WO1995028834A1

WO1995028834A1 - Enhancement of insect control with attractant pheromones

Info

Publication number: WO1995028834A1
Application number: PCT/US1995/004906
Authority: WO
Inventors: Robert K. Vandermeer; Jack A. Seawright; Steve Brocious
Original assignee: The United States Of America, Represented By The Secretary Of The Department Of Agriculture; American Cyanamid Company
Priority date: 1994-04-22
Filing date: 1995-04-21
Publication date: 1995-11-02
Also published as: AU2392695A

Abstract

An attractant pheromone-enhanced bait composition has been discovered for the control of arthropods, particularly pest social insects. The attractant pheromone-enhanced bait decreases the percentage of bait taken by nontarget arthropods and reduces the amount of toxicant currently in use to control pest arthropods. Testing is done with an olfactometer having 2 choice chambers (2) attached to an ant entry (6).

Description

ENHANCEMENT OF INSECT CONTROL WITH ATTRACTANT PHEROMONES

Background of the Invention Field of the Invention

The present invention relates to species- and/or genera- specific attractant pheromone-containing bait compositions for social insects, particularly ants, and more particularly fire ants. It also relates to the use of these compositions to control social insects while decreasing the percentage of bait taken by non-target arthropod species or other organisms. Furthermore, the present invention also relates to the use of these compositions to control social insects in order to decrease the quantity of insecticide required.

Description of the Related Art

Discovering pesticides that are effective against a broad range of insect pests and that also can be used safely has long been a problem. One such problem area has been in the control of social insects such as ants, yellow jacket wasps, other pest wasps, and termites. Various species of ants pose significant problems for man from both an agricultural and a health care point of view. Leaf-cutting ant species can defoliate a citrus tree overnight. Argentine ants endanger crops by domesticating and protecting other pest insects such as aphids and scale. Fire ants are particularly destructive by stinging humans and livestock, feeding on germinating seeds and crop seedlings thereby reducing yields, and damaging electrical equipment and farm machinery which strike the ants' mounds.

Requirements for an effective pesticide formulation for the control of pest social insect species such as ants are very stringent because the reproductive forms (queens) of social insects are buffered from the effects of insecticides by a large worker force and their often closed nest structure. Thus, control of social insect pests is inherently different from control of non- social insects. For example, a mature fire ant colony may contain 250,000 sterile workers and a single queen. Only 10 percent of the workers are on the surface foraging for food. Insecticide treatment with a fast acting insecticide will not affect the 90% of the workers in the nest or the queen and the effect is negligible. In fact 95% of the workers can be killed, but if the queen is unaffected, the colony will come back. Thus, an effective social insect toxicant must exhibit delayed toxicity, not repel the insects and be effective over a range of concentrations. Repellency can reduce or negate the effectiveness of a toxicant because the insects will avoid the treated bait. The bait must be of a form that is transferable either by carrying it back to the nest or by trophallaxis and the toxicity must be delayed because foraging worker insects constitute only a small percentage of the total colony and must survive long enough to pass the toxicant onto the main colony population, especially the queen (s) .

Pheromones play an integral role in the activities of social insect colonies such as recruitment to food sources, care of brood and queen, and defense of the colony. Pheromones can be used to enhance the control of pest insects by decreasing the time for discovery of the bait by the targeted insect which results in less bait available to non-targeted insects. This in turn leads to a reduction in the amount of insecticide used and predictably slower reinfestation rates.

Currently, the most efficient way to use insecticides to control pest social insects is in the form of a bait. Toxic baits take advantage -of social insect foraging systems to direct the toxicant to the entire colony. Thus, smaller quantities of insecticide are used then would be required for the inundation of colonies with, for example, insecticide emulsions (drenches) . However, competing insect species are also affected by bait toxicants. The phagostimulant, e.g. soybean oil, is non-specific. The concurrent removal of a pest species and non-target species may actually enhance reinfestation of the pest species.

In order to improve toxic baits and reduce the amount of pesticides used, the specificity of the baits needs to be improved. Pheromones have been added to baits for different insects such as cockroaches (U.S. Patent No. 5,238,681), flies (U.S. Patent No. 5,237,774), leaf cutting ants (WO 90/11012), and mint root boorers (U.S. Patent No. 5,236,715) . A trail pheromone-containing bait was tested on leaf-cutting ants (Robinson et al, Bull, ent. Res. V.72, 345-356, 1982; Robinson et al, Bull. ent. Res., V.68, 159-170, 1978: and Jutsum et al, Bull. ent. Res., V.71, 607-616, 1981) . Although there was an increase in bait pick-up and attraction to the bait, it was found that the ants dropped the bait particles outside the entrance to the nest (Vilela et al, An. Soc. ent. Brasil, v.l7(supl), 107-124, 1988) . Therefore, the bait was not ingested and/or transferred from one ant to another in the nest. House (WO 90/11012) discloses an alarm pheromone-containing bait which contains either 3-octanol alone, 3-octanone alone, or 3- octanol plus 3-octanone. 3-octanol and 3-octanone are two components of the alarm pheromone of the leaf cutting ant Acromyrmex octospinosus . However, these two components can be repellent to other arthropods including social insects such as the fire ant, Solenopois spp.

While various pest social insect pheromones have been fully identified, there remains a need in the art for genera- or species- specific pheromone-containing baits for the control of pest social insects. The present invention provides a bait formulation which is different from prior art baits.

Summary of the Invention

It is therefore, an object of the present invention to provide an attractant pheromone-containing bait formulation for the control of pest arthropods.

Another object of the present invention is to provide a racemic attractant pheromone mixture in a matrix which further includes a toxicant and phagostimulant.

Still another object of the present invention is to provide an optically pure attractant pheromone in a matrix which further includes a toxicant and phagostimulant.

A further object of the present invention is to provide a bait containing an attractant pheromone wherein said pheromone is imbedded in a controlled release formulation.

A still further object of the present invention is to provide an attractant pheromone-containing bait wherein said pheromone is exterior or interior to a matrix containing a toxicant and a phagostimulant.

Still another object of the present invention is to provide a method of controlling pest arthropods using an attractant pheromone-containing bait.

Another object of the present invention is to provide a method of controlling pest social insects using an attractant pheromone- containing bait.

Further objects and advantages of the invention will become apparent from the following description.

Brief Description of the Drawings

Figure 1 is a drawing of a Y-tube olfactometer bioassay used in detecting attractive pheromones.

Figure 2 is a bar graph showing Y tube olfactometer results for queen poison sac extracts and synthetic queen pheromone, racemic invictolide, at various concentrations and on various carriers .

Figure 3 shows the structures of the three components isolated from fire ant queen pheromone.

Figure 4 is a bar graph showing olfactometer bioassay of the nine possible three component mixtures of queen pheromone.

Figure 5 is a bar graph showing an olfactometer bioassay of 15 possible two component combinations of queen pheromone.

Figure 6 is a graph showing activity/concentration studies of a three component and a two component mixture of queen pheromone.

Figure 7 is a graph showing activity/concentration studies of a two component mixture and racemic invictolide (±) B.

Figure 8 is a graph showing activity/concentration studies of the three, two and one component mixtures.

Figure 9 is a graph showing percent fire ant workers with calico red-dye after the indicated time interval after exposure to dye treated bait and dye-treated bait containing 100 QE per particle of queen pheromone components A and racemic invictolide.

Figure 10 is a bar graph comparing polygyne and monogyne S. invicta worker olfactometer response to poison sac extract and control (PDCG) . Figure 11 is a bar graph showing mean discovery time for AMDRO® bait particles versus bait particles treated with 100 queen equivalents (QE) of racemic invictolide.

Figure 12 is a bar graph showing the percent AMDRO or pheromone enhanced AMDRO discovered first in a series of bait discovery assays.

Figure 13 is a bar graph showing a comparison of bait particle discovery time for standard bait and bait enhanced with various concentrations of racemic invictolide.

Figure 14 is a bar graph showing first discovery preference for bait particles treated with 1 queen equivalent of racemic invictolide versus standard bait particles. Site 1 = Polygene Population, Site 2 - Monogyne Population, Site 3 = Mixed Monogyne & Polygyne Population.

Figure 15 is a bar graph showing first discovery time for bait particles treated with 1 queen equivalent racemic invictolide versus standard bait particles. Site 1 = Polygene Population, Site 2 = Monogyne Population, Site 3 = Mixed Monogyne & Polygyne Population.

Figure 16 is a graph showing laboratory results (expected) of hydramethylnon (at 33% of recommended rate) efficacy when the theoretical amount of active ingredient was taken by a colony and field efficacy results (observed) where fire ants had to discover individual bait particles broadcast in the field at the same rate.

Figure 17 is the same as Figure 11 except pheromone enhanced bait results have been added.

Figure 18 is a bar graph showing results of a large scale broadcast field bioassay.

Figure 19 is a graph showing cumulative number of fire ants collected in pitfall traps located on the three treatment and control sites.

Figure 20 is a graph showing cumulative number of non-target ants in pitfall traps located on the three treatment and control sites .

Figure 21 is a bar graph showing comparisons of the cumulative pitfall trap data at the end of the experiment. Figure 22 is a graph showing chemical stability and biological activity of racemic invictolide-enhanced AMDRO bait particles. B = racemic invictolide.

Detailed Description of the Invention

The fire ant, Solenopsis invicta, is used as a model system. However the concept is applicable to the control of other pest ants and other pest social insects. In fact, social insects, such as wasps and termites, have well developed foraging mechanisms. Social insect pests are treated in accordance with the invention by dispensing the attractant pheromone-containing bait formulation in any suitable way in the vicinity of nests, in urban or rural areas, or anywhere pest social insects are a problem. For example, by scattering attractant pheromone-containing bait particles on the foraging surface surrounding a pest social insect nest.

The term bait is understood by those skilled in the art to be a combination of phagostimulant, insecticide, and a suitable carrier. The phagostimulant may be any substance that will entice the insect to ingest the toxicant. Suitable baits include edible oils and fats, vegetable seed meals, meal by-products such as blood, fish meal, syrups, honey, sucrose, and other sugars, peanut butter, cereals, amino acids, proteins, and the like. See U.S. Patent No. 3,220,921 which is herein incorporated by reference. Preferred phagostimulants for ants are mixtures of edible oils, which are also solvents for the toxicant, with granular carriers such as corncob grits, pregel defatted corn grits (PDCG) , and the like. These impregnated granular bait formulations readily fall to the ground, where insects forage, when dispersed by aerial or ground applicators. When found by the insects, the phagostimulant and toxicant is ingested, or the- bait particle is carried into the nest. In both cases the phagostimulant and toxicant are distributed to other colony members, most importantly the queen.

The toxicant or pesticide included in the bait can be any toxicant or insecticide or insect growth regulator effective against pest social insects as long as it is not repellent to the targeted insect, exhibits delayed toxicity, and is readily transferable from one insect to another. See for example, U.S. Patent No. 5,177,107, which is herein incorporated by reference. Environmentally acceptable insecticides and insect growth regulators with desired physical and toxic properties have been registered with the EPA for use in non-agricultural situations. These are amidinohydrazone, AMDRO, fenoxycarb, Logic® and abamectin, Affirm©. The toxicant component of AMDRO, disclosed in U.S. Patent No. 4,152,436, herein incorporated by reference, is used in all of the following examples as the toxicant component of the bait. The pheromone can also be added to the individual particles of AMDRO as detailed below. However, one of ordinary skill in the art could readily substitute any suitable toxicant in the bait formulation.

It has been found that components of a queen recognition pheromone of a social insect pest will attract the worker insects to a bait containing a toxicant and the worker insects will carry the toxicant back to its nest where the toxicant is distributed and kills the other insects. This is unexpected because no attractant pheromones have yet to be shown to be effective adjuvants to social pest insect control (wasps; bees; termites; ants) . In the case of ants and termites, the workers that forage for food are terrestrial (they do not fly) and; therefore, the pheromone utilization technology already developed for flying insects is of little or no use. In all social insects the reproductive females are few compared to the non-reproductive workers (for fire ants 1:250,000) thus the workers buffer or shield the reproductive queens. For non-social insects, each individual affected by an insecticide or pheromone decreases the reproductive potential of the pest population. In pest social insects, for example, fire ants, no more than 10% of the non-reproductive worker force is on the surface searching for food. Pheromones or insecticides or combinations that only affect these foraging workers have little or no effect on the reproductives (queens) or the colony as a whole. It should be noted that there is a use and need for a quick acting insecticide/pheromone combination that could be used to detect incipient pest social insect populations in areas where infestations are currently light or non-existent. Most social insect workers undergo age related polyethism and social insect behavior is largely regulated by chemical cues (pheromones) . Observations indicate that foraging workers are not very sensitive to physiological concentrations of their queen recognition pheromone. It was also not known if workers would respond to a bait with queen-related activities or with phagostimulant activity. Ants and other social ' insects exhibit what is called age polyethism or a process of temporal castes. The within colony tasks of an individual worker change as that worker becomes older. There are three defined age related temporal castes. The youngest are newly eclosed workers which tend brood and the queen. They are called nurse or broodtending workers and experience brood and queen related pheromones, but not recruitment pheromones. As workers age their behavior and main tasks change. The next defined caste is the reserve temporal caste. This caste normally maintains the nest structure and would be recruited from the colony nest to retrieve food. The reserve caste may experience brood or recruitment pheromones depending on their developmental stage. The last defined temporal caste category is the foraging worker who finds food and recruits other workers to do it if necessary. Foraging workers are often called the expendable caste because they are old and close to death. This caste is not very sensitive to any pheromone system due to senescence and the accompanying diminished sensory acuity. For example, foraging workers were significantly slower to respond to the recruitment pheromone when compared to their broodtending counterparts. Therefore, it was unexpected that, for example, the use of a queen recognition pheromone would both attract foraging workers to a bait and that after discovering the bait that the worker ant would feed on the phagostimulant in the bait and take the bait back to the nest and distribute the phagostimulant and insecticide.

In the test model system, the fire ant queen recognition pheromone component used is a synthetic pheromone called invictolide. Invictolide is an aliphatic δ-lactone, tetrahydro- 3, 5-dimethyl-6- (1-methybutyl) -2H-pyran-2-one. Invictolide can be used as a racemic (±)preparation or as an optically pure preparation of (-) -invictolide. The synthesis of racemic invictolide is disclosed in Tetrahedron Letters, Volume 24, No. 18, 1893-1896, 1983, which is herein incorporated by reference. Racemic invictolide can also be purchased from Nitto Denko Corporation, Japan. Optically pure invictolide can be synthesized as described by Mori et al, Tetrahedron, 42(23) 6459-6464, 1986, which is herein incorporated by reference. Invictolide can be used alone in the bait or can be used in combination with other components of the queen recognition pheromone, (E) -6- (1-pentenyl) -2H-pyran-2-one or dihydroactinidiolide. (E) -6- (1-pentenyl) -2H-pyran-2-one can also be purchased from Nitto Denko Corporation, Japan, or can be made synthetically as described in Tetrahedron Letters, Volume 24, No. 18, 1889-1892, 1983, which is herein incorporated by reference. Dihydroactinidiolide can be made synthetically as described in Tetrahedron 42(1), 283-290, 1986, which is herein incorporated by reference. Invictolide is used in a concentration range of from approximately about 0.1 queen equivalents to approximately about 250 queen equivalents per bait particle. A more preferred range is from approximately about 0.3 queen equivalents to approximately about 100 queen equivalents per bait particle and a most preferred range is approximately about 1 queen equivalent to approximately about 100 queen equivalents per bait particle. The most preferred concentration for bait formulation with fire ant queen pheromone is approximately about 1 queen equivalent per bait particle. With higher amounts of pheromone, there is a trend toward faster discovery of the pheromone-enhanced bait. However, since 1 queen equivalent per bait particle gives consistent results, it will be a cost-effective concentration for commercial purposes.

At one queen equivalent per bait particle, 1 gram of pheromone is enough to treat 272 acres. When two or more components of the queen pheromone are used, the total mixture will fall within the above ranges. 1 queen equivalent is defined as the estimated amount of a compound that a single queen contains at any given time (See Rocca et al . and Glancey et al . , Sociobiology, 9(1), 19-30) . It was found that queens contain approximately 4ng/queen of invictolide. The natural material is optically pure; however, the synthetic invictolide used in the examples is a racemic mixture where only one of the two enantiomers is active. Therefore, one queen equivalent for racemic invictolide equals 8ng, rather than 4ng of the optically pure invictolide. For a two component pheromone bait, 1 queen equivalent per bait particle for each pheromone is added. Therefore, 1 ng of (E) -6- (1-pentenyl) -2H- pyran-2-one and/or 1 ng of dihydroactinidiolide plus 8 ng of racemic invictolide equals 1 queen equivalent for the combination of pheromones. One of ordinary skill in the art could readily substitute any attractant pheromone in the bait formulation. Controlled slow release of the attractant pheromone may be effected by .encapsulation or absorption into a porous substrate that is incorporated into the bait or by any other controlled release technology known to those of ordinary skill in the art. See for example, Paul, Controlled Release Polymeric Formulations, ACS Symposium, Series #33, 1-14, 1976, and Lewis et al, Controlled Release Pesticides, ACS Symposium, Series #53, 1-16, 1977. Both references are herein incorporated by reference.

Granular carrier particles are sieved through American standard sieves #10 (2.0mm) and #12 (1.7mm) . Particles remaining in the #12 sieve are used to prepare the pheromone-containing baits. Next, the neat pheromone (s) is (are) weighed and an appropriate amount (defined by 1 QE/0.61 μl solution) of soybean oil with toxicant is added to give a solution of 1 queen equivalent per 0.61 ul to be added per granular carrier particle. For the test model, the toxicant is hydramethylnon which is the active ingredient in AMDRO (American Cyanamid Corporation, Princeton, New Jersey) bait particles. It was determined that a concentration of 0.26% (wt/wt) of hydra ethylnon/grit can be used in an attractant pheromone-enhanced bait to achieve effective control of fire ant colonies. This is 33.3% of the currently recommended amount of hydramethylnon for non-enhanced baits. One of ordinary skill in the art could readily determine the effective amount of any toxicant needed for an attractant pheromone enhanced bait. If bait particles containing toxicant are used, the pheromone is diluted in the appropriate amount of soybean oil to give 1 queen equivalent per 0.61 ul of soybean oil to add to each toxicant containing particle. The following examples illustrate the use of queen recognition pheromone-containing baits for control of pest social insects using the fire ant as a test model system. They are intended to further illustrate the invention and are not intended to limit the scope as defined by the claims.

EXAMPLE 1

The responsiveness of worker ants to queen pheromone release was tested using an olfactometer bioassay and different release medium.

The olfactometer (See Figure 1) consists of two 24/40 ground glass joints 2; each ring 7 is sealed to one of the arms 8 of a 5 cm Y tube 9 such that 1 cm of each Y tube arm 8 extends through the male half of one of the ground glass joints 4. A 5 cm piece of 0.6 cm ID tubing 5 was ring sealed 1 cm into the female half of the ground glass joints. Baffle 3 at the center of the Y tube controls air streams and prevents mixing of the sample, and gives ants a clear choice. The baffle also narrows the openings to the choice chambers to the minimum size required for passage of a major worker. A test sample of the extract and a solvent blank is applied to various release media and each is placed in one of the choice chambers 2. The sample concentration tested was 30 queen equivalent of racemic invictolide (±B) . The carriers used were AMDRO bait particles (AM CY BAIT) , pregel defatted corn cob grits (PDCG) , and filter paper. The controls were AMDRO bait particle alone, PDCG plus soybean oil (PDCG + SBO) , PDCG, and filter paper with 0.3QE of poison sac extract. Compressed air (breathing air quality) is split into two stems and passes into the two chambers 2. Each stream is regulated to 0.2 liters/ minute for a total effluent flow rate of 0.4 liters/minute. Approximately 50-70 ants from laboratory colonies are confined in a 2.5 cm piece of 0.9 cm ID TYGON tubing sealed at one end with wire gauze. The open end of the tubing is attached to the entrance stem 6. The initial choice of the first 20 ants that walk down the entrance tube and into one of the arms of the Y tube is recorded. Ants that are not trapped in a choice chamber and come back to the entrance stem are not counted if they make another choice. After each test, the olfactometer is rinsed with acetone and dried. Each test sample is retested with the same caste of ants from the same colony, but the choice chamber in which the sample and control is placed is reversed. This eliminates any bias inherent in the individual olfactometers . As seen in figure 2, workers respond equally well to pheromone regardless of the carrier. It can also be seen that, once refined, soybean oil (SBO) contains volatile attractants. The effect of adding queen pheromone to the grit/soybean oil mixture significantly increases the worker attraction over the grit/soybean oil alone but the two results are not totally additive.

EXAMPLE 2

Although it would be expected that foraging workers would not respond well to queen pheromone, example 1 demonstrates that foraging workers are attracted to the pheromone.

Queen pheromone is released by the queen and contains three components designated A, B, & C (see figure 3) . In the past, the attraction behavior of the ants was always to queen pheromone bioassayed within the context of other behaviors. Therefore, this example only measures worker response to volatiles of A, -B, +B, -C, and +C on AMDRO bait particles in an air stream in the above described olfactometer bioassay.

For an olfactometer result to be statistically significant (P < 0.05) the percent response must be greater than 65%. There are several mixtures that are significantly attractive. Figure 4 shows the results for the nine possible three component mixtures presented to the ants at 1 queen equivalent (lng A:4ng B: lng C; if racemic B or C were used the amounts were 8ng and 2ng, respectively) . From a commercial point of view A & (-) (+)B & (-) (+)C is the most desirable three component mixture since the synthesis of costly optically pure compounds is not necessary.

However, a two component system would be even better; ^■therefore we evaluated the 15 possible two component combinations as shown in figure 5 (at 1 queen equivalent) . Three mixtures showed significant activity. Two of these, (-)B & (-)C and (-) (+)B &(-)C have undesirable high standard deviations, indicative of inconsistent results. In contrast, A & (-) (+)B demonstrated good attraction and consistent results from replicate to replicate. In fact, the results of A & (-) (+)B are better than those for three component system, A & (-) (+)B & (-) (+)C. At one queen equivalent there were no active single components.

Although reducing the number of components from three to two and determining that optically pure components are not necessary is a big step forward, another important question is what is the concentration/activity profile? Ideally, we want significant activity over as broad a concentration range as possible, because if the activity is over a narrow range of concentrations, it will be difficult to formulate the material with the correct release rate.

Figure 6 shows the results of activity/concentration studies for the three component [A & (-) (+)B & (^_) (+)C] and two component [A & (-) (+)B] pheromone systems. Interestingly, the two component system has a slightly broader range of activity than that of the three components (racemic B was made by combining equal amounts of the two optical isomers) .

Although racemic B did not have statistically significant activity at one queen equivalent, tests at higher concentrations showed it to have significant attractive activity over about one and a half orders of magnitude (Figure 7) . This is an important result because it may now be possible to work with only a single racemic compound, which enhances the potential commercial viability of the queen pheromone.

Figure 8 shows a comparison of the three, two and one component mixtures.

EXAMPLE 3

In order to show that worker ants will exhibit phagostimulant activity when attracted to the queen pheromone-containing bait, the following study was conducted. Pregel defatted corn grits were loaded (20% wt/wt) with soybean oil containing 1% calico red, an oil soluble dye. This formulation was treated topically with a hexane control (10 ul) or with 10 ul of 100 queen equivalents of queen pheromone components, (E) -6- (1-pentenyl) -2H-pyran-2-one (A) and racemic invictolide (+B) . The two component mixture was used at a high concentration to maximize the chances of seeing a negative phagostimulation effect. Small subcolonies of fire ants were set up with 400-500 workers and brood. Two sets of sub¬ colonies were set up for both control and treatment. One was sacrificed two hours after exposure to the treated baits and the other four hours after exposure. There were three replicates of each. The colonies were maintained in porcelain trays with a small colony cell. The grits were scattered about the sub-colony tray for the ants to pick up or be available for contact. At the end of the designated time period the colonies were frozen, then 100 workers were randomly selected. They were sandwiched between two pieces of filter paper and squashed with a heavy duty roller. The results were evaluated by visually scoring the number of workers containing dye. The results are shown in Figure 9. After 2 hours, 61.5% of the control and 48.0% of the pheromone bait exposed workers contained dye. After four hours of exposure between 75 and 80% of the workers contained dye. Only 10-15 worker ants originally participated in the discovery and removal of the bait particles, thus the fact that after two hours almost 50% of the pheromone treated colony workers had dyed oil in them (100-200), means that significant trophallaxis occurs, even though it was less than control (61.5%) . After four hours exposure no differences were observed.

Workers at the pheromone treated baits appeared to remove them at a slower rate than the control baits. This may explain the initial differences seen after two hours, which disappears at four hours after exposure. A non-quantitative look at brood showed that the oil is distributed to the im atures in both cases. Therefore, this example shows that oil in both control AMDRO bait and AMDRO bait treated with pheromone is taken into the colony and distributed to nestmates. Furthermore, this experiment shows that ants are hardwired and respond to a single stimulus at a time. An ant follows an attractant concentration gradient without knowing what it is going to find. Once the ant comes to the source of the attractant, the ant receives additional stimuli. If the source is a queen, the ant receives- a queen stimulus; if it is a food source; phagostimulant activity is induced regardless of the pheromone present. This was not an obvious result considering previous literature.

EXAMPLE 4

The fire ant Solenopsis invicta has two distinct population types, monogyne and polygyne, which relate to the number of queens per nest. Monogyne nests have one queen and polygyne nests have many. It is important to know how both types of populations respond to baits. A Y tube olfactometer bioassay as described in Example 1 was set up to test both types of workers. 0.3 queen equivalents of poison sac extracts, which contains queen attractant pheromone, was used. As shown in Figure 10, worker response to queen pheromone (poison sac extract) is independent of the type of population it came from. Similarly, worker response to the blank carrier, pregel defatted corncob grits (PDCG), was not different for the two types. Monogyne and polygyne workers were significantly attracted to queen pheromone in the olfactometer. However, workers were not significantly attracted to the untreated grit carrier.

EXAMPLE 5

The next step was to test the grit/pheromone formulation in the field. A field bioassay was developed that measured discovery time of two paired bait particles, one standard and the other treatment.

Experimental sites were large grassy areas >l/2 ha and consisted mainly of bahia grass. The grass was maintained on a regular basis through mowing. The area was determined by inspection to be infested with fire ants. The onogyne/polygyne status of the fire ant population was determined by the aggression bioassay (Morel et al, Annals of the Entomological Society of America 3(3), 642-647, 1990) and by observation of mound characteristics and worker size distribution. The two test sites utilized were (a) Lake Alice Field on the University of Florida campus, Gainesville, Florida and (b) Santa Fe Community College site, Gainesville, Florida.

AMDRO fire ant bait was sieved through American standard sieves #10 and #12 as described above. Particles remaining in the #12 sieve were treated with 1QE racemic invictolide or hexane control in groups of five. The time to first ant discovery was recorded for each pair of bait particles. Worker ant contacts were recorded until one ant remained on the particle at least 3 seconds, after which the bait particle and ant were aspirated into a 2 ml vial and saved. The test was terminated after both particles had been discovered. If no contact (both particles) was made within 10 minutes the test was terminated. Similarly, if there was no contact at the second particle 15 minutes after the first particle was contacted the test was terminated.

There was no significant difference between the results from the 2 sites so the results were combined. The mean discovery time for pheromone treated AMDRO bait particle was 30 percent less than the discovery time for the standard AMDRO bait particle. The data were not normally distributed and were log transformed. Comparison by paired t-test showed the differences to be significant (D.F = 104; t₂._r.; P<0.005) . The data is presented in Figure 11.

EXAMPLE 6

Example 5 demonstrates that addition of attractant pheromone to a bait particle significantly decreases the time it takes fire ants to discover it. However, the pheromone was applied to already formulated AMDRO bait particles, a procedure that is not commercially viable. All subsequent paired bait particle assays were formulated by a procedure that would be used for commercial purposes. Neat pheromone was weighed and an appropriate amount of phagostimulant oil was added to give a solution containing IQE of racemic invictolide in 0.61 ul. Pregel defatted corn grits were sieved as described in Example 1. Then 0.61 ul of the pheromone/oil formulation were applied to each grit. Lower concentrations of pheromone in oil were prepared by diluting the 100 QE/0.61 ul solution with the phagostimulant oil. Four concentrations of 100, 10, 1, and 0.33 queen equivalents were evaluated. The results are shown in Figures 9 and 10. The percent bait discovered first for all but the 0.33 queen equivalent enhanced bait (Figure 12) was greater than 70% The 0.33 queen equivalent bait was below 60%. Fifty percent is expected if there was no effect. The mean time of discovery for all concentrations is shown in Figure 13. The results indicate that the mean discovery time of pheromone enhanced bait was considerably faster than the non-enhanced bait (44.8, 49.1, 40.4, and 27 percent faster, respectively, than the standard bait) . The differences were significant for all treatments, except the 0.33 queen equivalents. The results indicate that although there may be a trend toward faster discovery of pheromone enhanced bait at 0.33 queen equivalents, it is considerably diminished compared to higher concentrations. The 1 queen equivalent treatment gave consistent results and should be adequate for commercialization.

For large scale pheromone/bait preparation an appropriate amount of pheromone is added to and dissolved in soybean oil or any other appropriate phagostimulant/active ingredient solvent such as corn oil or peanut oil. This is thoroughly mixed and sprayed onto an appropriate carrier such as pregel defatted corn grits (PDCG) via a hand charged sprayer or other commercially available apparatus. The spraying process is conducted while the carrier is being mixed by rotation or other suitable means. After the specified amount of pheromone/phagostimulant is added to the carrier, the formulation is mixed for another 1 to 4 hours to insure uniform distribution.

EXAMPLE 7

In order to determine if bait discovery enhancement is dependent on location or the type of fire ant population (monogyne or polygyne) , the procedure of Example 6 was repeated using 3 locations which had different types of fire ant population. The sites were (1) Gator Getaway (development southwest of Gainesville, Florida) which has a polygyne population of fire ants; (2) Lake Alice perimeter (University of Florida campus) which has a monogyne population; and, (3) SW 23rd Terrace which has an unpredictable mixture of monogyne and polygyne colonies.

Racemic invictolide enhanced AMDRO fire ant bait and standard AMDRO fire ant bait were prepared as in Example 5. The particles were treated with 0.61 ul of soybean oil with hydramethylnon containing 1 queen equivalent of the pheromone. Controls were treated with 0.61 ul of soybean oil with hydramethylnon.

The results are shown in Figures 14 and 15. There were no differences in the percent of worker fire ants discovering the pheromone enhanced bait particle first based on site or on their polygyne/monogyne status (Figure 11) showing that pheromone enhancement works equally well regardless of location or population type. Similarly, there were no statistical differences (paired t- test) between the times of discovery, for pheromone enhanced bait particles or standard bait particles, for the three sites (Figure 12) . Since there were no significant differences between sites, the data were combined and analyzed. The combined data shows a highly significant difference between the enhanced bait versus the non-enhanced bait (t = 2.7768; DF = 62; p = 0.0037) .

EXAMPLE 8

To determine fire ant foraging efficiency, tests were conducted comparing AMDRO bait and pheromone enhanced AMDRO bait. A comparison of expected mortality curves from controlled laboratory studies where colonies took in 100% of the active ingredient presented to them to those of broadcast single mound field treatments at identical rates showed dramatically different^' results.

The laboratory studies were set up as described above in Example 3. The AMDRO bait contained 33.3% of the standard hydramethylnon concentration.

The field studies were conducted on pasture land. The pasture was first observed for the type of fire ant population present (monogyne versus polygyne) . Mound size, height, and worker ant size distribution were recorded for each field site to get a preliminary idea of the population. Mature polygyne populations have primarily minor workers with only a few major workers, whereas, colonies in monogyne populations have a range of worker sizes from minor to major. In addition, small sub-colonies from at least three different mounds were taken from various locations in each of the test fields. The sub-colonies were tested for their monogyne or polygyne status via an aggression assay (Morel, et. al, Annals of the Entomological Society of America 3(3), 642-647, 1990) . From the field observations and the aggression assays, the field could be reliably determined to be either monogyne, polygyne, or a mixed population. Mixed population fields and fields where populations of Solenopsis invicta and ____ σeminata were found together were not used in this study.

Once a test site was found to be satisfactory, 30 to 40 mounds were surveyed by observing the mound size (height and diameter) , then the population index of each colony was estimated by digging up a portion of the mound. A shovel was pushed into the mound from one side towards the center of the mound tumulus and the shovel lifted gently to observe the depth of ant activity and the density of ants' above and below the shovel slice. The mound depth, mound size, and ant density were all recorded, as were the qualitative variables - the presence or absence of worker brood, the presence or absence of sexual brood or alates, and general shape of the mound. This information and the mound location was used to match mounds and to calculate a Population Index for each mound. The Population Index was calculated using the categories in Table 1.

TABLE 1

Criteria Used to Calculate the Population Index for Individual Mounds

NUMBER OF WORKERS NUMERICAL RATING

Less than 1000 ants 0 1000 to 10, 000 ants 10 with brood 20 10,000 to 50,000 ants 20 with brood 40 50, 000 to 100, 000 ants 30 with brood 60 100, 000 to 150,000 ants 40 with brood 80 150,000+ ants 50 with brood 100

Modified from Banks and Lofgren 1991.

The mounds of a matched pair had to be at least 15m apart. For each selected mound, the area to be treated was physically marked with stakes. The target mound was in the center of a 15m square defined with four corner stakes. The square was oriented with the four faces perpendicular to the cardinal compass points. Maps of the 15m square to be treated were created on a grid where 1/2 inch = 5 feet. The left hand bottom corner (the southwestern corner) served as the origin (0,0) . All nontarget mounds were plotted on this map to aid in monitoring mound movement after treatment. Baits were prepared using blank corn grits, hydramethylnon soybean oil, and blank soybean oil provided by American Cyanamid Company. The hydra ethylnon-oil was diluted to 33.3% active ingredient using the blank oil. Half the 33.3% hydramethylnon strength oil was dosed with racemic invictolide (neat) at approximately 3.33 QE per bait particle. The oils were then added to the blank corn grits at a 20% by weight loading, shaken thoroughly, and left to equilibrate at room temperature overnight or longer.

One of the mounds of a pair was randomly assigned a treatment, AMDRO or AMDRO plus pheromone. The second mound of the pair received the opposite treatment. The pairs were treated at as close to the same time as possible, usually within 15 minutes of each other. Applications were made with an inexpensive broadcast spreader (Central Quality Industries, Polo., Illinois) that had the hopper and opening significantly reduced. Prior to treatment, flags were positioned starting ca.l meter from one apex and then ca. every 2 meters until reaching the other side (application on that side was also ca.l meter from the edge) . Eight passes were made with the applicator, guided by the flags marking two opposite edges of the square. Bait was applied at the outer edges first, moving inward toward the center of the plot until the final two passes were delivered on either side of the mound. Bait was applied at a rate in agreement with the dose specified on the label (between 1 and 1.5 lbs per acre) .

Mounds were checked at regular intervals after application to look for changes in the Population Index. Generally, evaluations were made after 3, 7, 10, 14, 17, and 21 days. Twenty one days was usually enough time to decrease the mound population below 10%, at which time the experiment terminated. The Population Index was determined as described above. Each mound was considered separately during the evaluation process (the rank of one mound in a pair was not linked with the other) . Because the mounds were not paired in order, the treatment of the mound was not known at the data readings, so the observations were made without knowledge of the treatment used. Population readings in the field were also taken independently of their previous reading, then checked afterwards for any gross inconsistencies with previous data. This was necessary because fire ant colonies will change location for a variety of reasons, including physical disturbance and environmental changes (temperature, rainfall, decreased competition from other colonies, etc.), thus the maps marking the starting position of all colonies within a treatment were extremely important. Generally, ten replicate pairs were set up per experiment. The use of 3.3 queen equivalents, which is a three¬ fold increase over the amount successively tested, was to enhance the bait's efficacy.

Figure 16 shows that fire ant colonies in the field are retrieving less than one third of bait presented to them when compared to laboratory generated concentration/activity curves. This may not be so surprising when one considers that at the application rate used (one pound of bait per acre) , there are 460,000 bait particles per acre. This requires 460,000 independent discoveries by the foraging workers to achieve 100% efficiency. When pheromone is added to bait, the fire ant is discovering and retrieving a higher percentage of the pheromone enhanced bait compared to non-enhanced bait as can be seen in Figure 17. Thus, pheromone enhancement does not just get the fire ant to bait particles faster than they normally would, but it also increases the number of bait particles discovered.

The degree of improved active ingredient recovery with pheromone enhanced bait was estimated by comparison with controlled laboratory concentration/activity studies using hydramethylnon. The foraging fire ant picked up less than 20% of the bait. However, pheromone enhancement almost doubles the amount of active ingredient retrieved. Based on these results alone, the concentration of hydramethylnon in currently used fire ant baits could be reduced in half if they were pheromone enhanced!

EXAMPLE 9

Large scale broadcast field bioassays were performed on half acre fire ant infested field sites. Treatments were full strength AMDRO, AMDRO containing one third the standard amount of active ingredient (1/3 Al), AMDRO containing one third the standard amount of active ingredient plus racemic invictolide (IQE) and untreated control. The bait was applied via standard methods (Banks and Lofgren, J. Entomol. Sci. 26(3), 331-338, 1991, herein incorporated by reference) . Three replicates of each treatment and control were applied. Evaluations of the treatments and control were made by pretreatment and posttreatment determinations of the population index of the center (Banks and Lofgren, J. Entomol. Sci. 26(3), 331-338, 1991) of each plot. In addition, pitfall traps (8 per plot) were set out for 24 hours twice a week, throughout the test period to determine the effect of the treatments on non-target ant species and to follow the decline in the target, fire ant, population. Pitfall traps results are another measure of population.

Figure 18 shows the population reduction 43 days after treatment. Significant reduction in the fire ant population occurred for all treatments; however, full strength AMDRO and one third the standard amount of active ingredient plus racemic invictolide 1/3AI+P) gave better population reduction than simply the 1/3 Al AMDRO.

The same trend is exhibited when the cumulative number of fire ants collected in pitfall traps after 60 days is evaluated (Figure 19) . All non-target ants collected in pitfall traps were lumped together and plotted in Figure 20. Control and the 1/3 Al+Pheromone treatment show a steady increase in non-target ants collected, whereas the 1/3AI formulation shows little or no non¬ target ant increase, which is indicative of detrimental effects on the non-target ant species. This is also demonstrated in Figure 21, which compares the cumulative pitfall trap data at the end of the experiment. The mean and standard error of the three replicates are shown. It is clear that the number of non-target ants collected in pitfall traps over the period of the experiment is greater for the pheromone enhanced baits than for the other two treatments. This again demonstrates the beneficial effects of pheromone enhanced bait.

EXAMPLE 10

The above examples demonstrate the successful implementation of attractant pheromone-enhanced baits. Another factor to consider is the storage longevity of these pheromone enhanced baits. Pheromone enhanced bait particles were prepared as described above using racemic invictolide at 1 queen equivalent. The particles were placed in 2ml screw cap vials and capped with a foil lined lid. The vials were stored at room temperature (24-27°C) . Periodically the samples were removed for chemical analysis and olfactometer bioassay.

The chemical analysis was carried out as follows. A bait particle was placed in a small tissue grinder tube and about 0.5ml of hexane added. Then lOOul of a 0.0001% hexane solution of n- tetradecane was added as an internal standard. The bait particle was ground thoroughly and the hexane extract filtered through a Pasteur pipet plugged with sylanized glass wool. The filtered solution was evaporated under nitrogen to about lOOul and transferred to a GC autosampler vial containing a lOOul insert. Samples were injected into the GC via a Leap Technologies Autosampler. Gas chromatograph (GC) analyses were carried out on a Varian 3700 gas chromatograph (Walnut Creek, CA) equipped with a flame ionization detector, split/splitless injector, and a 30 m x 0.32 mm ID, DB-1 fused silica capillary column (J&W Scientific, Inc., Rancho Cordova, CA) . Helium was used as the column carrier gas, and nitrogen was used as make-up gas. The oven temperature program was 100°C to 150°C at 5°/min. The detector temperature was set at 300°C and the injector temperature set at 170 C. The chromatograms were printed and peak areas calculated on a Maxima 820 data processor. The accuracy of peak integration was checked by re-plotting the chromatograms and manually optimizing the chromatogram baselines. Standards of racemic invictolide and tetradecane were run before and after the samples on a given day, to determine the response factor to be used in quantitating the amount of queen pheromone in each sample. The data were analyzed using Statview II on a Macintosh IlCi computer. No precautions were taken to exclude oxygen from the vials. As can be seen in Figure 22, the pheromone is stable in the formulation for at least 3 years and the biological activity remains high and does not diminish after 3 years. Therefore, the attractant pheromone bait formulation has a long shelf life and can be prepared and stored prior to use. The foregoing is for the purpose of illustration. The invention can be applied to other pheromone systems and/or other appropriate insecticides. Such detail is solely for that purpose and those skilled in the art can make variations therein without departing from the spirit and scope of the invention.

INDEX OF ELEMENTS

1. Air Inlet Tube

2. Sample Chamber

3. Baffle

4. Ring Seal Tube (front)

5. Ring Seal Tube (rear)

6. Entrance Stream

Claims

We claim:

1. A composition for the control of a population of pest arthropods comprising

(a) an effective amount of a pest arthropod attractant pheromone,

(b) a toxicant substance for pest arthropods, and

(c) a bait component for pest arthropods.

2. The composition of Claim 1 wherein said pest arthropod attractant pheromone is a social insect attractant pheromone, said toxicant is a toxicant effective for social insects, and said bait is a bait suitable for social insects.

3. The composition of Claim 1 wherein said social insect attractant pheromone is a queen pheromone, preferably a synthetically made queen pheromone.

4. The composition of Claim 1 wherein said attractant pheromone is an ant attractant pheromone, preferably a fire ant attractant pheromone, said toxicant is a toxicant effective for ants, preferably fire ants, and said bait is a bait suitable for ants, preferably fire ants.

5. The composition of Claim 4 wherein said fire ant attractant pheromone is a queen attractant pheromone, preferably queen attractant pheromone derived from fire ant poison sac.

6. The composition of Claim 4 wherein said queen attractant pheromone is synthetically made.

7. The composition of Claim 6 wherein said queen attractant pheromone comprises (E) -6- (1-pentenyl) -2H-pyran-2-one and racemic invictolide.

8. The composition of Claim 6 wherein said queen attractant pheromone is racemic invictolide.

9. A method for controlling pest arthropods comprising treating said arthropods with a composition comprising

(a) an effective amount of a pest arthropod attractant pheromone,

(b) a toxicant substance for pest arthropods, and

(c) a bait component for pest arthropods.

10. The method of Claim 9 wherein said pest arthropod attractant pheromone is a social insect attractant pheromone, said toxicant is a toxicant effective for social insects, and said bait is a bait suitable for social insects.

11. The method of Claim 9 wherein said social insect attractant pheromone is a queen pheromone.

12. The method of Claim 9 wherein said attractant pheromone is an ant attractant pheromone, preferably a fire ant attractant pheromone, said toxicant is a toxicant effective for ants, preferably for fire ants, and said bait is a bait suitable for ants, preferably fire ants.

13. The method of Claim 9 wherein said fire ant attractant pheromone is a queen attractant pheromone, preferably derived from fire ant poison sac.

14. The method of Claim 12 wherein said queen attractant pheromone is synthetically made.

15. The method of Claim 14 wherein said queen attractant pheromone comprises (E) -6- (1-pentenyl) -2H-pyran-2-one and racemic invictolide.

16. The method of Claim 14 wherein said queen attractant pheromone is racemic invictolide.