**3. Results**

#### *3.1. Effects of Sublethal Prallethrin Doses on Mosquito Fitness*

In the first experiment, the following were evaluated: (1) the effects of species, sex, and treatment on KD during the 90-min treatment trial; (2) the percentage of dead and affected mosquitoes 48 h into the post-treatment period; (3) the effects of species, sex, and treatment on long-term mortality (i.e., over the 4-week post-treatment period); and (4) the effects of species, sex, and treatment on fertility, egg laying, and F1 population size.

3.1.1. Effects of Species, Sex, and Treatment on KD during the 90-Min Treatment Trial

All three sublethal doses of prallethrin (0.4, 0.8, and 1.6 mg/h) caused more than 95% of mosquitoes to be knocked out, except in the case of *Cx. pipiens* females (87.2%; Figure 2). The higher the dose, the faster the KD. KD differed between the two control groups and the three prallethrin groups based on species and sex (Figure 2). In the untreated control, there was no KD. In the negative control, only a few male *Ae. albopictus* were knocked down (12.8%; Figure 2b).

**Figure 2.** Knockdown over the 90-min treatment trial in Experiment 1 for female and male *Ae. albopictus* and *Cx. pipiens* across the five treatment groups: (**a**) Female *Ae. albopictus*, (**b**) male *Ae. albopictus*, (**c**) female *Cx. pipiens*; and (**d**) male *Cx. pipiens*.

First, KD was compared within species. In *Ae. albopictus*, for both sexes, there was a significant difference in KD between the mosquitoes exposed to the 0.4 mg/h prallethrin dose and the mosquitoes exposed to the 0.8 and 1.6 mg/h prallethrin doses (Table 3). Exclusively in the case of male *Ae. albopictus*, there was no significant difference between the groups exposed to the 0.8 vs. the 1.6 mg/h prallethrin dose. In general, KD was faster

at the higher doses (Figure 3a,b). In *Cx. pipiens*, there were significant differences among all three prallethrin doses for both sexes (Table 3).

**Table 3.** Comparison of within species knockdown for female and male *Ae. albopictus* and *Cx. pipiens* across the five treatment groups in Experiment 1.


Pairwise comparisons of knockdown (KD) were carried out using Mantel–Cox log-rank tests in implemented in in SPSS (v. 15.0.1) for Windows (Chicago, SPSS Inc). All the statistical comparisons used an alpha level of 0.05. 1 No statistics were performed because no mosquitoes were knocked down in the controls. 2 Each control group (untreated and negative) was compared with each prallethrin group (0.4, 0.8, and 1.6 mg/h). This row summarises the results. Significant differences were observed between the control groups and the prallethrin groups in all the configurations.

> Second, KD was compared between species. At the lowest dose (0.4 mg/h), differences only existed between male *Ae. albopictus* and female *Cx. pipiens* (χ<sup>2</sup> = 6.562, *p* < 0.05). At the intermediate dose (0.8 mg/h), male *Ae. albopictus* experienced significantly faster KD than all the other groups (*p* < 0.0001 for all the comparisons). At the highest dose (1.6 mg/h), there were no differences among female *Ae. albopictus*, male *Ae. albopictus*, and male *Cx. pipiens* (female *Ae. albopictus* vs. male *Ae. albopictus*: χ2 = 0.787, *p* = 0.375; female *Ae. albopictus* vs. male *Cx. pipiens*: χ2 = 3.645, *p* = 0.056; male *A. albopictus* vs. male *Cx. pipiens*: χ2 = 1.419, *p* = 0.234). However, female *Cx. pipiens* experienced significatively slower KD than all the other groups (*p* < 0.0001 for all the comparisons). For example, at 10 min, KD was only 23% for female *Cx. pipiens* but 87–92% for all the other groups (Figure 3).

#### 3.1.2. Percentage of Dead and Affected Mosquitoes 48 h into the Post-Treatment Period

Mosquitoes displayed a variety of fates during the 48 h that followed the trials. Some died, some survived, and ye<sup>t</sup> others remained alive but were clearly affected by the prallethrin. The most obvious sign that surviving mosquitoes had been affected was the partial or complete loss of legs (Figure 3). This effect was observed for all the doses tested, although it was more pronounced at the higher doses (e.g., some individuals lost one or more legs and also died).

**Figure 3.** Photograph showing a sample of female *Cx. pipiens* that lost legs following prallethrin exposure. The numbers next to the mosquitoes indicate the number of legs lost.

At 24 h into the post-treatment period, dead and affected mosquitoes together accounted for more than 90% of all the mosquitoes in almost all the prallethrin groups. The only exception was female *Cx. pipiens* exposed to the 0.4 mg/h prallethrin dose (41.60% at 24 h and 75.2% at 48 h).

Similarly, at 48 h into the post-treatment period, dead and affected mosquitoes together accounted for more than 90% of all the mosquitoes (females and males combined) in almost all the prallethrin groups. The only exception was *Cx. pipiens* exposed to the 0.4 mg/h prallethrin dose (84.4%).

*Dead Adult Mosquitoes*. At 24 h into the post-treatment period (Figure 4), male mortality in both species exceeded 90% in almost all the groups exposed to prallethrin. The exception was male *Cx. pipiens* exposed to the 0.4 mg/h prallethrin dose, a group that displayed 80% mortality. In both species, female mortality was lower, especially when mosquitoes were exposed to the 0.4 mg/h prallethrin dose (49.6% and 30.4% for *Ae. albopictus* and *Cx. pipiens*, respectively). At the prallethrin dose of 0.8 mg/h, female mortality was 56% for *Ae. albopictus* and 43.2% for *Cx. pipiens*. At the prallethrin dose of 1.6 mg/h, female mortality was 71.2% for both species.

At 48 h into the post-treatment period (Figure 4), the only increases in male *Cx. pipiens* mortality were seen in the groups exposed to the 0.4 and 0.8 mg/h prallethrin doses (from 80% to 84% and from 95.2% to 96.8%, respectively). Female mortality rates had risen accordingly with higher doses for both species of mosquitoes from 67.7% to 83.2% for *Ae. albopictus* and from 49.6% to 86.4% for *Cx. pipiens* (Figure 4).

**Figure 4.** Percentages of affected and dead *Ae. albopictus and Cx. pipiens* at 24 and 48 h into the post-treatment period across the five treatment groups.

*Affected Adult Mosquitoes*. At 24 h into the post-treatment period, 5% at most (range: 0.8–4.8%) of male *Ae. albopictus* were affected; the rest of the mosquitoes were dead. In the case of female *Ae. albopictus*, there were 42.4% and 40.0% affected mosquitoes in the groups exposed to the 0.4 and 0.8 mg/h prallethrin doses, respectively. At 48 h, these percentages dropped to 26.4% and 25.6%, respectively, largely because the affected mosquitoes had died. For the group exposed to the 1.6 mg/h prallethrin dose, the percentage of affected mosquitoes went from 21.6% at 24 h to 13.6% at 48 h. The same general patterns were seen in *Cx. pipiens*.

At 48 h, the percentages of affected mosquitoes were lower because mortality had occurred. For male *Cx. pipiens*, the group exposed to the 0.4 mg/h prallethrin dose had the highest percentage of affected mosquitos (13.6% at 24 h and 9.6% at 48 h). In contrast, for female *Cx. pipiens*, the percentage of affected mosquitoes increased from 11.2% at 24 h to 25.6% at 48 h for the group exposed to the 0.4 mg/h prallethrin dose; for the groups at prallethrin doses of 0.8 and 1.6 mg/h, these percentages decreased from 48.8% to 32.8% and from 26.4% to 12%, respectively.

Mortality never climbed above 15% in the untreated and negative controls, except in the case of male *Ae. albopictus* (31.2% and 32%, respectively). None of the mosquitoes in the controls showed signs of having been affected (Figure 4).

3.1.3. Effects of Species, Sex, and Treatment on Long-Term Mortality

One week into the post-treatment period, total mortality for female and male *Ae. albopictus* was 90% across all the prallethrin groups; in the controls, however, total mortality was only 28%. For female and male *Cx. pipiens*, the total mortality for mosquitoes exposed to prallethrin doses of 0.4, 0.8, and 1.6 mg/h was 82%, 89.6%, and 94.8%, respectively; for the controls, it was 20.8%.

For both species and sexes, LTM was significantly higher in all the prallethrin groups than in the control groups (Table 4). Within species and sex, LTM did not differ between the untreated and negative controls; it was highest for male *Ae. albopictus* and lowest for female *Ae. albopictus* (Figure 5).


**Table 4.** Treatment effects on long-term mortality for female and male *Ae. albopictus* and *Cx. pipiens* across the five treatment groups.

1 Each control group (untreated and negative) was compared with each prallethrin group (0.4, 0.8, and 1.6 mg/h). This row summarises the results. Significant differences were observed between the control groups and the prallethrin groups in all the configurations. Pairwise comparisons of long-term mortality (LTM) were carried out using Mantel–Cox log-rank tests implemented in SPSS (v. 15.0.1) for Windows (SPSS Inc., Chicago, IL, USA). All the statistical comparisons used an alpha level of 0.05.

**Figure 5.** Mosquito mortality during the 4-week post-treatment period across the five treatment groups: (**a**) Female *Ae. albopictus*, (**b**) male *Ae. albopictus*, (**c**) female *Cx. pipiens*, and (**d**) male *Cx. pipiens*. Mortality at 24 h and 48 h is also shown to clarify the relationship between STM and LTM. LTM, long-term mortality; STM, short-term mortality.

LTM did not differ between the groups exposed to the 0.4 and 0.8 mg/h prallethrin doses, regardless of species or sex. It did, however, differ between the groups exposed to the 0.4 and 1.6 mg/h prallethrin doses. It was higher at the latter dose, except in the case of male *Ae. albopictus*—they died equally rapidly across all three doses (100% mortality at 2 weeks post-treatment; Figure 5 and Table 4). In both species, male but not female LTM was significantly higher in the groups exposed to the 1.6 mg/h prallethrin dose than in the groups exposed to the 0.8 mg/h prallethrin dose (Figure 5 and Table 4).

Sex also affected mortality in the prallethrin groups: LTM was higher for males than females, regardless of species (Figure 5 and Table 4). At 2 weeks post-treatment, male mortality was higher than female mortality by 13–20% for the groups exposed to the 0.4 and 0.8 mg/h prallethrin doses and by 7–10% for the groups exposed to the 1.6 mg/h prallethrin dose.

Species-specific differences in male mortality were present at the lowest prallethrin dose: at 1 week post-treatment, male *Ae. albopictus* exhibited 99.2% mortality, while male *Cx. pipiens* exhibited 90.4% mortality (0.4 mg/h: *p* < 0.0001). There was no such difference for the intermediate prallethrin dose (0.8 mg/h: χ2 = 0.011, *p* = 0.918) or the highest prallethrin dose (1.6 mg/h: χ2 = 3.806, *p* = 0.051). Species did not affect female mortality at any of the doses (0.4 mg/h: χ2 = 0.826, *p* = 0.363; 0.8 mg/h: χ2 = 0.256, *p* = 0.613; 1.6 mg/h: χ2 = 0.740, *p* = 0.390).

3.1.4. Effects of Species, Sex, and Treatment on Fertility, Egg Laying, and F1 Population Size over the 4-Week Post-Treatment Period

*Culex pipiens*. In this part of the experiment, the methodology diverged slightly for the two species because the *Cx. pipiens* strain did not need to consume blood (see the Methods section).

The number of eggs laid by *Cx. pipiens* could not be accurately counted because the eggs formed rafts. Furthermore, some of the rafts were not well assembled. Instead of forming the expected boat-like shape [37], unassembled eggs could be seen on the water surface (Figure 6).

**Figure 6.** Egg rafts produced by *Cx. pipiens* in the (**a**) untreated control group and (**b**) the group exposed to the 0.8 mg/h prallethrin dose. In (**b**), the poorly assembled egg rafts have been circled to make them easier to identify.

Forty-eight hours after the mosquitoes had been given access to water to lay their eggs, the number of females found dead in the tray was much greater in the prallethrin groups than in the control groups (Fisher's exact tests with Bonferroni correction: *p* < 0.001 for all the comparisons between the control groups (untreated or negative) and each of the prallethrin groups). In the control groups, fewer than 10% of females were found dead, while 23.81%, 38.78%, and 41.18% of females were found dead in the groups exposed to the 0.4, 0.8, and 1.6 mg/h prallethrin doses, respectively (Table 5).


**Table 5.** Treatment effects on mosquito fitness and F1 population size in *Cx. pipiens*.

ND, no data. In the negative control, algae began growing in some of the trays, creating a surface layer that choked off a large percentage of the larvae. This portion of the experiment thus had to be stopped for this group.1 This metric was calculated for the prallethrin groups based on the total number of adults in the F1 population in the untreated control.

The numbers of larvae to reach the third/fourth instar stage were similar in the untreated control (4137) and in the negative control (3985). Compared with the untreated control, the percentages of reduction in larvae that reached this development stage were 37.27%, 50.06%, and 84.60% for the groups exposed to the 0.4, 0.8, and 1.6 mg/h prallethrin doses, respectively. It is important to note that this result appeared to stem from a smaller number of adults being available to reproduce. When examining the ratio of third/fourth instar larvae to available females, there were no differences among treatments (Table 5).

The percentage of larvae reaching adulthood varied somewhat (64–74% across both sexes), although no treatment effects were observed (Fisher's exact tests with Bonferroni correction: *p* > 0.05 for all the comparisons between treatments). The sex ratio was nearly 1:1 in the untreated control and in the group exposed to the 0.4 mg/h prallethrin dose. The sex ratio was male-biased in the groups exposed to the 0.8 mg/h and 1.6 mg/h prallethrin doses.

There was a pronounced effect of treatment on the F1 population size. Using the untreated control as the standard of comparison, exposure to the 0.4, 0.8, and 1.6 mg/h prallethrin doses reduced the F1 population sizes by 31.25%, 53.13%, and 84.13%, respectively. Declines in population size were significatively different among the three prallethrin groups (Fisher's exact tests with Bonferroni correction: *p* < 0.005 for all the comparisons).

*Aedes albopictus*. The same data were collected for *Ae. albopictus*, but, in addition, egg number was quantified. As the eggs were laid on wet filter paper, females were not at risk of drowning. In all the groups, including controls, the percentage of females found dead in the egg-laying trays was less than 1%, except for the group exposed to the 0.8 mg/h prallethrin dose (5.41%) (Table 6).


1 This metric was calculated for the prallethrin groups based on the total number of adults in the F1 population in the untreated control.

When examining the ratio of third/fourth instar larvae to available females, no consistent pattern was seen. While there were 15.59 larvae for each female in the group exposed

to the 0.4 mg/h prallethrin dose, this figure was 3.86 and 8.24 in the groups exposed to the 0.8 and 1.6 mg/h prallethrin doses, respectively. A difference was also observed between the controls (untreated control: 13.20 larvae to 1 female; negative control: 9.44 larvae to 1 female; Table 6).

The percentage of larvae reaching adulthood (75–99%) displayed no treatment effects (*p* > 0.05), except the group exposed to the 0.8 mg/h prallethrin dose that differed from the other two prallethrin groups (*p* < 0.00001). The sex ratio was biased towards females, ranged from 0.7 to 1.0, and was unaffected by the treatments.

There was again a pronounced effect of treatment on the F1 population size. Population size declined by 32.95%, 60.6%, 91.55%, and 89.94% in the negative control group and in the groups exposed to the 0.4, 0.8, and 1.6 mg/h prallethrin doses, respectively. Dose significantly affected declines in population size in almost all cases (Fisher's exact tests with Bonferroni correction: *p* < 0.00001 for all the comparisons except that between the groups exposed to the 0.8 versus the 1.6 mg/h dose (*p* > 0.05)) (Table 6).

#### *3.2. Effects of Sublethal Prallethrin Doses on Mosquito Biting Behaviour*

Percent protection after 5 min of exposure ranged from 80.07% (±28.38) at the 0.4 mg/h dose to 100% at the 1.6 mg/h dose, but this difference was not significant (*p* > 0.05); (Figure 7. The control treatments provided no protection. At this same time point, KD was null for the two controls; it was 9.33% (±5.39), 17.67% (±49.62), and 51.67% (±7.44) for the 0.4, 0.8, and 1.6 mg/h prallethrin doses, respectively. No significant differences were observed in KD between the groups exposed to the 0.4 versus the 0.8 mg/h dose (*p* > 0.05); there were significant differences in KD at 5 min for the groups exposed to the 0.4 versus the 1.6 mg/h dose and the 0.8 versus the 1.6 mg/h dose (*p* < 0.00001 in both cases). After the diffuser had been running for 15 min, 100% protection was seen in all the prallethrin groups (*p* > 0.05). KD remained null for the two controls; it was 80.17% (±10.25), 95.83% (±4.92), and 100.00% (±0.00) for the 0.4, 0.8, and 1.6 mg/h prallethrin doses, respectively (Figure 7). There was a significant difference between the groups exposed to the 0.4 versus the 1.6 mg/h dose (*p* < 0.05) but not between the groups exposed to the 0.4 versus the 0.8 mg/h dose (*p* > 0.05) or the groups exposed to the 0.8 versus the 1.6 mg/h dose (*p* > 0.05).

**Figure 7.** Percent protection (%*p*) and knockdown (%KD) over time for *Ae. albopictus* across the five treatment groups in Experiment 2.

When assessing percent protection, there were no differences between the untreated and negative controls at any of the time points (i.e., *p* > 0.05 at all time points). The same pattern was seen for KD (*p* > 0.05 at all time points).

When the relationship between KD and percent protection was examined, it was found that once KD reached 10%, protection never dropped below 90%. In the controls, negative percent protection values were observed because there were greater numbers

of landings during the treatment trial than during the pre-treatment trial. KD was not observed in the control groups (Figure 8).

**Figure 8.** Relationship between knockdown and percent protection for *Ae. albopictus* across the five treatment groups in Experiment 2.

#### *3.3. Assessments of Human and Environmental Health Risks*

The HHRA models found that if a prallethrin dose of 1.6 mg/h were to be used, adults could be exposed for 24 h per day, but children could only safely be exposed for 12 h per day. At a prallethrin dose of 0.8 mg/h, children could be exposed for a maximum of 20 h per day. At the lowest dose, 0.4 mg/h, both adults and children could be exposed for 24 h per day.

In the environmental risk assessment, PECs and PNECs were determined for different environmental compartments. When the PEC/PNEC ratio is greater than 1, the AS poses a risk. If prallethrin were to be used 24 h per day and released using two diffusers per household, it would not be safe to use a dose of 1.6 mg/h (PEC/PNEC ratio for soils: 1.34). However, lower doses—0.8 and 0.4 mg/h—would be safe under the same usage conditions (PEC/PNEC ratio for soils: 0.75 and 0.33, respectively).
