocation and kin selection. Oxford: Oxford University Press.
Cupp EW, O'Neal J, Kearney G, Markin GP, 1973. Forced copulation of imported fire ant reproductives. Ann Entomol Soc Am 66:743-745.[Web of Science]
DeHeer CJ, Goodisman MAD, Ross KG, 1999. Simple genetic basis for a dispersal polymorphism in polygyne Solenopsis invicta. Am Nat (in press).
DeHeer CJ, Ross KG, 1997. Lack of detectable nepotism in multiple-queen colonies of the fire ant Solenopsis invicta (Hymenoptera: Formicidae). Behav Ecol Sociobiol 40:27-33.
Elmes GW, 1973. Observations on the density of queens in natural colonies of Myrmica rubra L. (Hymenoptera: Formicidae).J Anim Ecol 42:761-771.
Fletcher DJC, Blum M, 1983. Regulation of queen number
by workers in colonies of social insects. Science
219:312-314.
Fortelius W, Rosengren R, Cherix D, Chautems D, 1993. Queen recruitment in a highly polygynous supercolony of Formica lugubris (Hymenoptera, Formicidae). Oikos 67:193-200.[Web of Science]
Glancey BM, Glover A, Lofgren CS, 1980. The thoracic crop in Solenopsis invicta Buren (Hymenoptera: Formicidae).Sociobiology 5:272 -278.
Glancey BM, Lofgren CS, 1988. Adoption of newly-mated queens: a mechanism for proliferation and perpetuation of polygynous red imported fire ants, Solenopsis invicta Buren. Fla Entomol 71:581-587.
Goodisman MAD, Ross KG, 1997. Relationship of queen number and queen relatedness in multiple-queen colonies of the fire ant Solenopsis invicta. Ecol Entomol 22:150-157.
Goodisman MAD, Ross KG, 1998. A test of queen recruitment models using nuclear and mitochondrial markers in the fire ant Solenopsis invicta. Evolution 52:1416-1422.[Web of Science]
Herbers JM, 1986. Nest site limitation and facultative polygyny in the ant Leptothorax longispinosus. Behav Ecol Sociobiol 19:115-122.
Herbers JM, 1993. Ecological determinants of queen number in ants. In: Queen number and sociality in insects (Keller L, ed). Oxford: Oxford University Press;262 -293.
Hölldobler B, Wilson EO,1990 . The ants. Cambridge, Massachusetts: Belknap Press.
Hung ACF, Dowler MG, Vinson SB, 1979. Electrophoretic
variants of 
-glycerophosphate dehydrogenase in the fire ant
Solenopsis invicta. Can J Genet Cytol
21:537-542.[Web of Science]
Jouvenaz DP, Allen GE, Banks WA, Wojcik DP, 1977. A survey for pathogens of fire ants, Solenopsis spp., in the southeastern United States. Fla Entomol 60:275-279.
Keller L, 1993. Queen number and sociality in insects. Oxford: Oxford University Press.
Keller L, Genoud M, 1997. Extraordinary lifespans in ants: a test of evolutionary theories of ageing. Nature 389:958-960.[Web of Science]
Keller L, Passera L, 1988. Energy investment in gynes of the Argentine ant Iridomyrmex humilis (Mayr) in relation to the mode of colony founding in ants (Hymenoptera:Formicidae). Intl J Invert Reprod Dev 13:31-38.
Keller L, Passera L, 1989. Size and fat content of gynes in relation to the mode of colony founding in ants (Hymenoptera; Formicidae). Oecologia 80:236-240.[Web of Science]
Keller L, Passera L, 1990. Fecundity of ant queens in relation to their age and the mode of colony founding. Insect Soc 37:116-130.
Keller L, Ross KG, 1993a. Phenotypic basis of
reproductive success in a social insect: genetic and social determinants.Science
260:1107-1110.
Keller L, Ross KG, 1993b. Phenotypic plasticity and "cultural transmission" of alternative social organizations in the fire ant Solenopsis invicta. Behav Ecol Sociobiol 33:121-129.[Web of Science]
Keller L, Ross KG, 1995. Gene by environment interaction: effects of a single gene and social environment on reproductive phenotypes of fire ant queens. Funct Ecol 9:667-676.
Lee ET, 1992. Statistical methods for survival data analysis. New York: John Wiley and Sons.
Markin GP, Collins HL, Dillier JH, 1972. Colony founding by queens of the red imported fire ant, Solenopsis invicta.Ann Entomol Soc Am 65:1053-1058.[Web of Science]
Nonacs P, 1988. Queen number in colonies of social Hymenoptera as a kin-selected adaptation. Evolution 42:566-580.[Web of Science]
Pamilo P, 1991. Evolution of colony characteristics in social insects. II. Number of reproductive individuals. Am Nat 138:412-433.[Web of Science]
Porter SD, 1991. Origins of new queens in polygyne red imported fire ant colonies (Hymenoptera: Formicidae). J Entomol Sci 26:474-478.
Porter SD, Fowler HG, Mackay WP, 1992. Fire ant mound densities in the United States and Brazil (Hymenoptera: Formicidae). J Econ Entomol 85:1154-1161.
Porter SD, Williams DF, Patterson RS, Fowler HG, 1997. Intercontinental differences in the abundance of Solenopsis fire ants (Hymenoptera: Formicidae): escape from natural enemies? Environ Entomol 26:373-384.[Web of Science]
Queller DC, Goodnight KF, 1989. Estimating relatedness using genetic markers. Evolution 43:258-275.[Web of Science]
Rosengren R, Sundström L, Fortelius W, 1993. Monogyny and polygyny in Formica ants: the result of alternative dispersal tactics. In: Queen number and sociality in insects (Keller L, ed). Oxford: Oxford University Press;308 -333.
Ross KG, 1988. Differential reproduction in multiple-queen colonies of the fire ant Solenopsis invicta (Hymenoptera: Formicidae). Behav Ecol Sociobiol 23:341-355.[Web of Science]
Ross KG, 1993. The breeding system of the fire ant Solenopsis invicta: effects on colony genetic structure. Am Nat 141:554-576.[Web of Science]
Ross KG, 1997. Multilocus evolution in fire ants: Effects of selection, gene flow and recombination. Genetics 145:961-974.[Abstract]
Ross KG, Carpenter JM, 1991. Phylogenetic analysis and the evolution of queen number in eusocial Hymenoptera. J Evol Biol 4:117-130.
Ross KG, Fletcher DJC, 1985. Comparative study of genetic and social structure in two forms of the fire ant Solenopsis invicta (Hymenoptera: Formicidae). Behav Ecol Sociobiol 17:349-356.[Web of Science]
Ross KG, Keller L, 1995a. Ecology and evolution of social organization: insights from fire ants and other highly eusocial insects. Annu Rev Ecol Syst 26:631-656.[Web of Science]
Ross KG, Keller L, 1995b. Joint influence of gene flow and selection on a reproductively important genetic polymorphism in the fire ant Solenopsis invicta. Am Nat 146:325-348.[Web of Science]
Ross KG, Shoemaker DD, 1997. Nuclear and mitochondrial genetic structure in two social forms of the fire ant Solenopsis invicta: insights into transitions to an alternate social organization.Heredity 78:590-602.[Web of Science]
Ross KG, Vargo EL, Keller L, 1996. Social evolution in
a new environment: the case of introduced fire ants. Proc Natl Acad Sci
USA
93:3021-3025.
Shoemaker DD, Costa JT, Ross KG, 1992. Estimates of heterozygosity in two social insects using a large number of electrophoretic markers. Heredity 69:573-582.[Web of Science]
Shoemaker DD, Ross KG, 1996. Effects of social organization on gene flow in the fire ant Solenopsis invicta.Nature 383:613-616.
Stuart RJ, Gresham-Bissett L, Alloway TM, 1993. Queen
adoption in the polygynous and polydomous ant, Leptothorax
curvispinosus. Behav Ecol
4:276-281.
Sundström L, 1993. Genetic population structure and sociogenetic organisation in Formica truncorum (Hymenoptera; Formicidae). Behav Ecol Sociobiol 33:345-354.[Web of Science]
Sundström L, 1997. Queen acceptance and nestmate recognition in monogyne and polygyne colonies of the ant Formica truncorum. Anim Behav 53:499-510.
Toom PM, Cupp EW Johnson CP, 1976a. Amino acid changes in newly inseminated queens of Solenopsis invicta. Insect Biochem 6:327-331.
Toom PM, Cupp E, Johnson CP Griffin I, 1976b. Utilization of body reserves for minim brood development by queens of the imported fire ant, Solenopsis invicta. J Insect Physiol 22:217-220.
Toom PM, Johnson CP, Cupp EW, 1976c. Utilization of body reserves during preoviposition activity by Solenopsis invicta.Ann Entomol Soc Am 69:145-148.[Web of Science]
Tschinkel WR, 1987. Fire ant queen longevity and age: Estimation by sperm depletion. Ann Entomol Soc Am 80:263-266.[Web of Science]
Vander Meer RK, Glancey BM, Lofgren CS, 1982. Biochemical changes in the crop, oesophagus and postpharyngeal gland of colony-founding red imported fire ant queens (Solenopsis invicta).Insect Biochem 12:123-127.
Vargo EL, 1992. Mutual pheromonal inhibition among queens in polygyne colonies of the fire ant Solenopsis invicta.Behav Ecol Sociobiol 31:205-210.[Web of Science]
Vargo EL, Fletcher DJC, 1986a. Evidence of pheromonal queen control over the production of male and female sexuals in the fire ant, Solenopsis invicta. J Comp Physiol A 159:741-749.
Vargo EL, Fletcher DJC, 1986b. Queen number and the production of sexuals in the fire ant, Solenopsis invicta (Hymenoptera: Formicidae). Behav Ecol Sociobiol 19:41-47.
Vargo EL, Fletcher DJC, 1987. Effect of queen number on the production of sexuals in natural populations of the fire ant, Solenopsis invicta. Physiol Entomol 12:109-116.
Vargo EL, Fletcher DJC, 1989. On the relationship between queen number and fecundity in polygyne colonies of the fire ant Solenopsis invicta. Physiol Entomol 14:223-232.
Vargo EL, Laurel M, 1994. Studies on the mode of action of a queen primer pheromone of the fire ant Solenopsis invicta. J Insect Physiol 40:601-610.
Vargo EL, Porter SD, 1989. Colony reproduction by budding in the polygyne form of Solenopsis invicta (Hymenoptera: Formicidae). Ann Entomol Soc Am 82:307-313.[Web of Science]
Vinson SB, 1997. Invasion of the red imported fire ant (Hymenoptera: Formicidae). Am Entomol 43:23-39.
Vinson SB, Greenberg L, 1986. The biology, physiology, and ecology of imported fire ants. In: Economic impact and control of social insects (Vinson SB ed). New York: Praeger;193 -226.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. G. Ross, M. J. B. Krieger, and D. D. Shoemaker Alternative Genetic Foundations for a Key Social Polymorphism in Fire Ants Genetics, December 1, 2003; 165(4): 1853 - 1867. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||