POPULATION ECOLOGY Survival of Imported (: Formicidae) Species Subjected to Freezing and Near-Freezing Temperatures

1 2 SHANNON S. JAMES, ROBERTO M. PEREIRA, KAREN M. VAIL, AND BONNIE H. OWNLEY

Department of Entomology and Plant Pathology, The University of Tennessee, 205 Ellington Plant Science Building, Knoxville, TN 37996

Environ. Entomol. 31(1): 127Ð133 (2002) ABSTRACT Survival at low temperatures is an important parameter determining distribution of imported Þre in the . Supercooling points and survival at low temperatures, and the effects of species, individual size, and Thelohania solenopsae Knell, Allen & Hazard (Microsporida: Thelohaniidae) infection on these parameters, were examined. We tested Solenopsis richteri Forel, S. richteri X invicta hybrid, and Solenopsis invicta Buren. Great variation was observed in the supercooling points, which are not an appropriate measure of cold hardiness for imported Þre ants. When exposed to near-freezing temperatures above their supercooling points, Þre ants died at different rates depending on the species and T. solenopsae-infection status. Extended exposure to 4ЊC resulted in both the hybrid and S. invicta infected with T. solenopsae having signiÞcantly lower mortality rates than either the S. richteri or the uninfected S. invicta. At 0.5ЊC, the hybrids had signiÞcantly lower mortality than the uninfected S. invicta, but mortalities for S. richteri and T. solenopsae-infected S. invicta were not signiÞcantly different from each other or the hybrid. Ant mortality was 100% for all ant types after 7datϪ4ЊC. The uninfected S. invicta was consistently less cold-tolerant than the other ant types. The hybrid Þre ants and the T. solenopsae-infected S. invicta had the lowest mortalities. These results support the hypothesis that extended cold injury causes winter kill of Þre ants, and may partially explain the distribution of Þre ant species in the United States.

KEY WORDS Solenopsis invicta, Solenopsis richteri, Thelohania solenopsae, red imported Þre ant, black imported Þre ant, hybrid Þre ant

ORIGINALLY FROM , imported Þre ants ing Þre ant adults are nondormant and nonfreeze- are well known pests throughout the southern United tolerant (Francke and Cokendolpher 1986). Although States and are found in 13 states and Puerto Rico, imported Þre ants may be active and forage on warm including 29 counties along TennesseeÕs southern bor- days during winter, they cease activities outside the der. The black imported Þre ant, Solenopsis richteri mound at 10Ð15ЊC (Porter and Tschinkel 1987, Cok- Forel, is found in the northern portion of this range endolpher and Phillips 1990). Imported Þre ants were (Callcott and Collins 1996), an area approximately not expected to infest areas north of the Ð12ЊC min- Њ from 34.5 to 35.5 N (Shoemaker et al. 1996; R. Vander imum January isotherm (DifÞe et al. 1997), which runs Meer, unpublished data) covering parts of northern Ϸ100 km south of TennesseeÕs southern border. How- Mississippi and , and in southern Tennessee ever, the expansion of imported Þre ants beyond this from Shelby County to Giles County. The red im- isotherm has led to speculation on the Þnal limits of ported Þre ant, S. invicta Buren, occurs throughout the imported Þre ant range in the United States. southern portion of the United States. The red/black Imported Þre ants have a well developed sensitivity hybrid, which is not found in South America, occurs in a band between the parent species east of the to temperature and humidity gradients (Porter and Mississippi (Callcott and Collins 1996, Shoemaker et Tschinkel 1987, Cokendolpher and Phillips 1990), par- al. 1996, Drees et al. 2000). ticularly when tending brood (Cokendolpher and Originating from tropical and subtropical regions, Francke 1985, Porter and Tschinkel 1993). Imported imported Þre ants were not expected to have mech- Þre ants move within the column of the nest anisms to survive extreme cold weather. Overwinter- depending on daily temperature changes (Pinson 1980), going deeper in the mounds to avoid cold (Mor- rill et al. 1978). Other factors, such as colony size 1 Current address: Florida Keys Mosquito Control District, 5224 (Green 1959, Callcott et al. 2000), clustering behavior College Road, Key West, FL 33040. 2 Current address: USDAÐARS, CMAVE, P.O. Box 14565, Gaines- (Kaspari and Vargo 1995), and nest construction/lo- ville, FL 32608 (e-mail: [email protected]ß.edu). cation (Hubbard and Cunningham 1977, Morrill et al.

0046-225X/02/0127Ð0133$02.00/0 ᭧ 2002 Entomological Society of America 128 ENVIRONMENTAL ENTOMOLOGY Vol. 31, no. 1

1978), may also play an important role in imported Þre abaeidae), Galleria mellonella L. (Lepidoptera: Pyrali- ant winter survival. dae), and Periplaneta americana L. (Blattaria: The northward expansion of the imported Þre ant Blattidae)]. Gas chromatographic analysis of range, the lack of correlation between supercooling and cuticular hydrocarbons (Vander Meer et al. 1985) points and the location of imported Þre ant species, was used to identify types at the USDAÐARS, CMAVE and the lack of information on imported Þre ant ability laboratory in Gainesville, FL. to survive cold, nonfreezing temperatures, led us to Supercooling Tests. Five colonies of each ant type study imported Þre ant survival at low temperatures. were selected for each test. Ten female alates (when Also, we were interested in the effects of the recently available), small (head capsule Ͻ0.9 mm) and large introduced biological control agent Thelohania so- (head capsule Ͼ0.9 mm) workers were selected at lenopsae Knell, Allen & Hazard (Microsporida: The- random from each colony and individually frozen to lohaniidae) (Williams et al. 1999) on the imported Þre collect supercooling point data. Worker ants were ant ability to survive these low temperatures. The selected from the smallest and largest ants in each objectives of this research were to determine low colony. Based on preliminary data, supercooling point temperature survival of S. invicta, S. richteri, and the S. variation increased after a month or more exposure to richteri X invicta hybrid, and to examine the effects of laboratory conditions, thus ants were tested as soon seasonal acclimation and T. solenopsae infection on after collection as possible (1 wk for Tennessee col- imported Þre ant tolerance to low temperatures. onies, 2wk for Florida colonies). Ants collected in fall 1999 also were tested after 3 mo in the laboratory to Materials and Methods determine the effects of maintenance under labora- tory conditions. Testing of large and small workers (10 Solenopsis richteri, uninfected S. invicta, T. solenop- ants for each size from Þve colonies) using both 0.01 sae-infected S. invicta, and the S. richteri X invicta and 0.1-mm thermocouples was conducted to evaluate hybrid are herein called ant “types” so the improper the effect of thermocouple size on the supercooling term “species” would be avoided when referring col- point readings. lectively to all ants tested. To measure the supercooling point, ants were at- Imported Colonies. S. richteri and hybrid tached by means of petroleum jelly to either 0.01 mm colonies were collected from TennesseeÕs southern (small workers) or 0.1 mm (large workers and alates) border counties. S. richteri was collected from Fayette copper-constantan thermocouples (Omega Engineer- and Hardeman counties, on 1 November 1999 (fall S. ing, Stamford, CT) (Francke et al. 1986, DifÞe and richteri), and 25 May 2000 (spring S. richteri). Hybrid Sheppard 1989, Landry and Phillips 1996). Each ther- colonies were collected from Bradley County on 29 mocouple was attached to a 3-mm-diameter wooden November 1999 (fall hybrid) and 9 June 2000 (spring dowel protruding through the center of a stopper that hybrid). Additional hybrid colonies from Sequatchie County, Tennessee, were collected on 11 January 2000 held the ant/thermocouple array in the center of a 48 (winter hybrid). by 15-mm glass vial. Vials with the ant/thermocouple Colonies of S. invicta were collected in Florida, from arrays were placed in a test tube rack in a chest freezer set at Ð24ЊC. The temperature drop rate in the vials was polygyne populations. Colonies infected with T. so- Ϸ Њ Ϸ Њ lenopsae were collected from Alachua County on 20 5.6 C/min during the Þrst 4 min, slowing to 3.5 C/ March 2000 (infected spring S. invicta). Uninfected S. min by 10 min. Ant temperature changes were re- invicta colonies were collected from Taylor County on corded through a Campbell ScientiÞc CR10 data log- 20 March 2000 (uninfected spring S. invicta). Five ger (Logan, UT) connected to a computer running colonies of uninfected and Þve colonies of T. solenop- Campbell ScientiÞc PC208W data logger support soft- sae-infected S. invicta were also collected on 13Ð16 ware. The supercooling point of each was re- December 1999 from the same locations (winter in- corded as the lowest reading reached before the re- fected and uninfected S. invicta). At the time of col- lease of latent energy shown as an abrupt increase in lection, nest soil was shoveled into an 18.9-liter bucket. the temperature. Frozen ants were saved for head An intravenous drip-tube setup (Banks et al. 1981) was capsule measurement and veriÞcation of infection by used to separate ants from nest soil. Floating ants were T. solenopsae. transferred to ant boxes (plastic containers 40.6 by To determine if ant size correlated with supercool- 28.5 by 17.5 cm or 34.5 by 24 by 12.5 cm) (Rubbermaid, ing point, the head capsules were measured with a Wooster, OH) coated with Fluon (Asahi Glass Flu- wedge micrometer (Porter 1983). Ants from T. so- oropolymers, Chadds Ford, PA) to prevent escape. lenopsae-infected S. invicta colonies also were checked ArtiÞcial nests (20 by 150-mm test tubes with moist individually for T. solenopsae infection by microscopic cotton at the end or 150 by 15-mm petri dishes with examination (400ϫ) of a wet mount slide of the ant dental plaster [Castone, Dentsply-Trubyte, Jackson- abdominal contents. Because live ants do not show ville, FL]) were provided in each ant box. Test tubes external signs of T. solenopsae infection, infection sta- (20 by 150 mm) of deionized water plugged with tus could not be conÞrmed before supercooling tests. cotton were used as a water source. Ants were fed to The supercooling points were compared between T. repletion on a 15% honeyÐ15% dog food agar diet and solenopsae-infected ants and apparently uninfected live [Tenebrio molitor L. (Coleoptera: Tenebri- ants originating from infected colonies. This compar- onidae), Popillia japonica Newman (Coleoptera: Scar- ison allowed us to determine if the presence of T. February 2002 JAMES ET AL.: COLD-SURVIVAL OF FIRE ANTS 129 solenopsae infection in the colony affected the cold Statistical Analysis. Comparison across ant types tolerance of the uninfected ants. was done both with all ants in the supercooling tests, Extended Low Temperature Exposure Tests. The and with only those with head capsules within ranges spring-collected colonies used in supercooling tests shared by all ant types. The shared head width ranges also were used in the extended cold exposure tests. were 0.70Ð0.73 mm (small ants), 1.27Ð1.40 mm (large Groups of Ϸ20Ð30 (range, 14Ð85) worker ants of ants), and 1.40Ð1.43 mm (female alates). Only data for various sizes were collected at random and placed into these head-size selected groups are presented to max- test tubes (20 by 150 mm), with a Fluon ring to prevent imize uniformity of tested groups. The supercooling ant escape and a base of moist Castone to maintain point data were subjected to analysis of variance humidity. Tubes were closed with caps that allowed (ANOVA) followed by mean separation by Student- gas exchange and then placed in test tube racks. For Newman-Keuls or contrasts procedure at a P ϭ 0.05 each of Þve colonies from each ant type described level of signiÞcance using the SuperANOVA software above, one test tube was prepared for each of seven (Abacus Concepts 1989). Because supercooling points sampling days, for each of three temperature regimes were measured on individual ants (subsamples) from (4 Ϯ 0.5, 0.5 Ϯ 0.5, and Ð4 Ϯ 0.3ЊC), resulting in a total each colony (experimental unit), the colony within of 105 test tubes per test ant type. Mortalities at the ant type factor was used as the error term for the main different sampling dates were measured from inde- effect (ant type). The supercooling points of alates, pendent ant samples (tubes), which were discarded and small and large workers were analyzed separately. after use. Thus, the observations were based on ants Spring 2000-collected ants were compared with anal- exposed constantly to the temperature regimes. ogous fall- or winter-collected types to examine sea- Ants were sampled after 1, 2, 3, 5, 7, 10, and 13 d at sonal differences. Ant type and T. solenopsae infection the test temperatures. To avoid preparation of exces- effects were examined for ants collected in the same sive number of tubes, only enough tubes for four season. The supercooling points obtained 1 wk after sample days were initially prepared. Tubes for days collection were compared with those obtained 3 mo 7Ð13 were placed in refrigerators only after removal of after collection to determine the effect of laboratory initial samples, and only for those groups that had rearing on supercooling point. Pearson correlation surviving ants after 5 d. The racks of ant tubes were coefÞcients between head capsule size and supercool- ing points were also calculated to determine any as- placed inside Styrofoam coolers that had 4-cm walls sociation between ant size and freeze-tolerance of and sand to a depth of Ϸ2cm to provide temperature ants within the same type/size category. buffer against the refrigerator cycle. Percent mortality data from extended low temper- All temperature regimes were above the Þre ant ature exposure tests were arcsine-transformed and supercooling points as determined in the previous ANOVA and mean separation procedures were per- experiments. At all these temperatures, cold coma was formed as previously described. Different sampling induced within 24 h; therefore, the ants were not dates were used as a repeated measurement factor in provided with food and water. Without food or water, the analysis. Temperature, colony and experiments the ants kept at room temperature in preliminary tests were used as main effects. Probit analysis using POLO had excessive mortality and this control treatment was - PC (LeOra Software 1987) was used to determine not used subsequently. StowAway (Onset, Pocasset, LD50 of cold exposure as measured in the number of MA) data logger units were used to record tempera- days at the temperature regime. tures in each cooler in the refrigerators. Coolers were left open after being placed in the refrigerator, to facilitate the temperature drop in the test tubes, but Results were closed after 4.5 h to maintain uniformity at the Supercooling Points. Thermocouple Size. Large ants desired temperature. consistently broke the small (0.01 mm) thermocou- On sample days, ant tubes were removed from the ples; therefore, the use of larger, sturdier (0.1 mm) coolers for counting of living and dead individuals. thermocouples was necessary for the large ants. No Ants were considered to be alive if, after being turned signiÞcant difference was found between supercool- on their backs, they could return to an upright position ing points of small and large workers when tested on and walk. Ants at the 0.5 and Ð4ЊC regimes were the small thermocouples (small workers ϭ Ð5.9 Ϯ transferred into 60-ml plastic cups at room tempera- 0.07ЊC; large workers ϭ Ð5.8 Ϯ 0.07ЊC), but super- ture after removal from refrigerators to prevent cooling points were signiÞcantly different (F ϭ 4.044; drowning in condensation formed in the tubes and to df ϭ 1, 191; P ϭ 0.031) with large thermocouples (small revive them more quickly. These ants were allowed 2 ants ϭ Ð7.4 Ϯ 0.13ЊC; large ants ϭ Ð7.7 Ϯ 0.10ЊC). Also, and3htorecover from cold coma. The ants kept at the supercooling point measurements with large ther- the 4ЊC regime recovered from cold coma within min- mocouples were Ϸ1.5ЊC lower than the data obtained utes at room temperature, and were not subject to the with the more sensitive small thermocouple. condensation problem. This experiment was repeated Ant Type Comparisons. No signiÞcant differences in with ants from the same colonies except for two col- supercooling point were observed when the sub- onies of uninfected S. invicta and one colony each of groups of ants with similar head size were compared the other three ant types, which were substituted by (Table 1). Mean head size varied considerably among new colonies due to the low number of remaining ants. the ant types tested. Despite the head size-selected 130 ENVIRONMENTAL ENTOMOLOGY Vol. 31, no. 1

Table 1. Supercooling points (SCP) (mean ؎ SEM °C) and head capsule sizes (HS) (mean ؎ SEM mm) for head-size-selected groups of spring-collected Solenopsis ants

Ant type n SCP (ЊC)a HS (mm) Small workersb (head-size ϭ 0.70Ð0.73 mmc) S. richteri 27 Ϫ6.9 Ϯ 0.23a 0.73 Ϯ 0.000b Hybrid 19 Ϫ6.7 Ϯ 0.13a 0.72 Ϯ 0.003ab Uninfected S. invicta 9 Ϫ8.1 Ϯ 0.43a 0.71 Ϯ 0.005a T. solenopsae-infected S. invicta 7 Ϫ7.5 Ϯ 0.69a 0.71 Ϯ 0.006a Large workersb (head-size ϭ 1.27Ð1.40 mmc) S. richteri 31 Ϫ6.6 Ϯ 0.09a 1.31 Ϯ 0.007a Hybrid 34 Ϫ7.7 Ϯ 0.35a 1.34 Ϯ 0.007a Uninfected S. invicta 24 Ϫ6.9 Ϯ 0.21a 1.32 Ϯ 0.009a T. solenopsae-infected S. invicta 6 Ϫ6.0 Ϯ 0.17a 1.30 Ϯ 0.021a Female alatesb (head-size ϭ 1.40Ð1.43 mmc) S. richteri 37 Ϫ10.0 Ϯ 0.59a 1.42 Ϯ 0.003b Hybrid 24 Ϫ8.0 Ϯ 0.25a 1.42 Ϯ 0.004b Uninfected S. invicta 32 Ϫ8.8 Ϯ 0.42a 1.41 Ϯ 0.003ab T. solenopsae-infected S. invicta 18 Ϫ7.6 Ϯ 0.93a 1.39 Ϯ 0.005a

a Within a column, and within the ant size group, means followed by the same letter are not signiÞcantly different at the 0.05 level of conÞdence according to Student-Newman-Keuls means separation procedure. b Small workers were tested with a 0.01-mm thermocouple. Large workers and female alates were tested with 0.1-mm thermocouple. c Five colonies were tested per ant type. A maximum of 10 ants were tested from each colony but alates may not have been available in large numbers or in all colonies tested. Data is presented for head-size selected subgroups consisting of individuals with head capsule within a size range shared across all three species. groups within speciÞc ranges, signiÞcant differences 0.11; and uninfected S. invicta, F ϭ 0.30; df ϭ 1, 33; P ϭ in the mean head size also were observed for small 0.59). workers (F ϭ 3.98; df ϭ 3, 15; P ϭ 0.03), and female Extended Low Temperature Exposure. The results alates (F ϭ 8.4; df ϭ 3, 11; P ϭ 0.003). No signiÞcant for the two extended low temperature exposure tests correlation (P varying from 0.052for large S. richteri to were not signiÞcantly different (F ϭ 0.001; df ϭ 1, 95; 0.9688 for small S. richteri) was observed between P ϭ 0.97); thus, the results were combined for discus- head size and supercooling point for any of the ant sion (Fig. 1). Ant mortality rates increased as temper- types or castes. atures decreased (F ϭ 485.2; df ϭ 2, 95; P ϭ 0.0001). Colonies of S. invicta were generally composed of Also, signiÞcant differences (F ϭ 18.7; df ϭ 3, 95; P ϭ smaller individuals. Head capsule widths for small 0.0001) were detected among the four ant types workers ranged from Ͼ0.40 to Ͻ0.75 mm, and head tested. The hybrid ants were the most resistant to capsules for large workers ranged from Ͼ0.90 to Ͻ1.45 near-freezing temperature exposure. S. richteri and T. mm. T. solenopsae-infected and uninfected S. invicta solenopsae-infected S. invicta had intermediate resis- did not have signiÞcantly different head capsule sizes tance to the low temperature regimes, and the unin- in any size group. Small S. richteri and hybrid ants had fected S. invicta suffered the highest mortalities when head width values of 0.60Ð0.87 mm, whereas large exposed to cold. No signiÞcant interaction was ob- worker heads ranged from 1.07 to 1.50 mm. served among the main effects in the analysis. Maintenance in the laboratory and collection sea- During the Þrst5datthe4ЊC regime, no signiÞcant son also affected supercooling point of Þre ants. Large (P Ͼ 0.05) differences were observed as all four ant hybrid workers tested 1 wk after collection had sig- -niÞcantly (F ϭ 4.3; df ϭ 1, 9; P ϭ 0.044) lower super- Table 2. Mean supercooling points (mean ؎ SEM °C) for head cooling point than ants tested 3 mo after collection size-selected groups of fall-collected Solenopsis workers, tested from the Þeld, but no signiÞcant difference existed for after maintenance in the laboratory over different periods of time ϭ ϭ ϭ S. richteri (F 3.7; df 1, 9; P 0.063) (Table 2). The Ant Test time n Smallb n Largeb ant type effect was not signiÞcant for time comparison typea of small workers (F ϭ 1.6; df ϭ 9, 33; P ϭ 0.17). S. richteri 1wk 19 Ϫ6.7 Ϯ 0.08ac 39 Ϫ8.7 Ϯ 0.12a Seasonal differences in supercooling point were not 3mo 16 Ϫ6.6 Ϯ 0.15a 31 Ϫ8.0 Ϯ 0.14a signiÞcant for small S. richteri (F ϭ 0.063; df ϭ 1, 33; Hybrid 1 wk 27 Ϫ6.2 Ϯ 0.06a 38 Ϫ8.3 Ϯ 0.11a P ϭ 0.80), hybrid (F ϭ 0.23; df ϭ 1, 33; P ϭ 0.63), T. 3mo 29 Ϫ6.3 Ϯ 0.08a 39 Ϫ7.5 Ϯ 0.11b solenopsae-infected (F ϭ 2.1; df ϭ 1, 33; P ϭ 0.16), and ϭ ϭ ϭ a Five colonies were tested per ant type, and a maximum of 10 ants uninfected S. invicta (F 1.8; df 1, 33; P 0.19). were tested from each colony. Data is presented for head-size selected Fall-collected large S. richteri workers had signiÞ- subgroups consisting of individuals with head capsule within a size cantly lower supercooling point than the spring-col- range shared across all ant types. lected ants (F ϭ 24.8; df ϭ 1, 33; P ϭ 0.0001), but other b Small workers were tested with a 0.01-mm thermocouple and large workers with 0.1-mm thermocouple. seasonal comparisons for large workers were not sig- c ϭ ϭ ϭ Within a column, and within the ant type, means followed by the niÞcant (hybrid, F 3.7; df 1, 33; P 0.063; T. same letter are not signiÞcantly different at the 0.05 level of conÞ- solenopsae-infected S. invicta, F ϭ 2.7; df ϭ 1, 33; P ϭ dence (F test). February 2002 JAMES ET AL.: COLD-SURVIVAL OF FIRE ANTS 131

infected S. invicta (5.2d), but the LD50 for hybrid ants was 8.6 d. Close to 100% mortality was reached by5dinallant types exposed to Ð4ЊC and no survivors remained after 7 d (Fig. 1c). By day 3, the uninfected S. invicta displayed signiÞcantly (P Ͻ 0.05) higher mortality (82.1%) than all other types (hybrid ϭ 41.1%, S. rich- teri ϭ 44.6%, and T. solenopsae-infected S. invicta ϭ 65.6%). The LD50 was estimated to be between 2and 3 d for all ant types.

Discussion Supercooling points were not an adequate measure- ment of cold tolerance because they are highly vari- able even after groups of ants were selected with similar head-capsule size. Also, the differences in re- sults obtained with the 0.01- and 0.1-mm thermocou- ples prevent comparisons with previous supercooling point studies with Þre ants (Francke et al. 1986, Taber et al. 1987, DifÞe and Sheppard 1989, Landry and Phillips 1996), in which larger thermocouples (0.33Ð 0.59 mm) were used. Often, colonies of S. richteri and hybrid Þre ants from Tennessee are composed of larger individuals than S. invicta colonies as demon- strated by head capsule comparisons. Whether larger ant size is an adaptation to colder weather, or a con- sequence from slower development of larvae (Porter 1988), cannot be resolved by our results. Avoidance is usually the Þrst line of defense in cold weather survival of insects (Denlinger and Lee 1998). Imported Þre ants escape damaging cold temperatures by going deeper into their nests (Morrill 1977), but Fig. 1. Mean percent mortality of spring-collected work- ants may become trapped in the upper layers of the ers of S. richteri, S. richteri X invicta hybrid, uninfected S. Ϯ nest if a cold front moves in too rapidly (Green 1959, invicta, and T. solenopsae-infected S. invicta exposed to 4 Morrill et al. 1978, DifÞe et al. 1997). In the extended 0.5ЊC (a), 0.5 Ϯ 0.5ЊC (b), and Ð4 Ϯ 0.3ЊC (c) for up to 13 d. Data from two tests were combined. Bars represent standard low temperature experiments, no structural protec- error of the means. tion was available to the ants and they had succumbed to cold coma within 24 h. Thus, cold avoidance be- haviors were not factors in these experiments and the types had low mortality (Ͻ15%) (Fig. 1a). By the results represent a measurement of physiological, not seventh day, hybrid ants and infected S. invicta had behavioral, cold hardiness. lower mortalities (6.6 and 6.0%, respectively) than S. Short-term exposure to 4ЊC had little effect on ant richteri or uninfected S. invicta (30.6 and 35.5%, re- mortality as Þre ants may have derived some beneÞts, spectively). These differences persisted for the dura- such as lower energy use, from cooler temperatures tion of the experiment. LD50s could not be estimated (Calabi and Porter 1989, Porter and Tschinkel 1993). for this temperature regime due to low mortalities in However, without protective mechanisms, cellular all treatments. functions eventually are affected by continued low- SigniÞcant (P Ͻ 0.05) differences in ant mortality temperature exposure (Denlinger and Lee 1998). were observed on the second day of exposure at the With exposure to colder temperature, the deleterious 0.5ЊC (Fig. 1b). After 2d of near 0 ЊC temperatures, effects occurred sooner. uninfected S. invicta had 18.9% mortality compared A winter-mortality study of an isolated population with only 8.3% for the hybrid, 5.1% for S. richteri and of S. invicta in eastern Tennessee (Callcott et al. 2000) 1.7% for T. solenopsae-infected S. invicta. By5d,the corroborates our results from the extended low tem- uninfected S. invicta mortality was approximately dou- perature exposure tests. In this study, the greatest ble that of the other three ant types. After 7 d, the mortality (87.5%) occurred during winter 1993Ð1994, hybrid ants had Ϸ20% mortality, whereas all other ant with seven consecutive days at Յ1.1ЊC. In our study, types had mortality between 60 and 80%. Hybrid ant 81.4% mortality of uninfected S. invicta was observed mortality approached that of other ant types only after after7dattheϩ0.5ЊC regime. In the Þeld, sunny days 13 d of exposure to the 0.5ЊC temperature regime. The allow nest soil to warm up, reducing the extent of the estimated LD50 was similar for the uninfected S. in- exposure to low temperatures and ant mortality, but victa (4.0 d), S. richteri (5.7 d), and T. solenopsae- cloudy days may prolong exposure of ants to low 132E NVIRONMENTAL ENTOMOLOGY Vol. 31, no. 1 temperatures within the nest. In the future, monitor- Callcott A.M.A., D. H. Oi, H. L. Collins, D. F. Williams, and ing of nest microclimate in the Þeld may allow better T. C. Lockley. 2000. Seasonal studies of an isolated red understanding of ant survival during cold weather, and imported Þre ant (Hymenoptera: Formicidae) popula- would be useful in modeling the likelihood of Þre ant tion in eastern Tennessee. Environ. Entomol. 29: 788Ð794. colonization (Wiley et al. 2000). Further beneÞt may Cokendolpher, J. C., and O. F. Francke. 1985. Temperature be obtained from studies focusing on S. richteri and the preferences of four species of Þre ants (Hymenoptera: Formicidae: Solenopsis). Psyche 92: 91Ð101. S. richteri X invicta hybrid and on possible cell-level Cokendolpher, J. C., and S. A. Phillips, Jr. 1990. Critical advantages of hybrid ants that determine cold toler- thermal limits and locomotor activity of the red imported ance. Þre ant (Hymenoptera: Formicidae). Environ. Entomol. The geography of S. richteri infestations in both 19: 878Ð881. North and South America, and a lack of previous Denlinger, D. L., and R. E. Lee, Jr. 1998. Physiology of cold evidence of hybrid advantage (DifÞe and Sheppard sensitivity, pp. 55Ð95. In G. J. Hallman and D. L. Denlinger 1989, DifÞe et al. 1997), led to the expectation that the [eds.]. Temperature sensitivity in insects and application black imported Þre ant had an improved cold hardi- in integrated pest management. Westview, Boulder, CO. ness. However, the survival of S. richteri exposed to Diffie, S. K., S. M. Bass, and K. Bondari. 1997. Winter sur- extended periods of cold temperatures was not sig- vival of Solenopsis invicta and the Solenopsis hybrid (Hy- menoptera: Formicidae) in Georgia. J. Agric. Entomol. 14: niÞcantly different from that observed for the more 93Ð101. tropically oriented S. invicta. Despite the lack of sig- Diffie, S. K., and D. C. Sheppard. 1989. Supercooling studies niÞcant differences in cold tolerance between S. in- on the imported Þre ants: Solenopsis invicta and Solenopsis victa and S. richteri, the hybrid was more resistant to richteri (Hymenoptera: Formicidae) and their hybrid. J. extended exposure to low temperature than both par- Entomol. Sci. 24: 361Ð364. ent species. Drees, B. M., C. L. Barr, S. B. Vinson, R. E. Gold, M. E. Within their current range, Þre ants would rarely be Merchant, N. Riggs, L. Lennon, S. Russell, P. Nester, D. exposed to temperatures close to their supercooling Kostroun, and others. 2000. Managing imported Þre ants points inside the nest. However, exposure to occa- in urban areas. Tex. A&M Univ. Agric. Ext. Serv. Bull. sional extended periods of low temperature may occur B-6043. Francke, O. F., and J. C. Cokendolpher. 1986. Temperature frequently. Results from extended exposure to low tolerances of the red imported Þre ant, pp. 104Ð113. In temperatures support the theory that the distribution C. S. Lofgren, and R. K. Vander Meer [eds.], Fire ants and of S. richteri, S. invicta, and their hybrid is linked to leafcutting ants: biology and management. Westview, cold hardiness. As the Þre ant range expands north, the Boulder, CO. chances of Þre ants encountering damaging temper- Francke, O. F., J. C. Cokendolpher, and L. R. Potts. 1986. atures are increased. Based on our extended low tem- Supercooling studies on North American Þre ants (Hy- perature results, we expect that the northern range menoptera: Formicidae). Southwest. Nat. 31: 87Ð94. expansion may favor the hybrid Þre ant over the par- Green, H. B. 1959. Imported Þre ant mortality due to cold. ent species. J. Econ. Entomol. 52: 347. Hubbard, M. D., and W. G. Cunningham. 1977. Orientation of mounds in the ant Solenopsis invicta (Hymenoptera: Acknowledgments Formicidae: ). Insectes Soc. 24: 3Ð7. Kaspari, M., and E. L. Vargo. 1995. Colony size as a buffer We thank Dave Williams, David Oi, and Bob Vander Meer against seasonality: BergmannÕs rule in social insects. Am. (USDA-ARS, CMAVE), Jerome Grant, Ernest Bernard, Reid Nat. 145: 610Ð632. Gerhardt, and Charles Pless (University of Tennessee) for Landry, C. E., and S. A. Phillips, Jr. 1996. Potential of ice- their assistance during the course of these experiments and nucleating active bacteria for management of the red critical review of the manuscript. Also, we are grateful for imported Þre ant (Hymenoptera: Formicidae). Environ. assistance from John Wilkerson, Vernon Riech, David Rus- Entomol. 25: 859Ð866. sell, Stephanie Smithson, Jenny Croker, Derek Bailey, Jen- LeOra Software. 1987. POLO-PC a userÕs manual for probit nifer Manrod, Steve Schaller, Brian Leckie, Choo Hamilton, or logit analysis. LeOra Software, Berkley, CA. and Dane´sha Seth for assistance with experiments. Morrill, W. L. 1977. Overwinter survival of the red im- ported Þre ant in central Georgia. Environ. Entomol. 6: References Cited 50Ð52. Morrill, W. L., P. B. Martin, and D. C. Sheppard. 1978. Abacus Concepts. 1989. SuperANOVA, version 1.11. Abacus Overwinter survival of the red imported Þre ant: effects Concepts, Berkeley, CA. of various habitats and food supply. Environ. Entomol. 7: Banks, W. A., C. S. Lofgren, D. P. Jouvenaz, C. E. Stringer, 262Ð264. P. M. Bishop, D. F. Williams, D. P. Wojcik, and B. M. Pinson, C. K. 1980. The temperature regime in the Solenop- Glancey. 1981. Techniques for collecting, rearing, and sis invicta mound and its effect on behavior. M.S. thesis, handling imported Þre ants. USDA, SEA, AATS-S-21. Tech University, Lubbock, TX. Calabi, P., and S. D. Porter. 1989. Worker longevity in the Porter, S. D. 1983. Fast, accurate method of measuring ant Þre ant Solenopsis invicta: ergonomic considerations of head widths. Ann. Entomol. Soc. Am. 76: 866Ð867. correlations between temperature, size and metabolic Porter, S. D. 1988. Impact of temperature on colony growth rates. J. Insect Physiol. 35: 643Ð649. and developmental rates of the ant, Solenopsis invicta. Callcott A.M.A, and H. L. Collins. 1996. Invasion and range J. Insect Physiol. 34: 1127Ð1133. expansion of imported Þre ants (Hymenoptera: Formi- Porter, S. D., and W. R. Tschinkel. 1987. Foraging in So- cidae) in North America from 1918Ð1995. Fla. Entomol. lenopsis invicta (Hymenoptera: Formicidae): effects of 72: 240Ð248. weather and season. Environ. Entomol. 16: 802Ð808. February 2002 JAMES ET AL.: COLD-SURVIVAL OF FIRE ANTS 133

Porter, S. D., and W. R. Tschinkel. 1993. Fire ant thermal Wiley, S., L. C. Thompson, M. D. Korzukhin, and S. D. preferences: behavioral control of growth and metabo- Porter. 2000. A GIS representation of potential red im- lism. Behav. Ecol. Sociobiol. 32: 321Ð329. ported Þre ant range expansion across the United States, Shoemaker, D. D., K. G. Ross, and M. L. Arnold. 1996. De- pp. 110Ð114. In J. L. Croker, K. M. Vail, and R. M. Pereira velopment of RAPD markers in two introduced Þre ants, [eds.], Proceedings of the 2000 imported Þre ant confer- Solenopsis invicta and Solenopsis richteri, and their appli- ence. University of Tennessee, Knoxville. cation to the study of a hybrid zone. Mol. Ecol. 3: 531Ð539. Williams D. F., D. H. Oi, and G. J. Knue. 1999. Infection of Taber, S. W., J. C. Cokendolpher, and O. F. Francke. 1987. red imported Þre ant (Hymenoptera: Formicidae) colo- Supercooling points of red imported Þre ants, Solenopsis nies with the entomopathogen Thelohania solenopsae invicta (Hymenoptera: Formicidae) from Lubbock, (Microsporidia: Thelohaniidae) J. Econ. Entomol. 92: Texas. Entomol. News 98: 153Ð158. 830Ð836. Vander Meer, R. K., C. S. Lofgren, and F. M. Alvarez. 1985. Biochemical evidence for hybridization in Þre ants. Fla. Received for publication 5 March 2001; accepted 6 September Entomol. 68: 501Ð506. 2001.