The pollen beetle, Brassicogethes aeneus (F.), syn. Meligethes aeneus (F.) (Coleoptera: Nitidulidae), is a major pest in oilseed rape ( Brassica napus L.) (OSR) fields in most European countries ( Alford et al. , 2003 ), including Serbia ( Sekuliċ & Kereši, 2007 ; Milovanoviċ et al. , 2013 ). In Serbia, SE of Europe, only winter OSR is grown commercially, mostly in its northern parts (Vojvodina Province) and in Central Serbia, the production focusing in the downriver zo­ne of the Danu­be in Serbia. Winter OSR crops have been continually grown in the observed region over many years, and control measures (including biocontrol) have often included insecticides with different modes of action (organophosphates, pyrethroids and neonicotinoids) to control B. aeneus .

Yield loss assessments in Europe have been repor­ted to reach 50-60% ( Nilsson, 1987 ), or 80% in spring OSR in Denmark ( Hansen, 2003 , 2004 ), while official data for Serbia and its region are missing. Damage is greater in seasons with cold spring because such weather slows down and extends plant development during sensitive growth stages. Harmfulness of B. aeneus depends on their numbers on OSR inflorescences, the timing of their appearance with regard to bud development, the period of time elapsing from the moment of B. aene­us settling to OSR flowering, the variety-depending capacity of regeneration of OSR crops, the cultiva­tion methods applied, and of the weather conditions. Extreme damage occurs when a spell of warm weather at the end of winter causes B. aeneus to appear earlier in OSR crops, which only start to form flowering buds at the time. When cold weather delays OSR flowering, fewer B. aeneus adults are able to cause considerably more damage than more are able to do in seasons with shorter OSR flowering periods. Similar findings were reported by Bergant et al. (2005) for the developmen­tal dynamics of onion thrips Thrips tabaci (Lindeman) (Thysanoptera: Thripidae) in Slovenia.

The presence of B. aeneus adults in OSR crops is easy to confirm by inspecting of inflorescences because the glossy, black insects that are often covered with pollen dust are actively mobile and easy to spot, while yellow water traps or similar devices put up in fields are attractive to flying species. The importance of mo­­ni­toring B. aeneus populations is associated with rational and environment-friendly control measures for the pest because, based on their abundance assessment and adult infestation level, it is possible with greater precision to evaluate damage that the pest may cause to the crop, as well as pesticide treatments required ( Williams, 2004 ). Traditionally, control of B. aeneus in winter OSR is based on spring insecticide treatments, usually with pyrethroids. However, where populations are potentially exposed to intensive applications of insecticides with the same mode of action, such as pyrethroids, and treatments are often consecutive within season, there is a high level of concern regarding the increased potential for selection pressure, which leads to increasingly widespread resistance development and prevents insecticide products from remaining active against the pest.

Since the earliest reports on B. aeneus resistance to pyrethroids in France and Scandinavia ( Decoin, 2002 ; Hansen, 2003 , 2004 , 2008 ), its spreading in recent years across Europe has caused some major pest control problems in OSR fields in many European countries, e.g. France ( Ballanger et al. , 2007 ), UK ( Richardson, 2008a ), Denmark ( Hansen, 2003 , 2004 , 2008 ; Kaiser et al. , 2018 ), Poland ( Węgorek et al. , 2009 ; Philippou et al. , 2011 ), Germany ( Heimbach et al. , 2006 ; Thieme et al. , 2008 ; Heimbach & Müller, 2013 ), Switzerland ( Philippou et al. , 2011 ), Austria ( Slater et al. , 2011 ), Sweden ( Kazachkova, 2007 ) and Czech Republic ( Stará & Kocourek, 2017 ). Based on initial evidence, resistance monitoring activities have been initiated in many EU countries ( Zimmer & Nauen, 2011 ), while organizations such as the European Plant Protection Organization (OEPP/EPPO) and Insecticide Resistance Action Committee (IRAC) have developed methods for testing the susceptibility of B. aeneus to organophosphates, pyrethroids and neonicotinoides, na­mely the dipping test and adult vial test, respectively ( IRAC, 2006 , 2009a , b ; Thieme et al. , 2008 ).

Considering that B. aeneus studies have been mi­ssing in Serbia and no data are available on its populations dynamic and susceptibility to insecticides, three locations were chosen in the Danube downriver zone in order to create an appropriate pest management programme. The present study was design: 1) to de­termine the populations dynamic of B. aeneus during the critical phases of winter OSR growth by applying two sampling methods, the yellow water traps and the beating method, and 2) to conduct laboratory tests on the collected populations in order to determine toxicity/susceptibility levels to selected insecticides that have different modes of action (organophosphates, pyrethroids and neonicotinoid), applying two types of tests, the adult vial test and the dipping test.

B. aeneus emerged in winter OSR crops in the downriver stretch of the Danube in Serbia during March when crops were at the stem elongation growth stage (BBCH 30-31) with maximum daily temperature exceeding 10°C, which was noted on March 12 2009, and March 22 2010. According to Láska & Kocourek (1991) first B. aeneus adults emerged when temperature exceeded 10.2°C, or at a later growth stage (BBCH 51-54) in the second half of March ( Williams, 2006 ). The data show that emergence of B. aeneus adults depends on favourable temperature and that B. aeneus fly over from other early flowering plants into OSR crops when it exceeds 15°C ( Alford et al. , 2003 ).

B. aeneus adult numbers continued to grow during the later OSR growth stage (BBCH 50-59), and their highest counts were noted at the green-yellow bud growth stage (BBCH 57-59). After that, in May (BBCH 61-69), their numbers decreased substantially. Similarly, other authors showed that B. aeneus adults were most numerous at bud extension and yellow bud stages and thereafter declined ( Sedivy, 1993 ; Walter & Northing, 2007 ; Petraitiene et al. , 2008 ; Vaitelyte et al. , 2011 ; Metspalu et al. , 2015 ). A wider range of maximum B. aeneus abundance regarding OSR growth stages has been reported for stages from the late green bud until the beginning of flowering (BBCH 53-63), which occurs in April ( Williams, 2006 ).

However, the BM is required for determining the threshold for timely and cost-efficient treatments regarding insect number. Insecticide applications therefore may be delayed for some 10 days after the conventional timing "before flowering" ( Maceljski, 1999 ). In Serbia, other thresholds given by Maceljski are acceptable ( Maceljski & Jelovčan, 2007 ). During the test years, threshold was exceeded on the test location during the sensitive growth stages of winter OSR, and it was from 2.8 to 5.8 adult/terminal flower in 2009, and from 2.5 to 3.5 adult/terminal flower in 2010. Thresholds in winter OSR crops in European countries are given either as numbers per terminal flower (Croatia, Slovakia; range 0.8-3 adult/terminal flower based on growth stage), or as numbers per plant (in Switzerland, Czech Republic, France, Denmark, Hungary, Latvia, Luxemburg, Norway, Slovenia, Poland, Netherlands and Sweden) with numbers ranging from 1 to 6 per plant and commonly graduating according to growth stage ( Richardson, 2008b ). Critical thresholds in Germany and France are 3-4 adult /terminal flower at the BBCH 50-51 stage, 7-8 at the BBCH 52-53 stage, and more than 8 at the BBCH 55-59 stage ( Williams, 2010 ).

Our results show that considerably more B. aeneus were found on the location Kovin than in Smederevo and Požarevac over both years of trials. Also, more B. aeneus were caught in 2009 than in 2010. In a similar study in Lithuania, the least B. aeneus , 2.5 adult/flower were caught in 2009, and the most in 2007, 18 adults/flower ( Vaitelyte et al. , 2011 ). Differences in the abundance of B. aeneus over the years are common, and could be the result of several factors, e.g. differences in the rate of plant growth triggered by weather conditions, direct effects of temperature, humidity or plant species ( Petraitiene et al. , 2008 ; Vaitelyte et al. , 2011 ). In Lithuania, OSR developed faster in seasons with minimum daily temperature exceeding 15°C, buds developed over a briefer time period of 15 days, and 3 adult/flower were counted; in years with maximum daily temperatures close to 15°C, when bud formation took a longer time (21-27 days), the counts reached as much as 7 adult/flower ( Petraitiene et al. , 2008 ). In our present study, monitoring of the duration of bud development over the period 2009-2010 showed that this process was longer in 2009 (29 days), and showed a positive correlation with B. aeneus adult counts, compared with 2010. Harsh winter temperatures can also have harmful impact on overwintering B. aeneus and reduce the number of emerging B. aeneus in spring ( Hokkanen, 2000 ).

The results of susceptibility testing obtained in this study based on the parameters LC ₅₀ and LC ₉₅ and appropriate lc-p lines (for lambda-cyhalothrin and thiacloprid) showed that all tested insecticides were highly toxic to all three tested populations of B. aeneus with toxicity increasing with the duration of B. aeneus exposure. Also, the lc-p lines, especially the slopes and overlapping/non-overlapping of confidence intervals, showed that the resulting insecticide toxicity changed either not at all or very low, depending on population/location of B. aeneus , season in which they were sam­pled and tested, and the type of test (AVT and/or DT).

In AVTs, only bifenthrin and pirimiphos-methyl did not change their toxicity after exposing for 1 h all three test populations of B. aeneus collected in 2009 and 2010, and the same effect of lambda-cyhalothrin and alpha-cypermethrin was found regarding B. aeneus collected in 2009, and chlorpyrifos+cypertmethrin and thiacloprid as affecting B. aeneus collected in 2010. However, lambda-cyhalothrin was about 4-fold less toxic to B. aeneus from Kovin collected in 2010 than to populations from Smederevo and Požarevac, while alpha-cypermethrin applied to Požarevac popu­lation collected in 2010 and thiacloprid to those co­llected in 2009 were c . 3 times more toxic. After 5 and 24 h exposure, all test insecticides showed mostly unchanging level of toxicity to all three populations collected in 2009 and 2010 or a change only occur in B. aeneus collected in 2010 at the LC ₅₀ level, as decrea­sing of alpha-cypermethrin and bifenthrin toxicity for Kovin and Požarevac populations than for Smederevo and Požarevac populations, some 2.5- and 6-fold, and 4- and 3-fold, respectively. Also, the two-year AVT testing revealed that the recommended doses (showed in Table 1) and 75% of the recommended doses of all insecticides caused ≥ 95% mortality of B. aeneus in all populations, as well as some 50%, 25%, 10% and even 5% recommended doses. The only data exceeding the recommended dose (0.48 µg/cm ² ) referred to thiacloprid treatment of Kovin population for 24 h at the LC ₉₅ , 1.06 µg/cm ² in 2009 and 1.03 µg/cm ² in 2010.

Regarding the DT type of testing, all test insectici­des retained stable toxicity to all three populations collected in 2009 and 2010 after 1, 5 and 24 h of B. aeneus exposure, except pirimiphos-methyl, which was after 5 h exposure at the LC ₅₀ level 2 times less toxic to Kovin population than to population from Smederevo and Požarevac. The analysis of the two-year results of DTs for all test populations showed that the recommended concentrations and 75% re­commended concentrations of all insecticides, except thiacloprid, caused ≥ 95% mortality, even at 50%, 25%, 10% and 5% recommended concentrations. However, the LC ₉₅ data for Kovin population exposed for 24 h were higher than the recommended concentrations for: lambda-cyhalothrin 0.03 mg/L in 2010 and 0.04 mg/L in 2009, alpha-cypermethrin 0.07 mg/L in 2010, and bifenthrin 0.06 mg/L in 2009 and 2010. Concerning the population Smederevo, the data for 24 h exposure at the LC ₉₅ showed that they exceeded the recommended dose for bifenthrin 0.06 mg/L in 2009 and 2010, while in population from Požarevac the value for alpha-cypermethrin was 0.04 mg/L in 2009. Interestingly, thiacloprid achieved ≥ 95% B. aeneus mortality only to Smederevo population in 2009, while in all other trial variants after 24 hours of exposure at the LC ₉₅ it was higher than the recommended concentration (0.1584 mg/L), 0.22-1.06 mg/L.

Available literature has largely documented variable toxicity data for different classes of tested insectici­des (with different modes of actions) to B. aeneus populations, and their variation in sensitivity under laboratory tests, which depends on population and year of study. Studies in Germany on the suscepbility of populations of B. aeneus , and some other insect pest species to the pyrethroids lambda-cyhalothrin and cypermethrin revealed a notable reduction in suscepbility only in B. aeneus , while it was lower in C. napi and C. pallidactylus . However, data aquired in laboratory testing may be assumed valid only when they are considered in the context of corresponding field data ( Heimbach et al. , 2006 ). Therefore, special field trials Milovanoviċ et al. (2013) were organized to assess the efficacy of those insecticides using OEPP/EPPO methodology on the same locations during three successive vegetation seasons in 2008, 2009 and 2010 at the winter rapeseed development stage of visible flower buds but still closed (BBCH 55-59) by counting the present B. aeneus adults. In 2009, three days after treatment, all tested insecticides achieved efficacy ≥ 92%, except thiacloprid at Kovin and Smederevo locations, while all insecticides showed significantly lower efficacy, 79-92%, at all three locations after seven days of treatment. In those intervals, the highest efficacy (≥ 90%) was demonstrated by pirimiphos-methyl and chlorpyrifos+cypermethrin, at all three locations, and by lambda-cyhalothrin and alpha-cypermethrin applied to Požarevac population. In 2010, no statistically sig­nificant differences were detected between the tested insecticides at the locations Smederevo and Požarevac three days after treatment, while only thiacloprid showed a significantly lower efficacy against Kovin population than the other insecticides. As in 2008 and 2009, the efficacy of the tested insecticides seven days after treatment was again lower than it was after three days, 76-87%.

The results of the testing of three B. aeneus populati­ons at three downriver locations in the Serbian section of the Danube showed that the number of B. aeneus in winter OSR was highest during the green-yellow bud stage, exceeding the economic threshold. Considerably more B. aeneus were found in Kovin locality than in Smederevo and Požarevac, and in 2009 rather than in 2010. According to our susceptibility rating scheme, the populations tested in this ex­periment were found to belong to the second group, i.e. populations susceptible to the test insecticides, while B. aeneus from the location Požarevac were found as most susceptible, and Kovin population was least susceptible. Also, the data acquired by the AVT method proved more precise results. Lambda-cyhalothrin was found to be the most toxic insecticide in most trials, while thiacloprid was least toxic. However, as resistance of B. aeneus populations has been documented in many European countries, it is necessary to revaluate population susceptibility to lambda-cyhalothrin, thiacloprid and other insecticides in the same locations (especially Kovin) within the next ten years. Populations from other locations along the Danube should also be examined to provide a basis for creating a useful Insecticide Resistance Management (IRM) strategy and incorporating it into Integrated Pest Management program for B. aeneus control in Serbia and region.