Development of abamectin resistance in Tetranychus urticae in Australian cotton and the establishment of discriminating doses for T. lambi


Since the 1980s Tetranychus urticae Koch has dominated Australian cotton due to its ability to develop resistance. Here we give screening data for a range of chemicals tested against T. urticae including abamectin, bifenthrin, diafenthiuron, etoxazole and propargite and speculate why abamectin resistance emerged without warning. Abamectin resistance was not detected in T. urticae in Australian cotton before season 2007–2008 when a few resistant individuals were detected in a single strain. Resistance was detected again in season 2010–2011 and continued to be detected in every subsequent season comprising 80% of strains tested in 2018–2019. We speculate the reason may relate to prophylactic abamectin use to prevent mite flare with Creontiades dilutes Stål mirid sprays. With the introduction of transgenic Bt-cotton, spraying significantly reduced and anecdotally Tetranychus lambi became more abundant. Although T. lambi may now be more common than T. urticae its response to chemical controls is completely unknown. Tetranychus lambi conspecific dose responses were established to support resistance monitoring against abamectin, diafenthiuron and propargite that generated discriminating dose (DD) estimates of 0.0007 g/L abamectin, 0.03 g/L diafenthiuron and 0.7 g/L propargite. These DD were used in season 2018–2019 but resistance was not detected against any product including abamectin. The reason why T. lambi may now dominate despite T. urticae being still resistant is speculated and thought related to the progressive reduction in insecticide use in Australian cotton and/or the changing weed complex in the transgenic cotton era.


Spider mites are an induced secondary pest in Australian cotton. Historically, the application of pesticides against (the primary pests) Helicoverpa armigera (Hübner) and Helicoverpa punctigera (Wallengren) reduced the abundance of beneficial species allowing spider mite populations to build rapidly in number (Wilson et al. 1998). During the 1970s outbreaks of the two-spotted spider mite, Tetranychus urticae Koch, and the bean spider mite, Tetranychus ludeni Zacker, were common in Australian cotton. Outbreaks were controlled with organophosphate insecticides, resulting in both species developing organophosphate resistance (Herron et al. 1998). Tetranychus urticae was significantly more resistant and competitively displaced T. ludeni which declined to almost undetectable levels in Australian cotton during the 1980s (Herron et al. 1998). A third tetranychid species, Tetranychus lambi Pritchard and Baker, was reported from Australian cotton in 1977 (Gutierrez and Schicha 1983) though it was probably present earlier and mistaken for T. urticae. Fields with significant populations of T. lambi were first reported in 1992 (Wilson 1992) when it was sporadically widely present (~ 90–100% plants infested) at a low density per plant (~ 2–3 adult female mites) in a small proportion of fields. During the 1990s T. lambi rarely developed significant populations. Further, because T. lambi caused much less obvious damage than T. urticae (e.g., see Maas and Redfern 2018, p. 36), and was therefore unlikely to reduce crop yields, control was not recommended. In contrast, T. urticae remained a significant problem in commercial cotton requiring control (Wilson et al. 1995) to prevent yield loss (Wilson 1993) due to reduced photosynthetic capacity (Reddall et al. 2004).

Tetranychus urticae in Australia has a long history of developing resistance to miticides and causing control failures. In Australian cotton, T. urticae developed resistance to dimethoate/omethoate in the 1970s (Herron et al. 1998) and to monocrotophos and profenofos in the 1980s (Herron et al. 1998). In the mid-1990s a mite resistance management strategy (MRMS) was developed to manage the development of resistance to miticides newly registered in Australian cotton, such as propargite and dicofol. As further new compounds were registered against spider mites (e.g., bifenthrin, chlorfenapyr, diafenthiuron, abamectin and etoxazole) they were progressively added to the MRMS. Despite this strategy, escalating resistance in H. armigera, meant compounds such as chlorfenapyr and bifenthrin, which targeted both Helicoverpa spp. and spider mites, received increasing use. Consequently, resistance in T. urticae to bifenthrin was reported in the 1990s (Herron et al. 2001) and chlorfenapyr in the early 2000s (Herron et al. 2004).

The use of insecticides/miticides and the species they target changed dramatically in the mid-1990s following the advent of transgenic Bt-cotton. Since then increasing uptake and efficacy of Bt-cotton, which now accounts for more than 90% of the Australian cotton area, resulted in a dramatic decline in insecticide use against Helicoverpa spp. (Wilson et al. 2018). However, reduced insecticide use against Helicoverpa spp. allowed other pests to become problematic and by the early 2000s, following the introduction of two-gene Bt-cotton, the green mirid, Creontiades dilutes Stål, emerged as a key pest (Wilson et al. 2018). More recently the silverleaf whitefly, Bemisia tabaci (Gennadius) MEAM1 and solenopsis mealybug, Phenacoccus solenopsis Tinsley, have also emerged as key pests in Bt-cotton (Wilson et al. 2018).

Other changes in the pest complex were also observed, including perceived shifts in the spider mite complex. Each year T. urticae populations were collected from commercial cotton crops and tested against registered miticides to monitor changes in resistance frequency. This allowed the effectiveness of MRMS to be evaluated and modified to preserve miticide efficacy. During the late 2000s, and especially in the 2010s, mite collection trips found only T. lambi in many cotton crops (Herron and Wilson 2016). At this stage we did not collect the T. lambi to evaluate miticide resistance as it was not regarded as a pest. However, more recently some consultants and agronomists identified areas where T. lambi had built to densities that caused leaf loss in the lower cotton canopy. This raised the possibility that future control of T. lambi may be required but worryingly, the response of T. lambi to available miticides was unknown. Overall, there is little biological information available for T. lambi (Helle and Sabelis 1985) and no resistance related data has been published.

To understand whether the mite complex in Australian cotton is changing while still being prepared to manage the existing pest species complex, we (1) tested T. urticae for resistance against currently used miticides in Australian cotton using a discriminating dose method; (2) collected all mite species from a range of cotton fields over two consecutive seasons to evaluate the frequency of fields with T. lambi, T. urticae or both; (3) completed full dose–response studies on many of those field-collected T. lambi, to miticides likely to be used for their control, to establish a baseline for potential future resistance monitoring; and (4) applied those new discriminating doses generated against T. lambi to test for resistance in a subsequent season.


Sample collection

Collections were made annually in February/March each year by the authors or crop pest management consultants (professional pest management practitioners employed by most Australian cotton growers) or farm agronomists from commercially managed cotton fields from seasons 2005–2006 to 2018–2019. Cotton crops sampled were spread across the St George (Queensland, Qld), Darling Downs (Qld), McIntyre (Qld and New South Wales, NSW), Mungindi (QLD and NSW), Gwydir (NSW), Upper and Lower Namoi (NSW), Macquarie (NSW), Hillston (NSW), Griffith (NSW) and Murrumbidgee (NSW) cotton regions, though samples were sometimes received from other regions including Central Qld and the Burdekin (Qld).

Not all regions were sampled each year; instead, we targeted areas where mites were prevalent in that season. Selection of regions and of fields to sample was facilitated by first contacting crop pest management consultants or farm agronomists in each cotton region. For each field sampled we collected about 50–60 mite-infested leaves (confirmed using a 10 × hand lens) walking a transect through the crop using methods described in the Cotton Pest Management Guide (Maas and Redfern 2018). The leaves were placed in paper bags which were labelled and transported in cool boxes to the Elizabeth MacArthur Agricultural Institute (EMAI), Menangle, NSW, for processing.

Seasons 2005–2006 to 2015–2016

In the cotton seasons 2005–2006 to 2015–2016 collections targeted only T. urticae. When we requested consultants or agronomists identify spider mite-infested fields we specified that we wanted only T. urticae. Nevertheless, fields identified frequently contained T. lambi, either in conjunction with T. urticae or alone. Hence, during sampling we regularly inspected leaves with a hand lens (as above) to ensure the mites collected were T. urticae and leaves infested with T. lambi were discarded.

Seasons 2016–2017 to 2017–2018

In the 2016–2017 and 2017–2018 cotton seasons, collections of mites were made differently as we aimed to evaluate if the mite species complex was changing. Ideally such a survey of mite species in cotton would use a geospatially referenced grid system (or similar) to ensure each species had equal opportunity to be encountered. However, due to the patchy distribution of mites across the industry, patchy distribution within fields (Wilson and Morton 1993), travelling time to different regions and limitation with possible cross contamination (Chen et al. 2020), the number of strains that could be managed was limited. Fields to be sampled in these seasons were selected as having spider mites, without specifying to the consultant or agronomists which species. If they asked us to specify, we simply said mites. In these two seasons only about 27% of samples were collected by us and the remainder by the consultants or agronomists (Table 1). If consultants or agronomists agreed to collect mites on our behalf, they were mailed a mite sample collection pack with detailed instructions. This consisted of a cardboard postage box so mites were not crushed, paper bags to contain the mites that could be stapled shut to prevent escape and cross contamination and finally a prepaid self-addressed post package for express return mail.

Table 1 Season, mite sample location, host plant and if tested for resistance (yes/no) for Tetranychus lambi and T. urticae strains collected during cotton seasons 2016–2017 and 2017–2018

Season 2018–2019

In that final season we aimed to collect as many T. urticae and T. lambi as practical and since the system with consultants and agronomists was working well all samples were collected by them.

Mite rearing

At EMAI, mite infested leaves from each field were inspected and mite identity confirmed with the aid of a stereo microscope using morphological criteria outlined in the Cotton Pest Management Guide (Maas and Redfern 2018) mites transferred onto insecticide free French bean (Phaseolus vulgaris L.) plants housed in individual mite proof cages. Mite populations developed on the French bean plants for several generations until all bioassays were completed. In the period from 2005–2006 to 2015–2016 only T. urticae were transferred to initiate culture and other mite species discarded. From 2016 to 2017 onward cultures were established for whatever mite species was found. However, as no T. ludeni were collected, only T. urticae and T. lambi was established. A separate cage was used for each sample. If a collection contained more than one mite species and numbers were adequate, separate cultures in separate cages were initiated for each species. Otherwise, the dominant species only was established, and the other noted (Table 1). The cages were maintained in an insectary running at 25 ± 4 °C and under natural light and ambient humidity using methods previously detailed for T. urticae (Herron et al. 1998).

Chemicals tested

Chemicals tested included abamectin 18 g/L emulsifiable concentrate (EC) (Vertimec® Miticide/Insecticide, Syngenta Australia), bifenthrin EC (Talstar® 250 EC Insecticide/miticide, FMC Australia), pre-season 2016–2017 propargite 600 g/L EC (Comite® Miticide, Crompton Specialties), post-season 2016–2017 propargite 300 g/kg wettable powder (WP) (Omite® 300 W Wettable Powder Miticide [as EC product was not available], Arysta Life Science Australia) and diafenthiuron. All were proprietary commercial insecticide formulations except diafenthiuron (Pegasus® Miticide/Insecticide, Syngenta Australia) for which we used the UV activated technical grade (94%) carbodiimide derivative of diafenthiuron (CGA-140408) instead. Etoxazole was tested via a DNA method that did not require chemical product. Tetranychus lambi was not tested against all chemicals because we considered abamectin, diafenthiuron and propargite more likely to be used under current Australian field conditions.

Resistance testing of Tetranychus urticae using a discriminating dose or a DNA method

Tetranychus urticae were exposed to the discriminating dose for each miticide tested. The bioassay procedure required 20–30 young adult female mites to be transferred from culture to French bean leaf discs (Herron et al. 2004). Briefly, three batches of mites on leaf discs plus a water only control were then sprayed with commercial insecticide diluted in reverse osmosis (RO) water except CGA-140408 that was first dissolved into acetone and then diluted in RO water. Each test was replicated at least once on a different day with new chemical and included a water only sprayed control that did not exceed 15% mortality. All sprays were applied with a Potter spray tower (Busvine 1971). After spraying, mites on leaf discs were maintained at 28 ± 0.1 °C in constant light for 48 h after which mortality (defined as not being able to walk in a coordinated manner) was assessed.

Tetranychus urticae were exposed to our unpublished historical discriminating doses of abamectin 0.001 g/L, diafenthiuron 0.02 g/L and propargite 0.2 g/L (WP only, EC baseline available at Wilson et al. (1995). In support of unpublished doses we have again retested the ‘Cheltenham’ reference susceptible T. urticae (Herron et al. 1998) and give those responses here. LC99.9 level responses produced are approximately 2–3× less than the discriminating doses used (Tables 2, 3 and 4) and so consistent with the 0.2 g/L rate used for bifenthrin that was 3× more than the LC99.9 level at that time (Herron et al. 2001).

Resistance testing of T. urticae against etoxazole did not begin until season 2015–2016 and was based on the presence of the I1017F mutation mutation discovered by Van Leeuwen et al. (2012). The mutation was validated as a resistance marker by Riga et al. (2017) and is acknowledged globally significant (Van Leeuwen et al. 2020) and known to cause etoxazole resistance in Australian T. urticae (Herron et al. 2018). Initially etoxazole resistance was detected by extracting genomic DNA using Chelex-100 resin (Bio-Rad, Hercules, CA, USA) according to the manufacturer’s instructions and screened for the presence of the I1017F mutation (Van Leeuwen et al. 2012) by direct sequencing of PCR amplicons performed at the Australian Genome Research Facility. Sequencing results were analysed using Sequencher v.5.2.4 (Gene Codes Corporation). This was superseded in 2017 by a high-throughput real-time diagnostic assay previously developed by us to detect resistance alleles (Chen et al. 2014) where for each strain we pooled 40 mites and estimated the resistance allele frequency based on a standard curve (Chen et al. 2014).

Full log-dose dose–response bioassay of Tetranychus urticae and T. lambi

We didn't have a susceptible T. lambi strain from which to develop a comparative baseline so instead we used a known susceptible T. urticae strain as a surrogate to give context. The ‘Cheltenham’ T. urticae population tested is acknowledged susceptible and has been maintained at EMAI isolated from other cultures since the late 1980s (Herron et al. 1998).

The bioassay procedure to achieve full log-dose probability regressions (for both species) required 20–30 young adult female mites to be transferred from culture to individual French bean leaf discs (Herron et al. 2004). Mites on leaf discs were then sprayed as above with serially diluted commercial insecticide or CGA-140408 diluted in RO water. Tetranychus lambi or T. urticae exposed to serial insecticide concentrations always included 100% mortality and three or four lower sprayed concentrations to yield full dose responses. As above, each test was replicated at least once on different days with freshly made chemical solutions and included a water only sprayed control that did not exceed 15%. After spraying, mites on leaf discs were maintained at 28 ± 0.1 °C in constant light for 48 h after which mortality (defined above) was assessed.

Resistance testing of Tetranychus lambi with discriminating doses generated

In the 2018–2019 season we used the same discriminating dose procedure described above for T. urticae and the discriminating doses calculated in this report for T. lambi to assess T. lambi populations collected for resistance.

Bioassay analysis

Discriminating dose data were pooled and percent mortality calculated following control mortality correction (Abbott 1925). Full dose responses were analysed via a stand-alone PC based probit program developed by Barchia (2001). Here, full dose–response analysis was done without replicate pooling to account for variability between replicates. LC50 and LC99.9 values were estimated using the method of Finney (1971) after control mortality correction (Abbott 1925). Minimum effective concentration data (MEC) was interpreted directly from the bioassay data being the lowest dose across all concentrations that achieved 100% mortality.


Species frequency 2016–2017 and 2017–2018

Forty-nine mite strains were collected; 30 in 2016–2017 and 19 in 2017–2018 (Table 1). Of the 2016–2017 collections, 16 were T. lambi only, seven were T. urticae only and seven included both species. Of the 2017–2018 collections 10 were T. lambi only, eight were T. urticae only and one included both species. Across both years T. lambi was collected more frequently (34 strains) than T. urticae (23 strains). Tetranychus ludeni was not recorded in any collections.

Dose responses against Tetranychus lambi and a susceptible T. urticae strain

Conspecific dose responses against T. lambi were established against abamectin (Table 2), diafenthiuron (Table 3) and propargite (Table 4). Across the T. lambi strains the calculated abamectin LC99.9 ranged from 0.00013 to 0.0035 g/L and was higher than the observed MEC range of 0.000125–0.00025 g/L. Against abamectin the 0.0007 g/L LC99.9 of the susceptible T. urticae reference strain overlapped many T. lambi results but the 0.0005 g/L MEC was higher. Against diafenthiuron, the calculated LC99.9 of the T. lambi strains ranged from 0.01 to 0.025 g/L that was again higher than the actual 0.005–0.01 g/L MEC range. Again the 0.021 g/L LC99.9 of the susceptible T. urticae reference strain against diafenthiuron overlapped that of T. lambi and in this instance so did the 0.01 g/L MEC, although it was at the highest end of the T. lambi range. In contrast, the calculated LC99.9 of 0.071 g/L for susceptible T. urticae against propargite did not overlap the calculated LC99.9 ranged of 0.32–0.75 g/L propargite for T. lambi being approximately an order of magnitude less tolerant against some stains. Similarly, the actual 0.0625 g/L MEC for the reference susceptible T. urticae was less than the 0.125–0.25 g/L MEC range for T. lambi.

Table 2 Dose responses for field collected Tetranychus lambi and a susceptible T. urticae (TSM SUSC) against abamectin
Table 3 Dose responses for field collected Tetranychus lambi and a susceptible T. urticae (TSM SUSC) against diafenthiuron
Table 4 Dose responses for field collected Tetranychus lambi and a susceptible T. urticae (TSM SUSC) against propargite

Discriminating dose responses for Tetranychus urticae strains

Abamectin resistance was not detected in T. urticae in Australian cotton before season 2007–2008 when 6% resistant individuals were detected in a single strain (Table 5). Resistance was detected again in season 2010–2011. Thereafter, resistance was consistently detected in 50–80% of strains, peaking at 80% in the 2018–2019 season. Resistance to diafenthiuron was not detected in any strains although some vigour tolerant survivors were detected in season 2013–2014 in three strains (discussed below). Resistance to propargite was detected sporadically peaking at three of nine strains (33%) in 2015–2016 but was not detected in 2018–2019. Resistance to bifenthrin was detected frequently in all seasons, ranging in frequency from 54 to 100%. Interestingly, etoxazole resistance was found in three of 10 strains in 2016–2017 but not in other seasons.

Table 5 Number of Tetranychus urticae strains etoxazolea resistant via molecular diagnosis plus strains resistant via bioassay against bifenthrin, abamectin, propargite and diafenthiuron (as CGA-140408) with discriminating dose range (i.e., percent susceptible) for mites collected from Australian cotton during seasons 2007–2011 to 2018–2019

Discriminating dose responses for Tetranychus lambi strains

Tetranychus lambi collected from cotton fields in 2018–2019 showed no abamectin, diafenthiuron or propargite discriminating dose survivors (Table 6).

Table 6 Percent susceptible Tetranychus lambi via bioassay against abamectin, propargite and diafenthiuron (as CGA-140408) as percent mortality at the discriminating dose (i.e., percent susceptible) for mites collected during season 2018–2019


The data from the 2016–2017 and 2017–2018 seasons support earlier anecdotal observations that there has been a shift in the spider mite species complex in Australian cotton. Reports from the early 1990s indicated that T. lambi was rarely encountered in cotton (Wilson 1992). In contrast, it is now more likely to be encountered than T. urticae. The cause(s) of this change is unknown but may relate to the progressive reduction in insecticide use in Australian cotton since the introduction of Bt-cotton (Wilson et al. 2018) and/or possibly a shift in the weed complex on cotton farms related to changed herbicide use patterns in glyphosate-tolerant cotton (Koetz 2020; Manalil et al. 2017).

Decreased insecticide use on Bt-cotton may have reduced the selective advantage that resistance affords T. urticae to miticides. This species, and probably other mite species such as T. lambi, colonise seedling cotton crops in spring and early summer (Wilson and Morton 1993) as they migrate from senescent winter hosts such as weeds (Wilson 1995). Historically, young cotton crops were sprayed several times with broad-spectrum insecticides, such as endosulfan, thiodicarb and the organophosphates omethoate or dimethoate (Wilson et al. 1998). This was done to prevent crop damage from pests such as Helicoverpa spp. and phytophagous thrips. As dimethoate resistance is common, and endosulfan and thiodicarb have moderate and low activity against T. urticae, respectively (Wilson et al. 1999), these insecticides had little effect on T. urticae but reduced the abundance of beneficial species. The lack of beneficials allowed T. urticae to build to economically damaging levels (Wilson et al. 1998). In contrast, insecticide use in Bt-cotton is low, especially on young cotton (Wilson et al. 2018), which may have reduced the resistance advantage T. urticae once enjoyed in a sprayed agroecosystem allowing competitive displacement by T. lambi.

Similarly, changes in the spectrum of weed hosts on farms may favour one mite species over another, thereby, either alone or in combination with changed insecticide use patterns, contributing to changes in relative species abundance in crops. Over the last 30 years the abundance of key weed species in the complex on Australian cotton farms has changed significantly (Koetz 2020; Manalil et al. 2017). In volunteer glyphosate-tolerant cotton, the seventh most important weed on Australian irrigated cotton farms in 2001 is now the most important. Amongst the eight major ‘non-cotton’ weeds in the 1991 surveys, only two were still in the top 8 in 2014–2015. This is likely due to multiple factors including changes in agronomic practice, the emergence of herbicide resistance and the advent of herbicide-tolerant genetically modified cotton (Koetz 2020; Manalil et al. 2017). These changes in the weed complex may influence the relative abundance of different mite species. For instance, if the change in weed species included hosts more favourable to T. lambi than T. urticae, it may influence their winter period survival. Each spring the subsequent relative abundance of each species available to colonize seedling cotton would be different. These changes may also be influenced by the drought years 2015–2019 where weed abundance and species composition may have been further altered.

During our annual sampling trips into Australian cotton we were often directed to fields that were subsequently found to be infested with T. lambi, although we had specified, we were collecting T. urticae only. This highlights potential issues with consultant or agronomist confusion in correct identification of the different mite species and highlights the need for improved diagnostics (Chen et al. 2020). Application of miticides to control non-economic infestations of T. lambi incurs avoidable costs and increases the risk of selecting miticide resistance in co-incidental T. urticae populations (e.g., Table 1) as well as in other pest species susceptible to the miticide applied. For instance, diafenthiuron is registered for control of spider mites, cotton aphids (Aphis gossypi Glover) and suppression of silverleaf whitefly (B. tabaci MEAM1). The issue of concurrent insect and mite selection has been addressed in part through regular updated extension information made available to industry on an annual basis (Maas and Redfern 2018).

The shift in mite species also highlights the urgent need to determine whether T. lambi can cause economic damage to Australian cotton and hence, when it may need to be controlled. Research on species-specific yield effects (e.g., Wilson 1993) or photosynthesis (e.g., Reddall et al. 2004) may provide answers to these important issues. Importantly, if T. lambi is demonstrated to reduce yield, further research is required to understand its host use and season population dynamics. This would provide a valuable insight into factors affecting its abundance. If T. lambi is demonstrated to cause economic loss, then control may be required. Our results suggest all three miticides tested in this study and currently applied against T. urticae should have efficacy against T. lambi. However, we note that T. lambi already seems very tolerant to propargite. If, in future, miticides are targeted against T. lambi this species can potentially develop resistance so it will be important to be able to monitor populations for changes in susceptibility and if necessary, include this species in the MRMS.

This study provides the basis for developing a resistance monitoring capability for T. lambi by establishing discriminating doses for three miticides used in Australian cotton. Ideally a discriminating dose should be lethal to susceptibles in a population without affecting the resistant types. It is an empirical compromise based on a two-stage approach: first, define the limits of tolerance; and two, based on stage one select a dose that accounts for all the susceptibles (Herron et al. 2014). The approach is similar to that outlined by Robertson et al. (2007) using the LC90–95 level (that is multiplied by a factor) and here we use an LC extrapolation in combination with MEC data. We do this because field-collected discriminating dose samples forwarded by growers are often limited. We consider it more important then to be confident any survivor is resistant than to potentially miss low level resistance. For that reason, the baseline probit regression extrapolation was to LC99.9 and also included observed tolerance via MEC to set discriminating doses against abamectin, diafenthiuron and propargite. Although we are confident survivors will carry a resistant allele(s) that frequency relationship to any potential control failure is complex dependant on the interaction of multiple factors.

Accurate setting of reliable discriminating doses is critical if false positives are to be avoided. Resistance testing during season 2013–2014 against T. urticae and diafenthiuron showed a few discriminating dose survivors in three strains tested suggesting incipient resistance. The result is reminiscent of that seen for A. gossypii in season 2009–2010 against diafenthiuron. Then, discriminating dose survivors were detected and a significant 4.9-fold diafenthiuron change in response was documented in the A. gossypii tested. Aphis gossypii discriminating dose survivors were detected for a further two seasons until the discriminating dose was altered. However, when surviving aphids were carefully investigated, vigour tolerance rather than resistance was concluded and the discriminating dose eventually doubled (Herron et al. 2012). Similarly, when strains showing T. urticae survivors were further tested here they all died at double the discriminating dose. This again suggested diafenthiuron vigour tolerance rather than insecticide resistance. At the time it was speculated that the discriminating dose for diafenthiuron against T. urticae should double to 0.04 g/L as done for A. gossypii. However, the dose was not immediately altered and discriminating dose survivors have not been detected since.

To avoid such false positives for T. lambi data available must be well considered. For all dose responses against T. lambi it is noteworthy that the range of tolerance was higher for the theoretical extrapolations (i.e., LC99.9) than the observed data (i.e., MEC). Based on calculated and observed tolerance we consider robust DDs for abamectin, diafenthiuron and propargite would fall within the range of ~ 0.0005–0.001, ~ 0.02–0.04 and ~ 0.5–1.0 g/L, respectively. Therefore, we propose DDs of 0.0007 g/L abamectin, 0.03 g/L diafenthiuron and 0.7 g/L propargite. Using these doses there was no evidence of resistance in the T. lambi strains collected in 2018–2019.

Ideally any baseline study would comprise naïve strains with no previous exposure to the chemicals being tested (Robertson et al. 2007) and an established reference susceptible for comparison. Without good baseline and an established susceptible data can be difficult to interpret, as Herron et al. (2011) concluded “it was not the resistant field-collected strains that were missing but a reliable susceptible baseline for the purpose of resistance factor calculation”. For that reason, we included dose responses for a known T. urticae susceptible for comparison and noted the responses for T. lambi were mostly similar except propargite. The result for propargite is perplexing because the product now gets little use in Australian cotton (Wilson et al. 2018) and would suggest that T. lambi is innately tolerant to it. It implies that if T. lambi did ever require targeted control then propargite would likely need further evaluation in field experiments before it could be recommended.

Since the advent of transgenic Bt-cotton, insecticide use against Helicoverpa spp. has declined dramatically (Wilson et al. 2018), but selection for resistance in mites has remained, especially for abamectin. Development of abamectin resistance highlights the challenges of managing resistance and concurrent selection. When C. dilutus emerged as a significant pest mite, outbreak risk was managed via the prophylactic addition of a second chemical (Wilson et al. 2018). As a result, selection for abamectin resistance in T. urticae substantially increased and resistance developed. The abamectin resistance causing mechanisms in T. urticae are well studied (Xue et al. 2020), so it is interesting then that abamectin resistance appears absent from T. lambi although it likely experiences the same concurrent resistance selection as T. urticae. No resistance was recorded to diafenthiuron; however, use of diafenthiuron has increased targeting B. tabaci MEAM1 (Wilson et al. 2018) possibly selecting for vigour tolerance. Here it must be acknowledged that diafenthiuron is known to be metabolised in vivo by a P450 monooxygenase and decreased activation by down regulation is a known resistance mechanism (Feyereisen 2015). Therefore, using the carbodiimide (as used here) instead of diafenthiuron may cause resistance by decreased activation to be missed. Etoxazole resistant individuals were confirmed in a single season only and a lack of resistance to etoxazole is not surprising due to very low use with cross resistance from clofentezine/hexythiazox (Demaeght et al. 2014) also unlikely because neither are used in Australian cotton. The limited etoxazole detections may relate to Australian Cotton Research Institute (ACRI) based experimental use and if proven correct would suggest etoxazole resistance is quickly selected. This highlights the ongoing need to continue to monitor resistance in this species and to adapt the mite resistance management strategy in response to changes.


  1. Abbott WS (1925) A method for computing the effectiveness of an insecticide. J Econ Entomol 18:265–267

    CAS  Article  Google Scholar 

  2. Barchia I (2001) Probit analysis and fiducial limits. In: Paper presented at the genstat 2001 conference Goldcoast, Australia

  3. Busvine JR (1971) A critical review of the techniques for testing insecticides. Commonwealth Agricultural Bureaux, Slough

    Google Scholar 

  4. Chen YZ, Bogema DR, Barchia IM, Herron GA (2014) Quantification of the pirimicarb resistance allele frequency in pooled cotton aphid (Aphis gossypii glover) samples by TaqMan SNP genotyping assay. PLoS ONE 9:12.

    CAS  Article  Google Scholar 

  5. Chen YZ, Woolley L, Nguyen D, Gupta R, Chandler GT, Nehl D, Herron GA (2020) Development and use of a single real-time PCR assay to identify the three spider mite species Tetranychus urticae, Tetranychus lambi and Tetranychus ludeni (Acari: Tetranychidae). Austral Entomol 59:630–638.

    Article  Google Scholar 

  6. Demaeght P et al (2014) High resolution genetic mapping uncovers chitin synthase-1 as the target-site of the structurally diverse mite growth inhibitors clofentezine, hexythiazox and etoxazole in Tetranychus urticae. Insect Biochem Mol Biol 51:52–61.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  7. Feyereisen R (2015) Insect P450 inhibitors and insecticides: challenges and opportunities. Pest Manage Sci 71:793–800.

    CAS  Article  Google Scholar 

  8. Finney DJ (1971) Probit analysis, 3rd edn. Cambridge University Press, Cambridge

    Google Scholar 

  9. Gutierrez J, Schicha E (1983) The spider mite family Tetranychidae (Acari) in New South Wales. Int J Acarol 9:99–116

    Article  Google Scholar 

  10. Helle W, Sabelis MW (1985) Spider mites: their biology, natural enemies and control. World Crop Pests Volumes 1A and 1B. Elsevier, Amsterdam, The Netherlands

    Google Scholar 

  11. Herron G, Wilson L (2016) Mite resistance danger from over-use of abamectin. Aust Cottongrower 37:14–16

    Google Scholar 

  12. Herron GA, Edge VE, Wilson LJ, Rophail J (1998) Organophosphate resistance in spider mites (Acari : Tetranychidae) from cotton in Australia. Exp Appl Acarol 22:17–30.

    CAS  Article  Google Scholar 

  13. Herron GA, Rophail J, Wilson LJ (2001) The development of bifenthrin resistance in two-spotted spider mite (Acari: Tetranychidae) from Australian cotton. Exp Appl Acarol 25:301–310.

    CAS  Article  PubMed  Google Scholar 

  14. Herron GA, Rophail J, Wilson LJ (2004) Chlorfenapyr resistance in two-spotted spider mite (Acari: Tetranychidae) from Australian cotton. Exp Appl Acarol 34:315–321.

    CAS  Article  PubMed  Google Scholar 

  15. Herron GA, Langfield BJ, Tomlinson TM, Mo JH (2011) Dose-response testing of Australian populations of onion thrips Thrips tabaciLindeman (Thysanoptera: Thripidae) further refines baseline data and detects methidathion and likely imidacloprid resistance. Aust J Entomol 50:418–423.

    Article  Google Scholar 

  16. Herron GA, Langfield BJ, Wilson LJ (2012) A new discriminating dose to monitor diafenthiuron (CGA140408) resistance in cotton aphid Aphis gossypii glover (Hemiptera: Aphididae). Gen Appl Entomol 41:17–20

    Google Scholar 

  17. Herron GA, Langfield BJ, Bogema DR, Chen YZ (2014) Baseline susceptibility and cross-resistance in Aphis gossypii Glover (Aphididae: Hemiptera) to phorate and sulfoxaflor. Austral Entomol 53:32–35.

    Article  Google Scholar 

  18. Herron GA, Woolley LK, Langfield KL, Chen Y (2018) First detection of etoxazole resistance in Australian two-spotted mite Tetranychus urticae Koch (Acarina: Tetranychidae) via bioassay and DNA methods. Austral Entomol 57:365–368.

    Article  Google Scholar 

  19. Koetz E (2020) Herbicide resistance in Australian cotton farming systems—a journey through time. Aust Cottongrower 41:14–15

    Google Scholar 

  20. Maas S, Redfern R (2018) Cotton pest management guide 2019–2020. Greenmount Press, Toowoomba

    Google Scholar 

  21. Manalil S, Werth J, Jackson R, Chauhan BS, Preston C (2017) An assessment of weed flora 14 years after the introduction of glyphosate-tolerant cotton in Australia. Crop Pasture Sci 68:773–780.

    Article  Google Scholar 

  22. Reddall A, Sadras VO, Wilson LJ, Gregg PC (2004) Physiological responses of cotton to two-spotted spider mite damage. Crop Sci 44:835–846.

    Article  Google Scholar 

  23. Riga M et al (2017) The relative contribution of target-site mutations in complex acaricide resistant phenotypes as assessed by marker assisted backcrossing in Tetranychus urticae. Sci Rep 7:12.

    CAS  Article  Google Scholar 

  24. Robertson JL, Russell RM, Preisler HK, Savin NE (2007) Bioassays with arthropods, 2nd edn. CRC Press, Boca Raton

    Google Scholar 

  25. Van Leeuwen T et al (2012) Population bulk segregant mapping uncovers resistance mutations and the mode of action of a chitin synthesis inhibitor in arthropods. Proc Natl Acad Sci USA 109:4407–4412.

    Article  PubMed  Google Scholar 

  26. Van Leeuwen T, Dermauw W, Mavridis K, Vontas J (2020) Significance and interpretation of molecular diagnostics for insecticide resistance management of agricultural pests. Curr Opin Insect Sci 39:69–76.

    Article  PubMed  Google Scholar 

  27. Wilson LJ (1992) Identifying mites on cotton. Aust Cottongrower 13:68–69

    Google Scholar 

  28. Wilson LJ (1993) Spider-mites (Aari, Tetranychidae) affect yield and fiber quality of cotton. J Econ Entomol 86:566–585.

    Article  Google Scholar 

  29. Wilson LJ (1995) Habitats of twospotted spider-mites (Acari, Tetranychidae) during winter and spring in a cotton-producing region of Australia. Environ Entomol 24:332–340.

    Article  Google Scholar 

  30. Wilson LJ, Morton R (1993) Seasonal abundance and distribution of Tetranychus urticae (Acari, Tetranychidae), the 2 spotted spider-mite, on cotton in Australia and implications for management. Bull Entomol Res 83:291–303.

    Article  Google Scholar 

  31. Wilson LJ, Herron GA, Leigh TF, Rophail J (1995) Laboratory and field-evaluation of the selective acaricides dicofol and propargite for control of Tetranychus urticae Koch (Acari, Tetranychidae) in Australian cotton. J Aust Entomol Soc 34:247–252

    Article  Google Scholar 

  32. Wilson LJ, Bauer LR, Lally DA (1998) Effect of early season insecticide use on predators and outbreaks of spider mites (Acari: Tetranychidae) in cotton. Bull Entomol Res 88:477–488.

    CAS  Article  Google Scholar 

  33. Wilson LJ, Herron GA, Bauer LR, Lally DA (1999) Acaricidal and stimulatory effects of insecticides on Tetranychus urticae Koch (Acari: Tetranychidae) in cotton. Aust J Entomol 38:30–33.

    Article  Google Scholar 

  34. Wilson LJ, Whitehouse ME, Herron GA (2018) The management of insect pests in Australian cotton: an evolving story. Annu Rev Entomol 63:215–237

    CAS  Article  Google Scholar 

  35. Xue WX, Snoeck S, Njiru C, Inak E, Dermauw W, Van Leeuwen T (2020) Geographical distribution and molecular insights into abamectin and milbemectin cross-resistance in European field populations of Tetranychus urticae. Pest Manage Sci 76:2569–2581.

    CAS  Article  Google Scholar 

Download references


The Cotton Research and Development Corporation provided funding support for this research via project DAN1507. Consultants and agronomists who forwarded samples of mites for identification and resistance testing are gratefully recognised. The many staff that provided technical support over the many study years and are thanked. Susan Maas and Emma Cottage provided critical comment on an early draft.

Author information



Corresponding author

Correspondence to Grant A. Herron.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Herron, G.A., Langfield, K.L., Chen, Y. et al. Development of abamectin resistance in Tetranychus urticae in Australian cotton and the establishment of discriminating doses for T. lambi. Exp Appl Acarol (2021).

Download citation


  • Tetranychus urticae
  • Competitive displacement
  • Fitness
  • Resistance