Phytosanitary treatments are used to disinfest agricultural commodities of possible quarantine pests so that the commodities may be shipped from infested areas to areas where the pests do not exist but could become established (Heather & Hallman 2008). The major treatments used worldwide are cold storage, heating in air or water, fumigation, and ionizing radiation. Vapor heat treatments use heated air at about 47 °C with high humidity to kill insects that may be infesting fruit. Vapor heat treatment also is known by other names, such as high temperature-forced air treatment. Vapor heat treatment schedules dictate at minimum the treatment air temperature and the minimum temperature to be reached in the center of the fruit or at the seed surface for fruits with large seeds, such as mango.

For phytosanitary treatment schedules to be broadly applicable geographically there should not be significant differences in efficacy among populations of the same pest species from different areas. The phytosanitary treatment literature and treatment schedules indicate possible differences in efficacy among different populations of the same species of tephritids. For example, in a comprehensive review of the literature, Dohino et al. (2017) note that vapor heat treatment schedules for mangoes for Bactrocera dorsalis (Hendel) (Diptera: Tephritidae) differ in severity from a low of seed surface temperature ≥ 46 °C for 10 min in the Philippines to a high of seed surface temperature ≥ 48 °C for 20 min in India. Reasons why these schedules are different might include differences in susceptibility to heat among pest populations, the large-scale confirmatory testing was done at different time or temperature combinations, different manners of measuring efficacy were used, results were interpreted differently, and importing plant protection organizations required different levels of efficacy or simply added time or temperature to the treatment for additional security. The possibility that differences might exist may obstruct the evaluation and approval of phytosanitary treatments by international treaties dealing with trade, such as the International Plant Protection Convention (IPPC).

The objective of this study was to determine if populations of B. dorsalis from different geographic areas vary significantly in tolerance to the vapor heat treatment. A secondary objective was to estimate final seed surface temperatures required for efficacy for B. dorsalis in mango. Bactrocera dorsalis was chosen for this research because vapor heat treatments for this insect are being evaluated by the IPPC and the question of possible variation in tolerance to heat among different populations was raised by countries reviewing the recommendations.

A review of the literature indicates that the mid-egg stage is generally the most thermo-tolerant when studies are done with fruit flies infesting fruit and a common measure of efficacy of survival to the 3rd instar or pupariation is used. Only studies conducted with insects feeding their entire larval lives in fruit were considered, and not studies using insects reared in artificial diet and then inserted into fruit or done completely in vitro, to avoid making untested assumptions about the effect of artificial infesting techniques on relative efficacy. Hallman (2014) observed that 3rd instars of Mexican fruit fly, Anastrepha ludens (Loew) (Diptera: Tephritidae), reared on diet and inserted into grapefruit seemed more susceptible to hot water immersion treatment than those reared naturally in grapefruit via oviposition, although the results were not statistically significant in that case.

Data supporting the conclusion that the mid-egg stage is the most heat tolerant are from 8 species of tephritids including B. dorsalis (Table 1). In 10 of 13 studies (77%) the egg was the most thermo-tolerant stage or of equal tolerance as other stages that were among the most tolerant in that study. In the 3 studies where the egg was not the most tolerant stage it was the next most tolerant after 1st or 3rd instars.

Materials and Methods

VAPOR HEAT TREATMENT

In each replicate, 6 of the 12 mangoes of the same cultivar infested with 24-h-old eggs of each population (18 mangoes total) were placed in an environmental chamber (Pol-Eko Aparatura, Model KK 700 TOP+, Warsaw, Poland, about 1 m³ volume). This machine provides the basic features for a vapor heat treatment, namely a sustained source of heat with high humidity. The 18 mangoes infested with the 3 populations were randomly mixed together in the treatment chamber and heated at 47 °C and 95% RH for 150 to 180 min. The other 6 mangoes from each population were used as a control treatment. These were not treated and were held at 25 °C to allow larvae to develop to the size where they could be counted (3rd instar) and used to estimate egg population sizes in the treated mangoes.

Table 1.

Summary of thermo-tolerance studies among stages of Tephritidae using common measures of efficacy among the stages (egg and all 3 instars) and reared and treated within fruit.

Temperatures at the seed surface in 3 mangoes (1 from each population) were recorded with thermocouples (HOBO data logger, Onset Computer Corporation, Bourne, Massachusetts, USA) calibrated at 46.0 °C with a certified thermometer (H-B Instrument-SP Scienceware, Trappe, Pennsylvania, USA) traceable to the US National Institute of Standards and Technology.

The goal was to vapor heat treat the infested mangoes long enough to kill almost all of the Bactrocera dorsalis eggs, but leaving a few survivors so that possible differences in survival would be measurable and allow for comparisons among populations. The decision on when to terminate a treatment was made taking into account results from previous replicates; i.e., if no insects survived a treatment, the subsequent treatment target temperature time was reduced, and if many survived it was increased. All of the mangoes within a replicate were removed at the same time.

Results

VAPOR HEAT TREATMENT

In 1 replicate with the Kenyan population, no larvae were found in the controls, indicating lack of infestation by that population for that replicate (Table 2). In all other combinations the infestation rate ranged from 3.8 to 194.5 larvae per fruit in the non-treated control group. Infestation values do not necessarily reflect the population of eggs laid, only the population of larvae surviving to a size where they could be found. It is assumed that differences in the number of eggs deposited in mangoes and the number of larvae found later are relatively analogous among populations, enabling comparisons among populations.

Table 2.

Results of vapor heat treatment (air temperature 47 °C, RH 95%) of mangoes infested with 24-h-old eggs of Bactrocera dorsalis from 3 locations.

No survivors were found when the mean mango seed surface temperature reached ≥ 45.01 ± 0.56 °C in the Chinese population and ≥ 45.38 ± 0.29 °C in the other 2 populations (Table 2). Mean infestation levels at these points when no survivors were found were 28.5, 22.7, and 26.3 larvae per fruit for China, Kenya, and Thailand, respectively.

In 4 of 11 replicates the vapor heat treatment resulted in no survivors for any of the populations, leaving 7 replicates for analysis of differences in efficacy among the 3 populations (Table 2). ANOVA of the 7 replicates and 3 populations found that they were not significantly different at the 95% level of confidence (F = 3.22; df = 2,12; P = 0.076). The arcsine transformed means ± SEM for the Thai, Chinese, and Kenyan populations were 7.45 ± 4.00, 1.70 ± 1.28, and 0.79 ± 0.65%, respectively.

Estimates of final seed surface temperatures for 95, 99, and 99.9968% mortality for the B. dorsalis populations from Kenya and Thailand at 47 °C are presented in Table 3. The software did not provide estimates for the population from China due to lack of convergence in the model.

Estimates were obtained for all 3 populations combined and the data for the populations from Kenya and Thailand combined. The mean seed surface temperature to achieve 99.9968% control was estimated to be 47.1 °C when all data were pooled.

Discussion

Decisions on phytosanitary measures by plant protection organizations are sometimes made in situations that will not be entirely defined and would require an inordinate amount of time and resources to clarify beyond a reasonable doubt. Trade and international relations may influence the need to arrive at a decision. The consequences for concluding that there are significant differences among populations of the same pest species for tolerance to a phytosanitary treatment, when actually there are not, may result in treatments being regionspecific and create unnecessary obstacles to trade. The consequences of concluding that there is no significant difference, when actually there are, may result in failed treatments and new pest infestations if the treatment was developed with a particularly susceptible population and subsequently applied commercially to commodities infested with more tolerant populations. These potential consequences should be considered when treatments are approved by plant protection organizations.

Table 3.

Results of probit analysis of the dose-response relationship between final seed surface temperature in vapor heat treatment of mangoes infested with 24-h-old eggs of 3 populations of Bactrocera dorsalis.

Although differences in survival after vapor heat treatment among the 3 B. dorsalis populations studied were not significant at the 95% confidence level, they would be significantly different at the 92% level of confidence. The population from Thailand appeared to be more tolerant than the populations from Kenya and China. However, the apparently most susceptible population (China) still required a final seed surface temperature of 45.0 ± 0.56 °C to kill 100% of an estimated minimum of 171 eggs, which was only a mean of about 0.4 °C lower and within the standard error limits of the other 2 populations (Table 2). Vapor heat treatment of B. dorsalis infested mangoes in the Philippines found a similar result; at a temperature between 45 to 46 °C all eggs were killed (Merino et al. 1985).

Relatively large differences in survival among the 3 populations at < 45 °C (Table 2) may be largely due to the increased error in estimating populations from non-treated fruit as the proportion surviving increases. As the proportion surviving decreases these differences are minimized.

Considerable overlap in confidence intervals among the probit mortality estimates (Table 3) suggests there are no significant differences in susceptibility to vapor heat treatment among these 4 groups. Final seed surface temperature estimates to achieve the very high level of control demanded of phytosanitary treatments may give impractical results. For example, the mean final seed surface temperature for 99.9968% control when data from all 3 populations are combined (47.1 °C) exceeds the treatment air temperature (47.0 °C). Therefore, to confirm a phytosanitary treatment, normally large-scale testing is done where many thousands of insects are treated at a specific set of treatment parameters to demonstrate that a treatment protocol results in an acceptable level of efficacy (Heather & Hallman 2008).

Because the number of generations of each population in culture at the IPCL was 45 to 60, it could be argued that they may not reflect the heat tolerance of the original field populations from which they were collected. However, Corcoran (2001) found that Bactrocera spp. reared for 120 to 220 generations in the laboratory were not reduced in heat tolerance compared with wild flies.

This research has aided the approval process of vapor heat treatments being considered by the IPPC resulting in treatment approval without the need for considering country of origin. We feel this decision is justified by the results of this research.