Red imported fire ant control using individual mound treatment products, 1986

Are multiple queen mounds of the red imported fire ant more difficult to control than single queen mounds?

 Red imported fire ant control in a blueberry planting

 Aerial application of hydramethylnon fire ant bait (AmdroR) method demonstration

 Alternative red imported fire ant control method and result demonstration
 

There are a number of people who deserve thanks for permitting this compilation of result demonstrations to be completed. My secretary, Doris Pineda, deserves praise for the patience and drive needed to type these chapters and to proof my work.  My co-workers in the Texas Agricultural Extension Service that provided essential and much welcomed cooperation in conducting these result demonstration efforts are listed below:

Funding for these result demonstrations was provided, in part, by the following organizations:

American Cyanamid Company
Dow Chemical, U.S.A.
Merck AgVet Division of Merck and Company, Inc.
Stauffer Chemical Company
Union Carbide Agricultural Products Company, Inc.

The support of these companies and their representatives is greatly appreciated.  The majority of these organizations and others also provide quantities of insecticide products used in these tests.  Finally, I wish to thank producers and other individuals who cooperated in these result demonstrations by providing their labor and land, without whom these results could not have been obtained.

The information contained herein is for educational purposes only. Reference to products or trade names is made with the understanding that no discrimination is intended and no endorsement by the Cooperative Extension Service is implied.

Over 66 products are registered for the control of red imported fire ants. Although these products are all formulations of just over one dozen different active ingredients, there is a great deal of misunderstanding on the part of the general public as to what to expect in terms of product performance. In a continuing effort to document the efficacy of available products and obtain data needed to calculate the per-mound cost of control in terms of monetary input, required labor and time, result demonstrations were conducted in 1986 comparing "newer" products to more "standard" products used to control individual red imported fire ant mounds.
 

One result demonstration was conducted on the Lagow Ranch in Chambers Co. using the following materials in accordance with instructions provided on the products' labels:
 
 

A second result demonstration was conducted in a similar manner on the McInnis pasture in Port Arthur, Jefferson Co. Treatments applied in accordance to label instructions (except for Oftanol) are listed below:
 
 

In this result demonstration 20 large mounds were treated, 16 April, 1986, with each product and 20 mounds were treated with 1 gal of water only. Mounds were inspected 3, 7 and 14 days following treatment.

Mounds treated with each product, water or monitored as untreated controls were divided into subsets containing 5 mounds each. The percent active mounds from each of these groups (6 in Chambers Co., except for Enforcer, and 4 in Jefferson Co., considered as replications) were subjected to analysis of variance using the Least Significant Difference test at the 5 percent level of probability for each evaluation date.
 

The following observations were made during these result demonstration efforts:

Orthene 75S has extremely offensive odor and is difficult to apply as a dust to mounds in windy conditions.

Cessco Injectable* containing chlorpyrifos in a 1 1/2 lb aerosol can treated 15 large mounds.

Cotton's Fire Ant Insecticide treated 17 large mounds per gallons using a 11/2 gal pressurized sprayer with special Cotton's Fire Ant Probe applicator. The probe easily bent and clogged with mud. Blue colored liquid had petroleum smell. Grass around treated mounds was not harmed.

Enforcer retails for $7.99 per 1 lb can plus $1.49 plastic applicator rod and treats 7 large mounds.

Turcam produces an extremely quick kill of worker ants within the first few minutes after application. However, the product foams during mixing.

Oftanol** was applied at much less (1/9th) than the label rate, being sprinkled on to a 9 by 9 ft area (81 sq ft or 9 ft square) rather than a 3 by 3 ft area as specified on the label (9 sq ft or 3 ft square). Thus, lack of control with this product was expected.

Results of the Chambers Co. result demonstration (Tables 1 and 2) indicate that chlorpyrifos and acephate formulated insecticides were the most effective at the 5 and 15 day observation periods. Cotton's Fire Ant Insecticide began reducing fire ant mound immediately, as did chlorpyrifos aerosol, and later produced percent control statistically similar to the two numerically better products. Tetramethrin, although expected to produce an immediate initial reduction of mound activity, did not effect mounds until 5 and 15 day post treatment evaluation dates. Furthermore, percent control was only about half as high as that produced by better products. Amdro produced no effect, leading demonstrators to suspect that the product used was old, rancid and unattractive to the ants. However, activity from the other two bait-formulated products, the insect growth regulators Logic and Affirm, was not determined to be statistically different from that produced by Amdro treatment (Table 3). Untreated control mounds were not monitored during the 19 week post-treatment evaluation. Therefore, percent control could not be calculated.

In the Jefferson Co. result demonstration, Knox Out and Turcam produced 100 percent control within 3 days and maintained total control through the duration of the monitoring period (Table 3). Orthene produced better results when applied as an individual mound drench than as a dust treatment. In fact, although the dust treatment of large mounds produced great reductions in worker ant activity, workers could be found active around the edges of mounds for 14 days. The large mounds treated were difficult to cover entirely with 2 tsp of dust, the maximum labeled rate. Oftanol produced no control at 3 and 14 days post-treatment because it was not applied correctly. Amdro produced no control, perhaps because the material used was old (2 years) and spoiled. Of interest is that the 1 gal water poured on "untreated" control mounds resulted in a 36 percent reduction in active mounds.

 Table 1. Efficacy of individual red imported fire ant mound control products applied April 17, 1986, Chambers Co., Tx.
 

^1/ Means in columns followed by different letters are statistically difference according to the Least Significant Difference Test at the 0.05 level of probability.

 Table 2. Efficacy of selected individual red imported fire ant mound treatment products applied April 16, 1986 in Jefferson Co., Texas.
 

^1/ Means followed by different letters are statistically difference according to the Least Significant Difference Test at the 0.05 level of probability.

 Table 3. Efficacy of slow acting red imported fire ant insecticide bait-formulated products applied April 17, 1986, Chambers Co., Tx.
 

^1/ Means in columns followed by different letters are statistically difference according to the Least Significant Difference Test at the 0.05 level of probability.

Many anecdotal reports and "conventional wisdom" suggest that polygynous nests are more difficult to control than monogynous nests as all of the nest queens must be destroyed. Lammers (1987) found the survival of polygynous queens treated with an experimental juvenile hormone analog bait to be greater than monogynous colonies, suggesting that the polygynous form may be more persistent. We suspected that the use of individual insecticide mound drenches might be less effective on multiple queen colonies based on anecdotal reports which indicate the failure of products applied to areas with polygynous colonies. We compared the levels of control of the two forms of the ant in Texas using the insecticide chlorpyrifos.
 
 

Two sites were selected for treatment: 1) a pasture infested with polygynous RIFA in Brazos County, TX, and 2) a pasture with suspected monogynous colonies in neighboring Montgomery County, TX. The pastures at both sites were similar in vegetative cover and both were lightly grazed. The site in Brazos County consisted of sandier soil. Worker ants from ten randomly selected mounds were collected from each site, and head widths were measured from a minimum of 25 worker ants from each mound before calculating average values for determining colony type (Greenberg et al. 1985). To further substantiate the type of RIFA in each location, ten colonies were collected from plot areas. Queen ants collected from each colony were dissected to observe the presence of sperm in the spermathaeca to determine if they were inseminated and reproductively active.

Circular plots were established in order to be able to monitor 1) activity in 30 individually marked mounds, and 2) density of active mounds on sampling dates before, one and approximately 10 weeks following treatment in each location. In Brazos Co., all mounds within a 0.05 hectare (1/8 acre) circle were mapped prior to treatment. In addition, 30 randomly selected mounds within the mapped area were marked and numbered with Kerr^R canning jar lids nailed into the ground using 5 inch-long nails. All mounds within the mapped area were then treated with 7.39 ml (0.25 fl oz) chlorpyrifos (Dursban^R 4E) in 3.79 liters (one gal) water using plastic sprinkler cans (Hamman et al. 1986). In addition, all mounds in a 3.0 meter (10 ft) wide area around the circular mapped area were treated to produce a mound-free "buffer" zone. Four treatment plots (replicates) were established, one each week beginning July 16, 1986.
 
A control plot, mapped and marked as described above, but in which mounds were drenched solely with 3.79 liters water was established and treated on the same days as the chemically treated plot. This plot was monitored weekly and served as the non-chemically treated control plot from which to compare the effects of insecticide applications in treated plots established weekly (replicated over time). Populations in this non-chemically treated plot remained relatively stable with few of the 30 marked mounds becoming inactive during the treatment to observation period. When marked mounds did become inactive, additional active mounds were marked and numbered for the following week's (next replicate's) evaluation.

RIFA mound densities were much lower in Montgomery County, requiring larger, 0.1 hectare (1/4 acre) plots. Three to six circular plots were required before 30 mounds could be mapped, marked and numbered. The resulting set of circular plots was considered to be a single replication. All mounds within mapped areas, plus those within a 3 meter buffer zone were treated as in Brazos County. Four sets of circular treatment plots were mapped and treated weekly beginning July 14, 1987. One set of control plots were established, treated with water and monitored weekly in a fashion similar to the Brazos County location.
 
Mounds were monitored at one and eleven weeks after treatment using a minimal disturbance method: First, individual mounds were disturbed slightly to detect worker ant activity. If worker ants emerged in mass from the colony, that colony was determined active. If no activity was detected, the mound was further disturbed. The mound was determined to be inactive if no ant activity could be detected after severe disruption of the colony. This procedure resulted in a value for 1) the total number active mounds out of 30 originally marked mounds for each mapped plot or set of plots, and 2) the total number of active mounds per unit area which was converted to number of active mounds per acre.

Percent reduction of active RIFA mounds for each of the chlorpyrifos treated plots relative to their respective water-treated mound plots was derived for the 1 and 11 week post-treatment evaluations at each site by 1) calculating the percent negative difference in the number of active colonies from the 30 marked mound sets and 2) using Henderson's formula (Henderson and Tilton 1955) for the number of mounds per acre estimates. The Student's t test was used to separate mean worker head width as well as the mean percent control values derived to document any statistical differences between the two forms of RIFA.

The dimensions of the mounds differed between 30 randomly selected mounds measured from the Montgomery and Brazos County plots. Montgomery County plot mounds averaged 15.44 + 5.08 cm in height. Height of mounds in Brazos County plots could not be accurately measured since they were low. Diameters of mounds from both areas were similar ( 46.36 + 10.13SD cm in Montgomery County vs 50.11 + 13.06SD cm in Brazos County). Differences in mound height may have been due to other factors such as soil type rather than to fire ant type. Soils in Montgomery County were hard black clay while hardened sand was predominant in the Brazos Co. plots. No rain occurred at either location during the establishment and one week post-treatment evaluation phases of this trial.
 
 

Samples of worker ants randomly collected from 10 mounds from the Brazos County RIFA plots (August 7 and 22, 1986), showed that the population is predominantly (80 percent) polygynous (Table 1) with only two mounds (*) with an average head capsule size of over 0.74 mm. In addition, 80 percent of the mounds collected from the Brazos Co. plots with de-alate reproductive females contained more than one mated queen. The Montgomery County colonies sampled on July 3, 1986 were predominately (90 percent) monogynous (Table 1) with all but one (**) mound out of ten mounds sampled having an average worker head capsule size exceeding 0.74 mm. Furthermore, 25 percent of the mounds were found to contain a single mated female. The remainder contained numerous unmated de-alate females thought to have lost their wings after mounds were collected from the field. In the absence of the queen, female reproductives soon drop their wings (Fletcher and Blum 1981). Thus, the absence of mated females in these cases is interpreted as supportive evidence that these were monogynous mounds.
 
Table 2 presents the results of the treating four replicates of 30 marked mound sets of polygynous and monogynous mounds with 7.39 ml chlorpyrifos in 3.79 liters water per mound. No significant difference was found between the percent reductions of mounds between the two forms of ant colonies. The reduction in mound number in the controls of both forms were similar although more non-chemically treated mounds in Montgomery County became inactive. This may indicate that both forms respond similarly to natural mortality factors and perhaps that more movement of colonies from one mound site to another occurs in with monogynous forms.

The effect of mound drenches on the density of mounds is presented in Table 3. The density of polygynous mounds was 5 - 9 times the density of monogynous mounds in the control plots. Treatments reduced the number of mounds in both populations. The percent reductions was greater in the polygynous population at both 7 and 73+7 days with both populations showing a post treatment recovery by 73+ days (Table 4). Although not significantly different, recovery appeared greater in the monogynous populations (Table 4). The heavy insecticide use in the polygynous area may have reduced the resurgence of the polygynous mound densities as very few untreated sites existed for mound construction.

While chlorpyrifos drench treatments resulted in a greater percent reduction of polygynous mounds, because of the greater density, the absolute number of mounds per unit area remained 5 to 8 times greater post treatment (Table 3). The presence of 5 mounds per acre in the monogynous populations is generally tolerated better than 30 mounds per acre that remained in the treated polygynous area.

Other factors must be considered in evaluating treatment methods for control of polygynous vs monogynous RIFA populations, including labor, cost and environmental effect of product usage. Mixing, hauling and drenching individual mounds with a sprinkler can is labor intensive, particularly where a water source is remote. Obviously treating increased numbers of mounds in polygynous ant areas requires additional labor and time. Dursban^R 4E, containing 1.8 kg (4 lb) of chlorpyrifos per 3.79 liters (1 gal), retailed for $77.50 per gal in 1986. Brazos County polygynous mound densities required 640 individual mound drenches per 0.4 hectare (1 acre), using 1.25 gal Dursban^R 4E or 2.3 kg (5 lb) active ingredients per 0.4 hectare (1 acre) at a cost of $96.87. The Montgomery County only 29 individual mound drenches per 0.4 hectare were required to treat monogynous ant mounds, using 2.15 liters (0.57 gal) Dursban 4E or 0.11 kg (0.24 lb) active ingredients at a cost of $4.39. Current instructions on the product label do not allow for adjustments in the needed amount of active ingredient or water volume relative to mound size or S. invicta colony type. The label also does not suggest a maximum number of mounds to be treated per acre for economic purposes and in order to avoid excessive environmental contamination of areas treated.
 
Results obtained from using an individual mound drench to control monogynous or polygynous mounds of the RIFA in this experiment address the question, "Are multiple queen mounds of the red imported fire ant more difficult to control than single queen mounds?" The answer is not a simple 'yes' or 'no', as there are several different aspects which must be considered. Toxicologically, multiple queen RIFA mounds are as easy or easier to control than single queen mounds. From an analysis of percent control obtained with an individual mound treatment method of control, multiple queen mounds were eliminated to a significantly greater degree after one week than were single queen colony mounds, and resurgence was lower at 11 weeks. Numerically, however, multiple queen mound densities were consistently higher than those of single queen mounds to the point of being as numerous one week after treatment as were single queen colony mound densities prior to treatment. Furthermore, the labor, cost and amount of insecticide applied to the environment by treating multiple queen mounds renders the method used in this research impractical as a sole tactic for managing the polygynous RIFA.
 

Buren, W.F., G.E. Allen, W.H. Whitcomb, F.E. Lennartz and R.N. Williams. 1974. Zoography of the imported fire ants. J. N.Y. Entomol. Soc. 82:113- 124.

Fletcher, D.J.C. 1983. Three newly-discovered polygynous populations of the fire ant, Solenopsis invicta, and their significance. J. Ga. Entomol. Soc. 18:538-543.

Fletcher, D.J.C. and M.S. Blum. 1981. Pheromonal control of dealation and oogenesis in virgin queen fire ants. Science 212: 73-75.

Glancy, B.M., C.H. Craig, C.E. Stringer, and P.M. Bishop. 1973. Multiple fertile queens in colonies of imported fire ant, Solenopsis invicta. J. Ga. Entomol. Soc. 8:237-238.

Greenberg, L., D.J.C. Fletcher, and S.B. Vinson. 1985. "Differences in worker size and mound distribution in monogynous and polygynous colonies of the fire ant Solenopsis invicta Buren". J. Kansas Entomol. Soc. 58(1):9-18.
 
Hamman, P.J., B.M. Drees and S.B. Vinson. 1986. "Fire ants and their control", B-1536. Texas Agric. Ext. Serv., Texas A&M University System, College Station, Texas. 10 pp.

Henderson, C.F. and E.W. Tilton. 1955. Tests with acaricides against the brown wheat mite. J. Econ. Entomol. 63:1536-1539.

Hung, A.C.F., S.B. Vinson, and J.W. Summerlin. 1974. Male sterility in the red imported fire ant, Solenopsis invicta. Ann. Entoml. Soc. Am. 67:909-912.

Lammers, J.N. 1987. Mortality factors associated with the founding queens of Solenopsis invicta Buren, the red imported fire ant: A study of the native ant community in central Texas. Thesis, Texas A&M University. 206 pp.

Lofgren, C.S., and D.F. Williams. 1984. Polygynous colonies of the red imported fire ant, Solenopsis invicta, in Florida. Fla. Entomol. 67:484-486.
 

* indicates average worker ant head widths greater than 0.71 mm, indicating that these may not be polygynous colonies. The two that exceed 0.74 mm are considered to be monogynous.
** indicates average worker ant head widths less than 0.74 which may not be a monogynous colony.
 
^1/ Overall average head capsule sizes for each population were statistically different. Student's t test (t = -4.8762; P = 0.0001; DF = 18).

^1/ Means with Standard Deviation values of mounds per acre converted from mound counts of four circular one-eighth of an acre plots.
^2/Means with Standard Deviation values of mounds per acre converted from mound counts of four replicates of three to six circular quarter-acre plots.

^1/ Means with Standard Deviation values for percent reduction of mounds per acre converted from mound counts of four circular one-eighth of an acre plots.
^2/Means with Standard Deviation values for percent reductions of mounds per acre converted from mound counts of four replicates of three to six circular quarter-acre plots.
^3/Statistically different means at P>5% using the Student's-t test (t = 2.2903;P = 0.0310; DF = 6).
^4/Not statistically different at P>5% using the Student's t test (t = 1.7655; P = 0.0640; DF = 6).
 

Red imported fire ants inhabit blueberry plantings in east Texas, becoming a nuisance to workers or clientele in a "pick-your-own" operation. Insecticide products specifically registered for use in this site for this pest are virtually non-existent. Very little work has been undertaken to study fire ant management in blueberry plantings. Robinson (1983) compared root ball applications of acephate (Orthene^R 75SP) and individual mound treatment rates on amidinohydrazone (Amdro^R). However, Amdro can technically not be used "in agricultural land".

The result demonstration reported here was undertaken to determine if a broadcast application of Amdro in the turfgrass areas BETWEEN rows of blueberry bushes would reduce red imported fire ant mounds both in the rows and in the space between the rows. Treatment in this manner could be consistent with label instructions since actual rows of blueberries did not receive the application, but was effected by the product. A similar interpretation of the Amdro label is used when applications of this product are suggested AROUND home gardens, since technically Amdro can not be used IN gardens. Ants foraging around the garden would pick up the bait and bring it back to the colony wherever it was, resulting in a suppression of foraging activity inside and around the garden plot.
 

This result demonstration was conducted on Lamar Lynch's blueberry field. Amdro was first applied November 18, 1986 using a manual Cyclone^R seeder at a rate of 1-1.5 lb per acre. The demonstration covered 1.63 square acres with a total of 267 red imported fire ant mounds (163 mounds per acre). Plots were established that were 98 by 60 feet in a paired block design. Untreated plots were alternated with Amdro treated plots, replicated three times. The number of active red imported fire ant mounds were counted prior to application (Nov. 18), and on December 2, 14 days after application. Plots were re-treated on April 10, 1987, and evaluated on that date, and on May 8,22 and June 12.
 

Table 1 lists the results of these demonstration efforts. The November treatment of Amdro was unsuccessful in impacting on existing fire ant mound densities after 2 weeks and up until the April 10, 1987 second treatment date. Control failure may have been due to a number of possible factors including use of stale bait-formulated product, cold temperatures preventing foraging activity at the time of application, etc. However, after the spring application mound numbers in the Amdro-treatment plots were significantly reduced 80.5 percent after 4 weeks, 88.9 percent after 6 weeks, and 86.6 percent after 9 weeks.

To ensure a maximum level of control and no possibility in contamination of produce in proximity to treated areas from the use of this product, treatment of turf areas around blueberry plantings would best be applied at least 5 weeks prior to expected harvest activities. This pesticide is currently not specifically labeled for use in orchards or other agricultural land. The USER of any pesticide product is always responsible for directly following the instructions on the product label.

------------

Robinson, J.V. 1983. Red imported fire ant control using Orthene^R 75SP. in Red imported fire and control result demonstrations 1979-1986. B.M. Drees (ed). Texas Agricultural Extension Service. Texas A&M University, College Station, Texas.

 Table 1.  Number of red imported fire ants per plot (60x98 ft) in a blueberry field, Hardin Co., Texas, 1986-1987.
 

^1/ Amdro^R was applied at 1-1.5 lbs per acre to the turfgrass areas between rows of blueberry bushes November 11, 1986 and April 10, 1987.
^2/ Probability levels of less than 5 percent (*) are considered to be statistically significant according to the Student's t test.

Red imported fire ant insectides formulated as baits on light particles of processed corn cob grits coated with soybean oil, as is commonly used in the formulations of hydramethylnon (Amdro®), avermectin (Affirm®), fenoxycarb (Logic®) and Pro-Drone^TM, are registered for broadcast application at a rate of 0.8 to 1.5 lbs formulated insecticide per acre.  The methods used to achieve these low rates are no clear and difficulties exit in complying to the label directions with conventional application equipment. American Cyanamid Company had developed and distributed a bulletin in an attempt to assist individuals attempting to use their product (anonymous, 1980, Amdro^R Fire Ant Insecticide Information Bulletin, "Application equipment information", PE-5824, American Cyanamid Company, Wayne, New Jersey).

Aerial application of fire ant baits requires special modifications of application equipment, but most applicators are restricted to attempting application with conventional seeding/granular fertilizer application equipment and at most, turning the air scoop forward. This method demonstration was conducted in cooperation with Henry P. O'Neal, Dr. Jack D. Price and Dr. Rodney Holloway to determine the effectiveness of aerially applying Amdro at the 1987 Brazoria County Aerial Application Clinic.
 
 
Materials and Methods
 
The cooperator in this method demonstration was pilot James Knape, using an Ag Cat airplane fitted with a Stainless Steel Fabricators, Inc. Spreader. The pilot was instructed to fly over a series of 1 meter square funnels placed in a row 30 meters across the flight path. Granules collected in the funnels were weighed and weights were entered into a computer program that plotted the distribution pattern of the granules in the swath, and produced a printout including other technical and environmental information. The program was also capable of calculating optimum swath width based on the data collected.
 

Results and Discussions

Tables 1 and 2 list the results of this method demonstration. On the day of this demonstration (March 4, 1987, 2:06 pm), the temperature was 70^oF and wind was blowing at 12 mph. The pilot flew into the wind to make the trial application. By adjusting the settings of the gate opening to 11.5 turns, the pilot was able to achieve an application rate of 1.0 lbs per acre at a swath width of 40 ft, flying at an altitude of 12 ft and at a ground speed of 96 mph. However, the range of material applied within this swath was from 0 to 2.5 lbs per acre. The simulated optimum swath width for applying 1.1 lb per acre of material was 52 feet. In this simulation, the range of rates of bait applied within the swath was from 0.5 to 2.2 lbs per acre. The rate range within the actual and simulated swaths illustrate the difficulty of achieving a proper application of bait formulated fire ant insecticides by air, and explains, in part, the erratic results of product efficacy form trials in which these baits had been aerially applied.
 
 

1.)  AIRCRAFT ID:AC N48655                                                             PASS:  1-2
      SPREADER:  SSF                                                                            MATERIAL:  AMDRO
      VANE SPACINGS FOR SETTING #  :

2.)  GROUND SPEED:  96 MPH                                                           FLIGHT HEIGHT:  12 FT. OR 3.7 M

3.)  WIND SPEED:  12 MPH                                                                 WIND DIRECTION (090= IN WIND):  116
      TEMPERATURE:  70F, OR 21C                                                     RELATIVE HUMIDITY:  31%
 
4.)  NUMBER OF NONZERO WEIGHTS:  19                                     TOTAL RATE:    20.3KG/HA, OR   18.1 LB/AC
      LATERAL DISPLACEMENT OF MASS:  0.65 M LEFT
      VERTICAL CENTROID:    0.7 KG/HA, OR   0.6  LB/AC
      MEAN RATE:    1.1 KG/HA, OR    1.0 LB/AC                               SWATH AT MEAN:   12.0 M, OR 39 FT
      CALIBRATED SWATH WIDTH:  12.2 M, OR 40 FT                    MEAN AT SWATH:  1.7 KG/HA, OR 1.5 LB/AC
 

 

 Table 2. Coefficient of variation vs swath width racetrack lapping and racetrack lapping 16 meters or 52 feet reports, documenting results of computerized simulation of the optimum application pattern calculated from data obtained by aerially applying hydramethylnon (Amdro^R) fire ant bait insecticide, Brazoria Co., Texas 1987.
 

When it comes to killing fire ants, American ingenuity is alive and well - particularly in Texas. Currently there are several new methods on the market for the control of red imported fire ant (Solenopsis invicta Buren). These products are unique in several ways. First, several of them do not use insecticides but rather employ electricity or water to kill ants. Second, several of the new devices are intended for the equipment leasing and service markets rather than for homeowner use. Thus, although they may be initially expensive, their cost over time for killing fire ants in mounds may make these products competitive.

Historically, the first non-chemical fire ant control product was the McCoy Ant Stomper (advertized in the Progressive Farmer, 1978). This device was a windmill. As wind activated the windmill, the windmill turned a crank than operated a vibrating plate (stomper). When placed on top of a fire ant mound, the agitation of the mound by the stomper was thought to be unfavorable for the ants in the mound, causing them to migrate to a new location. This devise was based on a concept similar to that behind periodic disturbance of mounds with a shovel or other implement (i.e. lawn mower adjusted to a low setting) to force mounds to move to other, less disturbed areas. Although the McCoy Ant Stomper is no longer available, the distinction between any action which makes mounds move as opposed to actually killing a colony is still a difficult one to make. Careful experimental design and treatment or mapping of all mounds in an area is required to prove that a product is actually eliminating mounds.

Another device marketed recently was called the "Exterminator" (J&L Marketing). This device used a cartridge containing two explosive and burning elements placed on the end of an exterminating rod. The rod was inserted into the hill, creating a blast in the queens chamber that destroyed the entire colony. No data concerning the efficacy of safety of this device has been published.

The Texas Agricultural Extension Service has primarily conducted result demonstrations with insecticide products registered by the Environmental Protection Agency and the Texas Department of Agriculture. A report entitled, "Red imported fire ant control result demonstrations, 1979-1986" includes 43 reports documenting the effectiveness of numerous products containing one or more of 20 different active ingredients. This report also includes results of field trials using the straw itch or Pymotes mites currently on the market billed as "fire mites" and advertized as a biological method of fire ant control. These mites have not been found to be effective in eliminating fire ant mounds.

Biological control agents and devices currently on the market for the treatment of red imported fire ant mounds are not required to be registered by the EPA as long as there is no chemical insecticide involved. Therefore, efficacy data for these devices has not been generated by public agencies such as universities or the Extension Service. This result demonstration is designed to test the performance of several of the new (non-chemical) mechanical control devices and one method of applying a vaporized form of an insecticide (resmethrin).

Results of demonstration efforts are used to formulate management recommendations. These are published by TAEX, in a publication entitled, "Fire ants and their control" (B-1536), which can be obtained through the Department of Agricultural Communications at Texas A&M University or by visiting your county agent. This publication lists over 66 products currently registered for control of red imported fire ants and their nests. Products and other methods of control vary greatly in cost, environmental impact, labor involved in application, mode and speed of action and effectiveness. Because there are so many alternatives, there is a great deal of misunderstanding on the part of the general public as to product performance. The educational programs conducted by the Texas Agricultural Extension Service are designed to document product performance, enabling the user to have realistic expectations from investments made for fire ant suppression efforts.
 

Four techniques for eliminating ant activity in mounds were evaluated in this result demonstration:

1. The Yaard-Vark^TM, donated for experimentation by J. R. McCracken of Bryan, Texas, son of the inventor, uses 10 watts of electricity to electrocute ants attempting to crawl up the device past a wire coil similar to a miniature "bug zapper". The literature with the product claims that the ants' behavior is altered so that the ants "attack one another" and "fight each other to the death". This claim has not been scientifically proven. Although the instructions state that the device should be left in the mound three hours or more until ant activity ceases, Mr. McCracken suggested leaving the device in each mound for 15 to 20 minutes, probing it into several locations within the mound during that period. This treatment is used to "traumatize" the ants producing inactive colonies within 1-2 days (Pers. comm.).

2. The Anster^TM, developed and promoted by John C. Connolly of the Antser, Inc., is a mechanical devise using a two-cycle engine to rotate prongs or blades arranged to churn the soil of the mound, physically disrupting ants. Water is added during the agitation process, and the resulting slurry encases remaining ants in water-saturated mound medium. This device is being developed for the service sector (Private pest control operators) and/or the lease/rental market.

3. Ant Fire, Inc. has developed a device, called the Earthfire^R Injection System, that vaporizes the insecticide, resmethrin (Earthfire^R), to fumigate fire ant mounds. Resmethrin is a rapid knockdown product with no little or residual activity. It has a relatively low order of toxicity to mammals. This device is being developed for the service sector (Private pest control operators) and/or the lease/rental market.

4. The hot water method is probably the oldest tactic used to eliminate fire ant mounds. Recent reports in the scientific literature (Tschinkel and Howard, 1980) reported that 3 gal of hot water (90^o C) produced about 60% control of treated mounds. In this trial, mounds were treated using a water heating device rented from Hotsy of Houston to treat mounds with 2-3 gal of very hot (+170^o F) water.

Prior to treatment, a random sample of 30 mounds in the testing area were measured, with height and width recorded. Mounds were numerous (probably of the multiple queen or polygynous type). Mounds were 10.03 (+ 3.91 SD) inches in diameter and 2.43 (+ 0.98) tall (N = 30). Soil at the pasture site selected was sandy.

During treatments, the time required for each of the devices/methods to treat each mound was recorded for a sample of mounds treated. Results of these treatments for the Earth Fire and Antser devices was subjected to a Student's t test to determine differences in required to treat each mound.

Each device was used to treat a minimum of 30 marked and staked red imported fire ant mounds, May 27, 1987. An additional set of 30 mounds were marked but left untreated as a control group. Mounds were inspected 7 and 21 days after treatment and rated for ant activity as either active or inactive. Mounds were considered active if worker ants displayed defensive behavior when the marked mounds were minimally disturbed. The presence of a few worker ants was not considered to indicate the presence of an active colony, and thus although noted, these mounds were analyzed as inactive mounds. Conditions in the area of treated mounds were also evaluated for phytotoxicity to vegetation and physical disturbance. During evaluation periods, the existence of "new" mounds within 3 ft of the treated mound were assumed to be satellite mounds resulting from surviving ants in the treated mound selecting a new mound site. Certainly, the possibility exists that some of these "satellite" mounds migrated close to treated mounds from other locations.

Rainfall, monitored during the treatment and evaluation period, was heavy (Table 1), flooding portions of areas containing treated mounds. These areas were abandoned and evaluations continued in areas where flooding did not occur.

For analysis, mound evaluations for each of the treatment and control groups were divided into five consecutive sets of five mounds each (Note: Although a minimum of 30 mounds were treated with each method or device, flooding conditions due to heavy rains prohibited all originally-treated mounds from being evaluated). Percent active mounds and percent of mounds with satellite mounds within a three-foot radius were calculated for each set. The resulting percent values were subjected to an analysis of variance using the Least Significant Difference test at the P < 0.05 level for each post-treatment evaluation date. Unfortunately, the 7-day post-treatment data for Earthfire was misplaced, and the analysis was performed with existing data.
 

The Yaard-vark required 30 minutes to treat each mound. Five devices were operating simultaneously using a portable generator, and thus treatment of 30 mounds required 3 hours (15 hours if only one device had been available). The Earthfire injection system required significantly less time to treat each mound than did the Antser, requiring 33.36 (+ 9.42) vs 44.55 (+ 4.16) seconds per mound (N = 22 and 11, respectively, D.F. = 31, t = -3.776; P = .0004). Hot water treatments were not timed since the water source was located remotely from the treatment site and individual five gallon buckets of hot water needed to be delivered via pickup truck. Producing large quantities of hot water is an obstacle for the practice large-scale use of this method.

All treatments produced an almost immediate effect of ants in mounds except the Yaard-vark. This device did produce a ring of dead ants at the base of the device while it was inserted into the mound. However, the number of mounds rendered inactive using this tool was not significantly different from those in the untreated sets of mounds throughout the observation period (Table 2).

The Antser produced 12 inch diameter circle of water-saturated mud, having obliterated all vegetation. Shortly following treatment, ants could be seen within the slurry attempting to rebuild the mound. The Earthfire Injection System eliminated all activity in treated mounds within several minutes of application. The use of this system is accompanied with a distinctive odor of the vaporized insecticide formulation. This odor was somewhat persistent, being detected in some of the treated mounds 21 days after application. At both post- treatment evaluation periods, grass around some treated mounds was yellow, indicating some phytotoxic reaction to the treatment. Hot water eliminated ant activity wherever ground was thoroughly soaked. This treatment also killed vegetation that was drenched. After 21 days, treated mounds could easily be spotted because they were surrounded by dead grass.

On the 7 and 21 day post-treatment evaluation dates, the percent of active mounds in sets treated with the Antser and hot water were found to be significantly lower than those of the untreated and Yaard-vark treated mounds, having achieved 80 to 96 percent control of ants colonies in treated mounds. However, the number of satellite mounds occurring in association with these treatments was significantly greater than those associated with the untreated mound sets, except for the hot water treated mounds at the 21 post-treatment evaluation date, ranging from 40 to 68 percent. Thus, "percent control" in the area treated with these methods may be less than that expressed by the percent elimination of ant activity in individually treated mounds when the presence of satellite mounds is taken into consideration.

Satellite mounds are commonly formed by surviving worker ants and may persist for several weeks until the worker ants perish. However, if the queen(s) and brood survive initial treatment as well, satellite mounds can persist and form new permanent mounds. Although brood was detected in some of the satellite mounds, their long-term survival was not monitored.

The Earthfire Injection System treatments resulted in a statistically significant reduction of 84 percent in mound activity relative to the untreated mound sets, but similar to reductions resulting from treatments with hot water or the Antser. Sixteen percent of the Earthfire treated mounds where found to be associated with satellite mounds 21 days after treatment but this value was not significantly different from that of the untreated mounds. Results of this treatment are statistically similar to those produced by the application of hot water.

In summary, these results should be considered to be preliminary. These methods or devices need to be tested under various conditions, including different soil types (clay) and environmental conditions (different times of the year) before their actual performance can be adequately documented. Excessive rainfall during the observation period may have confounded these results, causing an increase in the normal rate of colony migration and satellite mound formation.

________________

Tschinkel, W.R. and D.F. Howard. 1980. A simple, non-toxic home remedy against fire ants. Georgia Entomol. Soc. 15(1):102-105.
 

 Table 1. Rainfall in Harris County, Texas, from May 17 through June 17, 1987.
 

 Table 2.   Percent active red imported fire ant mounds and percent treated mounds with satellite (newly established) mounds within a 3 ft radius, 7 and 21 days after application of alternate fire ant control devices or methods, Harris County, Texas, May 27,1987.
 

^1/ Means in columns followed by different letters in columns indicate significant difference according to the Least Significant Difference test at P < 0.05.

Click  to view the entire report in ".pdf" format so the document can be easily and accurately printed.. Adobe Acrobat Reader will launch automatically. If you do not have Adobe Acrobat Reader, click