Chapter Two: Pesticide Interactions

As an oversimplification, it may be stated that all pesticides work by interrupting some vital life system in the pest. This same action may also impact a non-target organism such as a honey bee, a bird or the pesticide applicator. Although different pesticides may have different modes of action, the combined effects of seemingly unrelated pesticides can sometimes give surprising results. Because the numerical reality is that no one person, group or research effort could examine all the possible combinations of pesticides on any one organism, this study chose to look at the effect of three common pesticides on bees simultaneously exposed to synthetic pyrethroid insecticides at a sublethal dose.

Background and Objectives

Because honey bee colonies are often exposed to several different pesticides during the year, the question of the interaction of synthetic pyrethroid insecticides and other pesticides becomes important in assessing the total impact on bees. Some of these other pesticides may be other synthetic pyrethroid insecticides, some may be other types of insecticides and some may be fungicides or herbicides which by themselves may not be particularly harmful to honey bees (Moffett 1972, Morton 1974, Stevenson 1978, Stoner 1985). Synergism of the toxicity of insecticides to dipterous pests by herbicides was documented by Lichtenstein (1973) but was not seen by Sonnet (1978) with carbamate and organophosphate insecticides fed in combination with various herbicides to honey bees.

In general, honeybees keep nectar from various sources largely segregated in the combs of the hive. That is, within a given cell which makes up the comb, usually only honey from a single floral source is stored. Further, usually the cells of a given area of the hive are filled at the same time and will represent those honeys made from nectar collected in a rather narrow time range. Most of the honey consumed by the colony over the winter is stored in an area of the hive referred to as the brood nest. This is the area of the hive in which brood is reared through the spring and summer.

In the fall as brood production declines, the bees begin to fill cells no longer needed for brood production with the honey that is being made at the time. This generally leads to honey from various sources being stored in concentric circular areas in the brood nest. Later in the season, the bees will begin to move honey stored in other parts of the hive into the brood nest in preparation for winter. These honeys come from a variety of nectar sources, and may be contaminated with a variety of pesticides. The possibility of pesticide interactions in the honey bee colony is increased by their unique method of storing and utilizing food. In the winter, the cluster may be exposed to any of these products simultaneously.

The importance of interactions between unrelated compounds is well documented. The natural pyrethrins, from which the synthetic pyrethroid insecticides take their name, are isolated from the flowers of certain chrysanthemum plants. They are known for their very quick knockdown of many pest species. They are often formulated in combination with piperonyl butoxide, a compound which by itself has little toxicity to insects. Together, the piperonyl butoxide acts as a synergist increasing the toxicity of the pyrethrins by a factor greater than the additive toxicities would suggest.

The behavior of bees as a unit is highly dependent on chemical cues being passed from individual to individual in the colony. Sharing food among worker bees as they pass in the colony facilitates this form of communication. Because of this food exchange behavior and the pattern of food storage in the brood nest, it is likely that any individual worker may be consuming honey from a number of sources on any one day. In order to simulate this, a study of the effect of exposure to synthetic pyrethroid insecticides simultaneously with one of three types of pesticides was devised.

The fungicide chosen was mancozeb, the active ingredient in the product Dithane M-45, among others. This product is frequently used in vegetable and fruit production (Sine 1988). Bees are attracted to blooming orchards in the spring at a time when they are replenishing honey stores depleted by the winter. Honey produced at this time may be stored for long periods. Because stored honey must be diluted with water prior to consumption by adult bees or for feeding larvae, nectar is preferred as a food over stored honey. If subsequent conditions are good, honey reserves are not needed to sustain the colony through poor nectar flows later in the season. Surpluses that are produced in good nectar years in the spring and stored in empty cells in the brood nest are likely to remain there until winter.

The herbicide chosen was paraquat, the active ingredient in the product Gramoxone 1.5EC. This product was chosen because it is a nonselective contact herbicide used to kill growing vegetation and because it has a higher toxicity to animals than most herbicides (Sine 1988). The areas in which it is applied are more likely to contain vegetation in bloom which is attractive to bees than preplant, preemergence or early postemergence herbicides. It is used throughout the growing season and has seen increased use as the practice of no-till, or reduced tillage farming has become popular in Indiana. In this practice weeds are killed with a herbicide, such as paraquat, in preparation for planting into ground that is not tilled. This may occur early in the season in the case of full season planting or later in the year in preparation for a second or double crop on land from which one crop has already been harvested. In either case, the possibility of some weeds being in bloom and attractive to the bees is high.

The insecticide chosen was Sevin 50W, a formulation of carbaryl, which is a common, general purpose insecticide that is especially toxic to bees. This product is used in nearly all crops, but is of special interest in some parts of Indiana because of the frequency with which it is used in soybeans for control of leaf feeding insects. Soybeans are often treated while in bloom and are often attractive to honeybees as nectar producing plants (Erickson 1979, Kettle 1979, Robacker 1983). Carbaryl is also used in other attractive crops such as sweetcorn, melons, cucumbers and in home gardens and orchards.

Because of the number of pesticide combinations to be examined, only two synthetic pyrethroid insecticides, permethrin and fluvalinate, were used. These products represented the most and least toxic of the insecticides examined in the previous study. The lower temperature range of 12 degrees C was also dropped because of the results of the previous work and to reduce the number of treatments. All of the pesticides were examined at two concentrations, 1 and 10 PPM.

The objective of this study was to determine if there was any synergism or antagonism between the products. This was determined by examining the mortality of bees fed a synthetic pyrethroid insecticide in combination with another pesticide in comparison to the mortality of the two products alone.

Materials and Methods

Bees were collected from the same colony as used in the previous study. The study was conducted one year later, but was headed by the same queen as evidenced by markings placed on her thorax at the time of her introduction to the colony. Bees were collected and handled as previously described. Twenty five bees were again used in cups constructed as before. Each treatment was tested at 18 degrees C and 25 degrees C.

Table 3

Table 3

The experimental design was again a split plot randomized complete block design with the restriction on randomization being the two temperature ranges. Treatments (Table 3) consisted of the assigned pesticides mixed in 50% sucrose solution and prepared to give concentrations of 1 or 10 PPM as assigned when mixed with equal volumes of the paired pesticide. All solutions were prepared fresh before the trial with dilutions of 1000 PPM solutions. The study was replicated four times, with each temperature repeated twice in each of two environmental chambers.

Formulated product was used for all solutions assuming the concentration stated on the packaging to be correct. Fresh formulated product was obtained from the manufacturer before the start of the experiment. Tests for possible fumigant action of the products in 50% sucrose indicated no effect from a concentration of 1000 PPM. Concentrations of 1 PPM and 10 PPM were tested for stability in 50% sucrose solution for seven days and had no significant decrease in toxicity.

Bees were checked for mortality at 12 and 24 hours following introduction to the cups and each 24 hours thereafter for 5 days. As previously described, bees not responding to gentle stimuli were considered dead. Two environmental chambers were used for the trial and each chamber was kept at 18 degrees C for two replications and 25 degrees C for the other two replications. The bees were kept in the dark throughout the experiment and at 50% to 70% relative humidity.

Results and Discussion

No evidence of synergism or antagonism was shown in any of the combinations examined. The toxicities of the fungicide and herbicide were not found to be significantly different from the control of 50% sucrose under the conditions examined. Bee mortality for the treatments containing permethrin showed higher mortality at 18 degrees C than at 25 degrees C, as expected from the previous tests.

Data Handling.

The ANOVA and GLM procedures of SAS were used for data analysis. As in the previous study, mortality is indicated by a mean of the percentage of bees dead at each of the six observations made over five days. Means of three or more groups were tested for significance of differences with the Duncan’s Multiple Range Test with Type I error rate set at 5%.

Overall Results.

Permethrin and carbaryl were the most toxic (Figure 5) and were not significantly different from each other at the 5% level under the conditions tested. There were also no significant differences between fluvalinate, paraquat, mancozeb or the control of 50% sucrose. A control was grouped with both the synthetic pyrethroid insecticides and the three other pesticides and assigned to both concentrations for completeness of balance in design.

The mean percentage of bees dead per observation for each pesticide tested.

The mean percentage of bees dead per observation for each pesticide tested.


As expected from the previous work, mortality was higher at 18 degrees C than 25 degrees C for all treatments containing permethrin. Figure 6 shows the mean percentage of dead bees per observation for all the treatments by pesticide and temperature. That is, all observations of bees exposed to permethrin, in any combination, at 25 degrees C are represented by the first bar of the graph.

Percentage of bees dead per observation for each pesticide by temperature.

Percentage of bees dead per observation for each pesticide by temperature.


Because the value for each pesticide is a mean of all occurrences of that product in combination with all others in a balanced design, much of the increased mortality at lower temperature is due to the combination of that product with premethrin. The mean percentage of dead bees per observation for each of the 48 treatments is given in Figure 7.

The mean percentage of bees dead per observation by treatment.

The mean percentage of bees dead per observation by treatment.


Pesticide Interactions.

Comparing the expected mortality of permethrin or fluvalinate with the other pesticides did not indicate that any synergism or interaction was present. The expected mortality was determined by adding the increased mortality over the control for the non-pyrethroid to the increased mortality of the pyrethroid over the control. This calculated, expected value was compared to the observed mortality of the combination minus the control mortality estimated by the control/control mean.

For example, the permethrin/control mean mortality was 43.13%, a 32.17% increase attributable to permethrin over the control/control combination value of 10.96%. The carbaryl/control combination gave a 41.65% mean mortality indicating that 41.65% – 10.96%, or 30.69% was due to carbaryl.

Therefore, the expected mean percent mortality of the permethrin/carbaryl combination was 32.17% + 30.69% or 62.86%. The observed mean percentage of dead bees per observation for the permethrin/carbaryl combination was 70.63%. This value minus the control mortality (70.63% – 10.96%) was 59.67% is the observed combination mortality.


The results of this study indicate that two of the more toxic insecticides to bees, permethrin and carbaryl, have an additive and not a multiplicative toxicity to adult bees. It was also shown that any effects of paraquat and mancozeb are very small in terms of adult mortality and these products do not significantly change the impact of permethrin or fluvalinate on bees. This is especially interesting in light of the perception of many beekeepers and applicators about the impact of herbicides and fungicides on honeybees (see Chapter 4).