Effects of Insect Growth Regulators on Honey Bees and Non-Apis Bees. a Review

Apidologie 32 (2001) 527–545 527 © INRA/DIB-AGIB/EDP Sciences, 2001

Review article

Effects of insect growth regulators on honey bees and non-Apis bees. A review

Jean-Noël TASEI*

Unité de Recherches de Zoologie, INRA, 86600 Lusignan, France

(Received 26 December 2000; revised 17 May 2001; accepted 24 July 2001)

Abstract – The insect growth regulators (IGRs) are ecdysone or juvenile hormone mimics, or chitin synthesis inhibitors. They are more likely to be hazardous to larval insects than to adults. Application of JH mimics to adult honey bees may affect foraging behaviour and some physiological traits. Top- ical and feeding tests revealed that application of IGRs to larvae may result in death and larval ejection by workers, malformed larvae and pupae with typical rimmed eyes, or malformed adults. Sev- eral laboratory “larvae tests” using artificially contaminated diets have been described for honey bees and bumble bees. Field and cage methods have also been published for honey bees and bumble bees respectively. Diflubenzuron was generally safe for honey bee brood in fields treated at 35 to 400 g.ha–1 a.i. and harmful to bumble bees at 300 g.ha–1 a.i. Fenoxycarb was safe for bumble bees at 1200 g.ha–1 a.i. and hazardous to honey bees causing damage to honey bee brood at 140 g.ha–1. insect growth regulator / honey bee / non-Apis bee / toxicity / risk assessment

1. INTRODUCTION insect growth regulators or IGRs, have been used in insect pest control for more than After Wigglesworth’s pioneering studies 25 years. These insecticides of the third gen- performed during 1933–1940 (Wiggelsworth, eration appeared after the early generation of 1972) on the control of insect growth by arsenates, insecticides of mineral origin, and endocrine organs, Williams (1956) suggested the second generation of organic synthetic that in addition to their theoretical interest, compounds such as DDT. The mode of juvenile hormones could be accurately iden- action of IGRs is quite original since they tified, then synthesised and used as insec- are not stomach poisons and they do not ticides. His prospects have been achieved exhibit neurotoxicity but they disrupt moult- since several commercial compounds called ing process or cuticle formation. Larval

* Correspondence and reprints E-mail: [email protected] 528 J.-N. Tasei characters are maintained by the juvenile the toxicity of different compounds to adults hormones (JH) which are secreted by cor- or larvae or to assess the risks of field treat- pora allata. The JH are hormones of iso- ments. Fewer articles report studies on non- prenoid nature which prevent the breakdown Apis bees, most of which are concerned with of the thoracic gland (Wigglesworth, 1972). the bumble bee, Bombus terrestris L. Pupal moulting is determined by a circulat- ing hormone: the moulting hormone, secreted by the prothoracic glands which 2. EFFECTS OF IGRS ON HONEY are activated by neurosecretory cells. This BEES hormone triggers changes in the epidermis and deposition of the new cuticle. The 2.1. Symptoms of IGR intoxication purified form, of steroid nature is called in adults ecdysone. Similar substances can be isolated from many plants. IGRs are commer- The effects of juvenile hormone mimics cial hormones mimics that influence moult- such as methoprene were studied on ing as hormones do, acting at the cellular workers by Redfern and Knox (1974). They level in various ways depending on their tested this chemical, formulated in acetone, chemical constitution: tebufenozide influ- by topical application on adult workers ences moulting as ecdysone does, while lightly anaesthetised and did not find any pyriproxyfen and fenoxycarb are JH mimics. mortality for concentrations ranging from 1 A third class is that of chitin inhibitors such to 1000 µg/bee. Robinson (1985) also applied as diflubenzuron, flufenoxuron, hexaflu- methoprene dissolved in acetone topically on muron, and lufenuron. In addition, a natural workers which were marked and reintro- substance from neem tree, Azadirachta duced into their colony for behavioural indica, proved to be a potent IGR (Rembold observations. He found that bees treated et al., 1980). According to Engels (1990) with 250 µg of JH analogue started foraging IGRs cause little or no damage to adult earlier than control ones. Lower doses, 25 honey bees and generally the typical effects and 2.5 µg caused no significant effect. For- of these compounds can be seen after the aging performances estimated by the num- moult of the exposed stage. All natural or ber of foraging trips per hour and the dura- synthetic compounds with IGR properties tion of the foraging period were not may be suspected of being hazardous to the influenced by methoprene whereas the pro- brood of Apis or non-Apis bees since their duction of two alarm pheromones was larvae can be exposed to insecticides by an induced prematurely. Jaycox et al. (1974) oral route. Davis (1989) contributed to the tested another synthetic JH mimic, the Law- understanding of the oral pathway of insec- Williams mixture. They injected experi- ticides, mediated by adult nurse honey bees mental doses ranging from10 to 200 µg/bee, which regurgitate honey sac content to feed between metasomal tergites, using either larvae. In solitary bees, George and Rincker olive or mineral oil as a carrier and found (1985), Tasei and Carré (1987), and Tasei bees ate less pollen after injection of JH et al. (1988), showed that Megachile rotun- mimic. Treated bees could not develop their data Fabr. females provisioning their nest hypopharyngeal glands and started to move with pollen and nectar can also transfer out of the brood nest, to guard the hive insecticides to their larvae via the food entrance, to fly and to collect pollen, sooner stored in the cell prior to egg laying. than control bees. There was a dose-effect Most of scientific articles on IGR effects relation in the onset of activity outside the on bees are concerned with Apis mellifera L. colony. JH mimic treatment also reduced They describe the symptoms observed on longevity by 39%. Rutz et al. (1974) applied larvae, and report on various methods to test JH III and the IGR triprene on freshly Impact of insect growth regulators on bees 529 emerged workers by feeding or injection and observed abnormal leucocytes, the inhi- and found that both hormones caused a bition of hypopharyngeal glands, the reduction in the total protein content of decrease of the total haemolymph proteins the haemolymph and a change in the form and the reduction of longevity. of leucocytes. At a high dose of JH III (1 µg/bee) the hypopharyngeal glands regressed which hampered larva feeding by 2.2. Symptoms of IGR intoxication workers. Triprene at 1 µg/bee shortened the in larvae life of queen-less workers. Atkins et al. (1976) measured the effects of hundreds of After topical application of the juvenile pesticides on bees and listed active sub- hormone analogue methoprene on 3 and stances and formulations according to their 5 day old larvae Hussein and Abdel-Aal hazards to bees in field tests for commonly (1978) observed malformations of the used materials. For less frequently used abdomen, wings and wax glands in adults. In compounds they referred to laboratory data. addition, more than 50% of treated larvae were removed from their brood cell. Zdarek Dimilin®, formulated with diflubenzuron, and Haragsim (1974) studied the morpho- one of the most common commercial IGR genetic action of 33 structurally unrelated appeared in the group of the “relatively non- JH analogues chemicals by applying them toxic” insecticides. Usha and Kandasamy on worker bee larvae either during their (1986) exposed Apis cerana Fabr. for 90 min feeding period or later. The early treatment to surfaces treated with several insecticides induced the development of imaginal char- including diflubenzuron. As no mortality acters of queen and the late administration . –1 occurred even at 10000 mg kg the authors inhibited the differentiation of imaginal concluded that this chitin synthesis inhibitor structures. Beetsma and Ten Houten (1975) was the safest and should be recommended tested JH I, II, III, IV, V and VI by mixing for integrated pest management. Difluben- the compounds with the food supplied to zuron was also tested on newly emerged small colonies or by spraying it on rape- adults of A. mellifera and A. cerana indica seed flowering in a flight cage. After treat- by Gupta and Chandel (1995). They found ment with JH I, II, IV and V they did not that topical applications of 100 µg doses observe any mortality in workers but found were tolerated by treated bees but resulted in incompletely coloured adults with queen- a reduced weight gain. The same dose like characters. With JH IV and V, the administrated per os proved fatal, but 50 µg queens died and all larvae disappeared were tolerated and suppressed the develop- within the week following application. JH V ment of hypopharyngeal glands in both inactivated hypopharyngeal glands of adult species. This oral treatment also affected workers. Almost no brood was affected by weight gain, indicating that at high doses, JH III and VI, and JH I and II resulted in an this IGR can be harmful to adults. Acute increase of brood. Similar methods were toxicity tests performed on adult A. mellifera used by Hrdy and Skrobal (1976) who found with another IGR, flucycloxuron (Andalin®) that application of JH I and JH II reduced proved this material safe in integrated pest brood quantity. Gerig (1975) found that control (Ceparano and Job, 1989). Pyriprox- kinoprene inhibited the development of yfen, applied at concentrations higher than hypopharyngeal glands. 1.25 µg, to newly emerged workers, impaired Feeding small colonies with difluben- vitellogenin synthesis in the haemolymph zuron reduced the amount of capped and (Pinto et al., 2000). Gerig (1975) fed work- uncapped brood while increasing the num- ers with pollen, artificially contaminated by ber of eggs laid (Chandel and Gupta, 1992) the juvenile hormone analogue kinoprene which was presumably a sign of an ovicidal 530 J.-N. Tasei action of this compound as demonstrated workers. The application of both substances on other species by Grosscurt (1978). After resulted in growth disruptions, causing mal- addition of diflubenzuron to royal jelly fed formations and high pupal mortality. The to queen larvae Nitsch et al. (1994) observed less pure fraction showed an anti-feedant that at the concentration 0.05%, treated lar- effect but did not affect weight gain of larvae either died or were ejected by workers. vae. One tenth of the dose effective for pests, Even at the lowest tested concentration caused disturbance in larval development 0.00625%, the authors found some larvae of honey bees. removed from their cells. When queens died before emergence they showed typical damage on head and thorax. El-Din et al. (1990) 2.3. Methods for testing the toxicity topically applied diflubenzuron and also two of IGRs to bee larvae other IGRs, triflumuron and chlorfluazuron at the LD 25, to 3 day old larvae and reported Atkins and Kellum (1986) were con- reduced weight in emerging adults by 19.3, vinced that pesticides could be deposited in 39.6, and 29.1%. When treated at LD 50 or the hive through contaminated pollen. Com- LD 90, larvae were removed by nurse bees pounds which are hazardous to brood and and no sealed brood could be found. likely to be transferred to hives by foragers are those which are absorbed slowly, those Observations were reported on the effects which are encapsulated, and those which of fenoxycarb on honey bee brood after are of low toxicity to adults and high toxic- application on flowering fruit trees (Anony- ity to larvae. These authors were probably mous, 1989). The authors found that 8 to the first scientists to devise a method for 20 days following treatment, contaminated testing the effects of pesticides on bee brood. and malformed larvae and pupae were Their “Bee Larval Morphogenic Test”, removed outside the hive by workers. devised in 1974 and unpublished at that Ejected pupae had atrophied wings and time, allowed the assessment of the effects an abnormally flat and short abdomen. of pesticides on individual larvae inside the According to Van der Steen and de Ruijter hive. In the experimental colony the queen (1990), Gerig (1990) and Marletto et al. was confined for 24 h in a small cage made (1997), dead pupae showed typical eyes of queen excluder material, covering 500 with a white or reddish rim on their inner worker cells of an empty comb. Then the border. Gerig also observed undeveloped treatment solutions were applied to the food wings and shrunk abdomen in young work- in the bottom of each cell with a micro- ers. Fenoxycarb applied to hives by various syringe delivering 1 µl droplets per cell. A means resulted in ejection of malformed sample of 100 larvae living in several hori- pupae and prepupae, which began 1 to zontal adjacent rows was treated for each 2 weeks after treatment and continued 2 to dilution of pesticide and a control was 3 months. Besides, this IGR predisposed the included in each test. The number of sur- hives to attacks by the European foul brood viving treated and control larvae was and to sac brood (Marletto et al.,1992). assessed after cell capping. Before adult Teflubenzuron treatments in feeding and emergence, the experimental comb was contact tests affected brood development as removed from the hive and kept in an incu- early as larvae hatching from eggs (Gromisz bator at 35–37 °C. After emergence, sur- and Gromisz, 1996). vival and amorphogenic effects were Two of the partially purified fractions assessed. This brood toxicity test was extracted from neem seed (Azadirachta applied to different age groups of larvae. indica, Meliaceae), was tested by Rembold Barker and Taber (1977) reared bees from et al. (1980, 1982) on third instar larvae of egg to adult using nurse workers selected Impact of insect growth regulators on bees 531 from moved colonies containing no foragers. of the in-hive “Apis-larva-test” Wittmann They considered bees suitable for nursing et al. (1985) used an artificial diet they fed if they remained 2–3 h on a lone frame with to larvae of various stages. These larvae eggs. Eggs of this experimental unit should were grafted into 0.5 ml depressions of titra- be attended by approximately 250 nurse tion trays, kept in the incubator. Engels bees. The frame with eggs was introduced (1990) used the same in vitro method and into a wooden box holding two other frames recommended applying the test substance on either sides containing a pollen-sugar mixed with the artificial diet to third or mixture making a dough. A caged virgin fourth instar larvae because they were eas- queen was added to the workers and the ier to handle. He claimed this procedure pro- wooden box was held 10 days at 25 °C in vided highly reproducible LD 50s and LC the dark. Control or contaminated syrup was 50s. weighed and replaced every 5 days. After In his Apis-larvae-test Czoppelt (1990) 10 days the brood was separated from the reared larvae from the first to fifth instar in workers and incubated at 32 °C for 10 days the incubator at 35 °C, on a semi-artificial until emergence of adults which should be diet consisting of a mixture of royal jelly, examined for morphological characteristics. sugar solution and yeast extract (Rembold Naumann and Isman (1996) also used lar- and Lackner, 1981). Young larvae were vae reared by colonies after treatment. They transferred into artificial cells to feed on the applied the test substance on first and fourth mixture and one day later into new cells instars which were in demarcated areas of where chemicals could be applied after dilu- combs containing at the beginning only first tion in royal jelly at concentrations ranging –1 instar larvae. Series of 45 or 100 larvae were from 0 to 10 µg.ml . Larval weight gain treated with each concentration. Treatment was measured during the 48 h period of was applied by injecting 0.5 µl of test solu- intense growth between the fourth and the tion into the larval food at the bottom of fifth instar. This method is an in vitro stan- each cell. Six concentrations were used and dard test allowing the assessment of larvi- frames were examined for the survival of cidal effects of IGR and observations of lar- first instar larvae 6 and 10 days after treat- val growth until pupation. ment and for the fourth instar larvae, 10 days Van der Steen and de Ruijter (1990) after treatment and at adult emergence. The described a feeding test and a field test. In authors recommended to give the result as the first test the pesticide was dissolved in a LD 50 expressed in active substance weight sugar solution which was fed to colonies. per body weight. About 200 cells containing all stages from egg to pupa were marked by means of a Wittman (1982) described the first ver- transparent sheet. The larval stage was sion of a new “Apis-larvae-test” which divided into young and old larvae. The mor- enables the determination of the LC 50 of tality in each cell was checked once a week. agrochemicals and in particular IGRs. The In the field, test hives were placed in the principle was to dissolve the compound in crop at the beginning of flowering, 80 cm2 royal jelly and apply the food to groups of of brood cells were marked as in the feeding larvae of known age reared in strong test, and a trap was fixed in front of the hives colonies. Test areas comprised 50 larvae in to collect dead pupae. The test substance the second, third and fourth instar. Each was applied to the crop during its full bloom larva was fed 10 µl of the test mixture. Six and the marked cells were checked three different concentrations of the test substance times a week until brood cells were empty. were applied in 3 or 4 series of 100 larvae each and the LC 50 was calculated by Oomen et al. (1992) published the official regression analysis. In their in vitro version method of the EPPO (European Plant 532 J.-N. Tasei

Protection Organisation) for honey bee feed- It is more acceptable to use water or royal ing tests with IGR insecticides. The princi- jelly as a dilution medium, than acetone ple was to feed colonies with 1 l of a sugar (Atkins and Kellum, 1986) which may exert solution containing the test substance at the a variable negative influence on the treated concentration recommended for field use. larvae according to their stage. Barker and The trial compared a test substance with an Taber’s method was intermediate between IGR reference and a pure sugar solution. in-hive and in vitro test, since the authors Before the start of feeding, 300 cells per kept the experimental bees in the incubator colony were marked, 100 with eggs, 100 with and relied upon nurse bees to transfer the young larvae and 100 with old larvae. Brood test substance from contaminated syrup to development was checked once a week dur- larvae. Nurse bee interference was likely to ing 3 weeks following application. Dead be reduced compared to free flying colonies pupae were counted in a dead bee trap fitted since they have not access to food collected in front of the hives. Three replicates at least from outside. per product and per concentration should Van der Steen and de Ruijter (1990) and be performed. To describe brood develop- Oomen et al. (1992) recommended in-hive ment, it was recommended to mark each tests where feeders enabled permanent or cell on an overhead sheet, using a colour limited consumption of the test substance. code for each stage. Then, copies of the Their tests can not be conducted when there sheet should be used for further inspections is a high natural nectar flow because of and juvenile stages should be indicated by the possible risk of storage of the contami- their colour code. nated solution in empty frames. The reproducibility of their tests depends on the exter- 2.4. Discussion of the methods nal food resources which may influence the results due to a dilution effect of the test substance. Data recorded include only the The recommendations of the nine authors percentage of dead or malformed individu- or groups of authors mentioned above are als when a known concentration of formu- summarised in Table I. Toxicity tests com- lated IGR is diluted in the sugar solution. prise (a) in hive methods described by A similar method for testing the short term Barker and Taber (1977), Wittman (1982), and long term effect of a non-IGR insecti- Atkins and Kellum (1986), and Nauman and cide on bee brood was published by Webster Isman (1996), (b) the in vitro methods of and Peng (1989). Engels (1990) and Czoppelt (1990). For risk assessment, field methods were described The in vitro test described by Wittmann by Van der Steen and de Ruijter (1990) and et al. (1985); Czoppelt (1990) and Engels Oomen, de Ruijter and Van der Steen (1990) requires artificial cells and artificial (1992). diets where yeast extracts replace pollen. Their procedure is that which ensures the The in-hive tests required a unique appli- highest reproducibility of LD 50s or LC 50s. cation of test substance and implied perma- Other authors published slightly different nent feeding by nurse bees. Therefore an in vitro “Apis-larva-tests”, in particular influence of nursing can not be excluded Davis et al. (1988), who studied the effect of and may result in low reproducibility of LD non-IGR insecticides at adult sublethal lev- 50s. Another source of variability lies in the els on larval growth. syringe application into the brood cell where the contaminated food is both ingested by When considering the registration the larva and is in contact with its cuticule. scheme, it is recommended to conduct an This mixed action is also modified from one in vitro “Apis-larva-test” to estimate the stage to another due to size variation. toxicity of IGRs, then risk assessment of Impact of insect growth regulators on bees 533 3 colonies) 100 eggs × ( (dead bee trap) and Van der Steen Van and 2 (all ages) (all ages) 100 old la. capped cells capped cells 100 la. – 200 80 cm × solution LD 50 LD 50 100 la. 4 –1 × kg . bee mg –1 g . g µ emergence after treatment 3 weeks nurse bees with effect –1 bee . g colony queen caged virgin colony colony cells artificial cells colony colony colony µ acidified +yeast and KellumTaber and and Isman and de Ruijter and de Ruijter de Ruijter Methods for testing toxicity and hazards of IGRs to honey bees, according authors. Larvae. Other amorphogenicobservations malformed longevity effects none individualslarvae malformedmalformed of adults none malformed individuals pupae pupae pupae adults Table I. Table Authors Atkins Barker Nauman Wittman Engels Czoppelt der Steen Van der Steen Van Oomen a individualsper test concentration eggs brood cells brood cells 100 young la. Age of testlarvae (dor instar)Rearing 1–2 dconditions 3–4 d 5–6 dApplication syringe allof test solution flying free permanentApplicationmedium 25 °C. 10 d 3 frames syringe 1st Insecticide grade flying acetoneSurvival 4th syringe technicalassessment free pure or 2nd feeder water formulated flying syringe 3rd cell capping formulated artificial free syringe after cell capping 3rd 4th formulated water 10 d or after 35 °C. formulated permanent incubator 4th formulated all after incubator royal formulated flying field after 6 and artificial formulated free eggs 4, 7, d flying formulated feeder royal – jelly young + free eggs flying water old larvae young+ 1, 2, 3 diet permanent old larvae free flowers check jelly weekly water per week feeder 3 times 3 weeks application weekly check 1 l. Reportedparameter LD 50 concentration LD 50 LC 50 LC 50 LC 50 (%) dead (%) dead (%) dead Year (1986)Year Number (1977) (1996) 100 la.ª (1982) 148–1083 45–100 la. (1990) 4 (1990) (1990) (1990) (1992) 534 J.-N. Tasei registered concentrations should be per- Concentrations lower than 1.0 mg.kg–1 had formed with Oomen et al.’s field test (1992). no detectable effect on brood and adults. Tomic et al. (1985) feeding small colonies with contaminated syrup did not report dam- 2.5. Toxicity data on some IGRs age on sealed or unsealed brood with concentrations of 5 mg.kg–1 and 50 mg.kg–1. 2.5.1. Fenoxycarb Nation et al. (1986) used caged, small colonies for testing diflubenzuron in syrup Fenoxycarb dissolved in a sugar solution at 10 mg.kg–1 and did not see any reduction . –1 at the rate of 100 mg L and fed to colonies in the consumption of pollen and the quan- caused the death of almost all larvae and tity of brood reared. In the laboratory, those which reached the pupal stage were Czoppelt and Rembold (1981) reported that . –1 malformed. At 200 m L , all the larvae died topical application of 30 ng diflubenzuron (Van der Steen and de Ruijter, 1990). In resulted in delayed larval development and Czoppelt’s study (1991) all pupae died when dead prepupae. At doses of 50 and 100 ng, . –1 the food contained 0.5 mg kg fenoxycarb. larval survival was less than 26% and pre- . –1 This author found an LC 50 of 0.2 mg kg , pupal death rate more than 35%. In their µ . –1 i.e. 0.12 g larva while Nitsch and feeding test, concentrations ranging from Vorwohl (1992) established several LD 50 0.6 to 1.2 µg.mL–1 diflubenzuron resulted for each caste larvae fed at the 2 day stage in reduced larval survival and a total fail- and found for queens, workers and drones 7, ure of pupal stage. Larval growth was . –1 17 and 16 ng larva , respectively. impeded at high concentrations (1.0 and 1.2 µg.mL–1). The authors found a topical 2.5.2. Diflubenzuron LD 50 of 50 ng per larva and an oral LD 50 of 120 ng, which was contradictory to Chan- Barker and Taber (1977) found in labo- del and Gupta’s (1992) results who studied ratory conditions that diflubenzuron mixed the toxicity of diflubenzuron to third and with syrup at concentrations 0.59, 5.9 and fourth larval instars. The topical LD 50s 59 mg.kg–1 did not affect food consump- were 2.42 and 6.01 µg.larva–1, respectively, tion. Sealed brood was significantly reduced for A. mellifera, and 1.49 and 3.65 µg.larva–1, for the highest concentration 59 mg.kg–1, respectively, for A. cerana indica. LD 50 while lower concentrations had no effect on expressed per body weight showed a similar brood. No abnormalities were observed in susceptibility of both species to the com- newly emerged adults in any treatment. pound. Wittman (1982) found an LC 50 of Barker and Waller (1978) obtained similar 3.7 mg.kg–1 diflubenzuron. results in field conditions when studying hives exposed to 100 mg.kg–1 diflubenzuron in their water and to 60 mg.kg–1 difluben- 2.5.3. Azadirachtin zuron in their syrup. They reported a higher egg laying in treated hives but interpreted Rembold and Czoppelt (1981) studied this effect as a consequence of hatching fail- the effects of azadirachtin on honey bee lar- ure or a compensation for killed larvae. vae. They purified the compound from neem Using field colonies in an area of limited seeds and treated third instar larvae by top- bee forage, Stoner and Wilson (1982) fed ical application. Larvae were fed with a the bees for 12 weeks with sugar cake con- royal jelly and yeast mixture and reared in taminated with diflubenzuron and found the incubator. The lowest dose causing the concentration of 1.0 mg.kg–1 reduced observable effects was 0.25 µg.larva–1. sealed brood while 10 mg.kg–1 reduced both Naumann and Isman (1996) did not use sealed brood and the population of adults. seed extracts with unknown amounts of Impact of insect growth regulators on bees 535 azadirachtin but an emusifiable concentrate collected by honey bees, sorted apple and with an undiluted azadirachtin content of pear pellets, and found residues ranging 46000 mg.kg–1. Oral application of increas- from 1.9 to 18 mg.kg–1 fenoxycarb. After ing doses of azadirachtin on first and fourth tests using more or less contaminated pollens instar larvae resulted in larval ejection by fed to newly emerged workers, Gerig con- nurse bees in a dose dependent manner. The cluded that fenoxycarb sprayed on flowering LD 50 for both instars was 37 µg.g–1 body orchards was not likely to cause damage to weight and 61 µg.g–1 body weight. honey bees since the compound must be diluted through mingling of species and be 2.5.4. Bay Sir 8514® present only in insignificant amounts. Besides, during the years 1983 and 1984, Herbert et al. (1986) prepared a pollen sub- no damage on bees was reported. Although . –1 stitute containing 1, 10, 50, and 100 mg kg these results may have been due to the Bay Sir, which is a chitin synthesis inhibitor. lack of appropriate methods of investiga- At the two highest concentrations, bees fed tion, the author recommended this material for 12 weeks with the test diets were not should not be applied during foraging peri- able to rear brood despite the presence of ods. During the years 1983–1988 Gerig eggs in the frames. The eggs, removed from (1991) found damaged brood in colonies their cells and placed in free flying colonies foraging orchards sprayed with fenoxycarb were either ejected or eaten by workers. At . –1 . –1 the concentrations 1 and 10 mg.kg–1, bees at 200 g ha and 600 g ha . Maximum reared brood during 12 weeks and the residues in mixed pollens collected by bees . –1 . –1 amounts of larvae obtained was equal to ranged from 1.93 mg kg to 11.3 mg kg , those reared by the bees feeding on the con- respectively. Studies conducted with trol diet. Free flying colonies fed 3.5 l of a colonies visiting rape grown under cages sucrose solution at 100 mg.kg–1 Bay Sir, showed that sprays at the dose rates of –1 –1 contained 2 to 3 day old larvae and dead 200 g.ha and 600 g.ha were hazardous to –1 pupae one week following the treatment. brood, whereas 20 g.ha were tolerated. Eighteen days after the initial test feeding, The authors recommended that fenoxycarb the colonies had new healthy sealed brood. should not be applied to flowering fruit trees When the concentration was 150 mg.kg–1, and that flowering weeds should be elimi- the colonies examined 6 days after feeding, nated before treatment. Moreover, care contained no young larvae and 2 weeks after should be taken to avoid drifting to neigh- feeding all pupae were removed. Normal bouring attractive crops (Arzone et al., brood was observed only 3 weeks after the 1989). De Ruijter and van der Steen (1987) initial feeding. At 200 mg.kg–1, 11 days after sprayed a 12 ha orchard with 700 mg.kg–1 feeding, no brood was present and all pupae a.i. fenoxycarb and 200 l.ha–1 and observed died. the brood of four colonies in the treated crop and two in a control orchard. They marked eggs, young larvae, old larvae and capped 2.6. Risk assessment after field cells on overhead sheets 3 times a week and treatment with an IGR noted the first damage 10 days after appli- 2.6.1. Fenoxycarb cation. Contaminated larvae developed into pupae that died at the capped stage and all In experiments conducted from 1982 to brood which reached the second larval phase 1984, Gerig (1985) observed bee colonies within 5 days after spraying, died. Mal- foraging apple tree orchards which were formed individuals with white rimmed eyes treated with fenoxycarb (Insegar®) during could be found for 10 days and abnormal flowering period. He sampled pollen mortality of pupae was noted for 12 days. 536 J.-N. Tasei

2.6.2. Diflubenzuron placed in a 4 ha treated grove or in an adjacent control grove. Over a period of After treatment of apple orchards in full 7 months, the authors could not observe any bloom at 110, 200 and 400 g.ha–1 difluben- difference between the sealed brood of the zuron Emmet and Archer (1980) did not find treated hives and that of the check. Moreover any damage in adult bees or brood of no residues were detected in honey collected colonies foraging the treated flowers. Small after the sprays. Robinson and Johansen quantities of residues (0.11 and 0.39 mg.kg–1 (1978) found that spraying forests at 140 or . –1 diflubenzuron) were determined in honey 280 g ha diflubenzuron caused no dam- from these colonies. Bauml (1982) reported age to adults or brood in hives placed in the that diflubenzuron sprayed at 75 g.ha–1 on treated area and examined 10 and 46 days spruce forests did not affect colony weight after the spray. These dose rates resulted in . –1 and brood rearing in 3 hives placed in the residues in pollen of 1.2 and 6.2 mg kg treated area. Robinson (1979) studied the respectively (Davis et al., 1978). effects of a total of 8 spray treatments with diflubenzuron applied at one week interval 2.7. Registration of IGR treatments on cotton fields. Two doses were tested: on crops and risk assessment 140 g.ha–1 and 35 g.ha–1, and colonies were examined before and after each spaying. Adult bee activity and brood development According to a recent Directory (ACTA, assessed by marking individual cells, were 2001) four chitin inhibitors and four hor- not affected. Residues were not detected in mone mimics have been registered in France any wax, honey or pollen sample from for pest control in crops which can be for- colonies foraging the plot treated at the aged by bees (Tab. II). Target pests gener- lower rate but in the other plot treated at ally belong to Lepidoptera, Homoptera, 140 g.ha–1, pollen samples from the hives Coleoptera and Acarina, harmful to maïze, . –1 fruit trees, vine, vegetables and ornamental contained 0.06 to 0.19 mg kg difluben- . –1 . –1 zuron. As no residues could be detected in flowers. The highest dose ha is 168 g ha of tebufenozide to control Lepidoptera of other hive products, it was concluded that . –1 residues in pollen resulted from a direct con- pear trees, and the lowest is 25 g ha of tamination by the spray after the pellets pyriproxyfen to control white flies (Aleu- dropped in the trap. Because experimental rodes) on tomatoes. None of these com- plots were small, the author presumed for- pounds is systemic. agers dispersed on competing nectar and Laboratory and field data (Tab. III) show pollen sources and that their exposure to that even at the highest dose rate (400 g.ha–1) residues was limited. Therefore, further tests diflubenzuron sprays did not affect honey should be conducted in places where large bee brood. This is in accordance with the areas are treated in order to confirm these laboratory data. Residues of diflubenzuron preliminary results. Egger (1977) found that in pollen were maximum (6.2 mg.kg–1) after an aircraft application of diflubenzuron in spraying forests at 280 g.ha–1 while no a forest supplying a good honeydew flow residues could be detected after a treatment caused little damaged brood in bee colonies of cotton at 35 g.ha–1. In honey, maximum foraging the honeydew 11 days after spray- residues were 0.39 mg.kg–1 when dose rate ing. Twenty five days after treatment simi- ranged from 110 to 400 g.ha–1. If the maxi- lar losses in larvae, prepupae and pupae mum registration rate is 125 g.ha–1 (Tab. II), were still observed though at a lower extent. residues in bee forage are expected to be After 8 aerial applications at 350 g.ha–1 lower than 6.2 mg.kg–1 and 0.39 mg.kg–1 in diflubenzuron on Citrus, Schroeder et al. pollen and honey respectively. In semi-field (1980) examined the brood of hives either feeding tests with diflubenzuron, Wilson Impact of insect growth regulators on bees 537

Table II. Registration of 8 IGRs in France (ACTA, 2001).

Mode of action Active substance Crop Target pest Dose g.ha–1

Chitin inhibitor Diflubenzuron Maïze Sesamia 75–125 Fruit trees Cydia Forest trees Bombyx Chitin inhibitor Flufenoxuron Vine Mites 40–100 Fruit trees Cydia Chitin inhibitor Hexaflumuron Potato Leptinotarsa 35–125 Fruit trees Psylla, Cydia Chitin inhibitor Lufenuron Vine Lobesia 50–150 Ornamental Thrips plants Hormone mimic (Ecdysone) Tebufenozide Vine Lobesia 144–168 Fruit trees Cydia

Hormone mimic Buprofenozin Vegetables Aleurodes 132 (anti-Ecdysone) Hormone mimic(JH) Fenoxycarb Vine Lobesia 75–150 Fruit trees Cydia Saissetia Hormone mimic (JH) Pyriproxyfen Tomato Aleurodes 25

(1982) assessed a lowest observable effect damage caused to brood was in agreement concentration of 1 mg.kg–1 while the high- with the LC 50 of 0.2 mg.kg–1. The reduced est non observable effect concentration was dose of 140 g.ha–1 applied to orchards was 50 mg.kg–1 (Tomic et al., 1985). Such dis- also harmful to brood (de Ruijter and Van crepancies may be due to the different envi- der Steen, 1987). ronmental conditions of the two field tests and can be interpreted in the worst case as a The LD 50 and LC 50 of fenoxycarb and consequence of an exceptional low dilution diflubenzuron estimated for Apis mellifica of the contaminated food by external floral and B. terrestris larvae show a greater tol- resources. Therefore, diflubenzuron treat- erance of B. terrestris to fenoxycarb which ments can be regarded as generally safe for is harmless to this species when applied at dose rates up to 125 g.ha–1. However some doses up to 1200 g.h–1 (Tab. III). Therefore damage can not be excluded if environ- applying fenoxycarb at 140 g.h–1 will be mental conditions do not allow enough dilu- safe for bumble bee larvae but will cause tion of the contaminated food. damage to honey bee brood. Conversely B. terrestris young larvae are more suscep- In mixed pellets from bees foraging in tible to diflubenzuron than honey bee lar- an area where orchards were treated at vae. This explains losses in young brood of 200 g.ha–1 and 600 g.ha–1 fenoxycarb, bumble bee colonies exposed to crops residues reached 1.93 mg.kg–1 and sprayed at 300 g.ha–1 which is a dose rate 11 mg.kg–1 respectively (Gerig, 1991). The safe for honey bees. 538 J.-N. Tasei ) –1 kg . nd Van der Steen, 1987 (f), Emmet and Van nd (mg tenkord and Drescher, 1996 (l). tenkord and Drescher, ) in pollen or honey –1 ha . (l) 300 (l) 62 [day 1] (l) damaged brood (l) (l) 1200 (l) 217 [day 1] (l) none (l) (i) 350 (i) 0 [ho.] (i) none (i) Citrus cotton (h) 35–140 (h) 0– 0.19 [po.] (h) none (h) forest (j, k) 140–280 (j, k) 1.2–6.2 [po.] (j, k) none (j, k) Phacelia Phacelia ) (g a.i. –1 kg . 137.8 2 [day 7] (l) [6 day la.] (l) larva) (mg –1 bee . (ng 7.7 [1 day la.] (l) 1.18 5112 [6 day la.] (l) [1 day la.] (l) 2420 [3 day la.] (c) 3.7 (d) orchard (g) (g) 110–400 [ho.*] (g) 0.11–0.39 none (g) > 650 [1 day la.] (l) >100 > 3710 [6 day la.] (l) [1–6 day la.] (l) 7.5 [day 7] (l) 16 [2 day male la.] (a) 7 [2 day queen la.*] (a) orchard (e) 200–600 (e) [po.*] (e) 1.9–11 damaged brood (e) 17 [2 day worker la.] (a) 0.2 (a) orchard (f) 140 (f) damaged brood (f) Apis Apis Bombus Bombus terrestris mellifera mellifera terrestris Oral toxicity of two IGRs to honey bee and bumble larvae risk assessment after application test crops. Diflubenzuron Diflubenzuron Fenoxycarb Archer, 1980 (g), Robinson, 1979 (h), Shroeder et al., (i), Robinson and Johansen, 1978 (j), Davis (1978) (k), Gre Archer, Table III. Table IGR Species LD 50 LC 50 Crop* la. = larva, po.= pollen, ho. honey. Dose 1982 (d), Gerig, 1991 (e), De Ruijter a Wittman, 1992 (a), Czoppelt, 1991 (b), Chandel and Gupta, (c), Vorwohl, Nitsch and Residues Effects Fenoxycarb Impact of insect growth regulators on bees 539

3. EFFECTS OF IGRS ON NON-APIS 3.1.2. Cage and greenhouse tests BEES The same authors modified a cage test 3.1. Methods for testing IGRs method previously devised for conventional on bumblebees compounds testing (Gretenkord and 3.1.1. Laboratory tests Drescher, 1993). In their new procedure (Gretenkord and Drescher, 1996), colonies De Wael et al. (1995) comparing the of 50–70 workers were placed in cages effects of several IGRs on B. terrestris, used (3 × 4 × 2 m) covering Phacelia plots. When queen-right colonies, each containing 30 to about 10 foragers could be observed, 50 workers. They were kept in the dark at colonies were moved to the laboratory and 29 °C and fed daily with 50% sugar solu- foragers left in the cage. The colonies were tion and pollen collected from honey bees. reduced to 5 workers with the queen and a Test IGRs were administrated in syrup for defined amount of brood of all stages. These 24 h, then the amount ingested was deter- colonies were then reintroduced into their mined by weight loss. Dead adults and lar- cage and foragers could go back to their nest vae were removed and counted daily and thus forming standardised units which could photographs of the nest were taken daily for develop with flower resources and syrup 5 weeks. As photographic records started a but without additional pollen feeding. This week before treatment it was possible to procedure allowed an accurate assessment of estimate the effect of IGRs on the develop- the effects of contamination of pollen by an ment of all brood stages, from egg to pupa. IGR on brood. The test substance should be The authors used only one colony per treat- sprayed on Phacelia the day following ment and a control colony fed with sucrose colony reduction. After a cage period of solution. 2–3 weeks, colonies should be reared in the As bumble bee larvae can not be isolated laboratory for 2 weeks until adult emer- successfully in vitro as it is done with honey gence. bees, Gretenkord and Drescher (1996) Tornier (1999) described a test method devised a larval test adapted to bumble bees. adapted to the greenhouse. He used 4 queen- They prepared “test groups” by removing right colonies per treatment, with 30 work- egg cells from colonies until hatching, and ers and similar amounts of brood. Pho- equalising the number of larvae in all the tographs of each brood clump were taken groups. They obtained standard cells with before and after application. All adults were 10 young larvae which were each kept at marked before the colonies were introduced × × 29 °C in rearing boxes (12.5 7 5 cm) into the greenhouse and a trap was fixed at with 3 nurse workers. Test groups were fed the hive entrance for collecting dead adults syrup and pollen dough until pupation. The and larvae. Three parameters were recorded workers were then removed until adults during the test period: the food consump- emerged. For testing IGRs it was recom- tion, the weight of colonies and the wing mended that 1,4 and 6 day old larvae be fed size of adults. separately for 24 h with pollen dough or syrup in which the test substance was dis- Thompson and Barrett (1999) also tested solved. Trials, comprising 3 replicates of IGRs in a greenhouse. They used 5 × 3 m each treatment and control, should start with compartments containing tomato plants and the recommended concentration for field a single queen-right colony with 100–200 use. If negative effects are observed, the tri- workers. Tomato was treated at 10 days als should continue with lower concentra- intervals. Colonies were fed additional tions. From the mortality records and food pollen which was also treated at the same consumption measurements per larva, LC rate as plants. Each IGR treatment and the 50 and LD 50 should be calculated. control spray with water were repeated in 540 J.-N. Tasei

2 compartments. Diflubenzuron at 0.03% zuron at 150 mg.kg–1g a.i. De Wael et al. a.i., was used as a “positive” control. From (1995) found that the last treatment resulted day 3 to day 23, the authors monitored the in a higher larval mortality than the control number of dead adults, dead larvae, foraging and the two other IGRs. With teflubenzuron, bees and the general appearance of the larvae died and were all removed by work- colonies. ers during the week following treatment. In addition, the queen continued to lay eggs 3.1.3. Field tests but egg development was arrested for 5 weeks. Fenoxycarb and pyriproxyfen did Schäfer and Mühlen (1996) conducted a not cause any damage to brood. field trial to test the effects of an IGR on B. terrestris by placing 6 colonies of approx- Testing two IGRs on one and 6 day old imately 50 workers each in a 2400 m2 larvae, Gretenkord and Drescher (1996) Phacelia field. Before and after treatment reported an LC 50 higher than 100 mg.kg–1 the following parameters were determined: for fenoxycarb for both ages, while that for the density of foragers the flight activity at diflubenzuron was 1.18 and 137.79 mg.kg–1, the hive entrance, the origin and amount of respectively. The LD 50 of fenoxycarb was the pollen collected, the number of work- higher than 650 and 3710 ng.bee–1 for one ers, the mortality of larvae, and the number and 6 day old larvae respectively. For of egg cells, larvae and cocoons. There was diflubenzuron, the LD 50 was 7.7 and no observation beyond the fourth day after 5112.0 ng.bee–1 respectively. In a cage test treatment. The authors concluded their where both compounds were sprayed procedure did not allow a correct interpre- at a normal and double dose (i.e. 600 and tation of the data and that standardisation 1200 g.ha–1 fenoxycarb and 300 and of such method was not easy due to the 600 g.ha–1 diflubenzuron) residues in pollen unpredictable development of bumble bee ranged from 217 mg.kg–1 the first day to colonies. 7.5 mg.kg–1 fenoxycarb the seventh day, and from 62 mg.kg–1 diflubenzuron the first day to 2 mg.kg–1 the seventh day. In all of 3.2. Effects of IGRs on bumble bees the five test cages with fenoxycarb, no brood 3.2.1. Effects on adults damage and no malformed adults were detected while in the eight cages with In a greenhouse test diflubenzuron diflubenzuron all the larvae died, except old sprayed on tomatoes at the concentration ones, within the 2 days following treatment. 0.03% a.i. proved harmless to adults (Thomp- During 3 weeks no brood was reared though son and Barrett, 1999). Fenoxycarb applied egg laying continued, which suggested a at 1200 g.ha–1 to Phacelia grown in cages, negative effect on the queen’s ovaries. More- did not cause any trouble to caged adults of over the authors observed malformed the test colonies (Gretenkord and Drescher, cocoons which were spherical with abnor- 1993). In a Phacelia field, Schäfer and mal brown dots on the surface. Fenoxycarb Mühlen (1996) found that an application of was safe for caged bumble bees though the triflumuron at 800 g.ha–1 did not influence concentration in pollen on the first day was flight activity of experimental colonies. twice the concentration fed to larvae in the laboratory test. Diflubenzuron residues in 3.2.2. Effects on brood pollen even after the seventh day following application were within the range of the When pyriproxyfen at 20 mg.kg–1 a.i. was LC 50 values for one to 4 day old larvae, fed to colonies for 24 h and compared to but residues on the first day did not reach fenoxycarb at 100 mg.kg–1 a.i. and tefluben- the LC 50 for 6 day old larvae which was in Impact of insect growth regulators on bees 541 accordance with the survival of the old and typical malformations have been brood in the caged colonies. observed in colonies after some IGR applications. These troubles were due to the properties of these products, which all interfere 3.3. Effects of IGRs on other with embryo development and moulting wild pollinators process and which contaminate food resources collected (nectar and pollen) and Kawashima (1989) tested 3 chitin syn- stored in the colony by foragers. As a con- thesis inhibitors, diflubenzuron, chorflu- sequence, the effects of intoxication by IGRs azuron and teflubenzuron on the orchard are always delayed. They also cover a longer pollinator, the solitary bee Osmia cornifrons, period than in the case of non-IGR insecti- by spraying bees directly and confining them cides. These characteristics justify new test- with treated apple leaves. The three IGRs ing of laboratory and field methods for tox- affected neither the survival of adults nor icity and risk assessment. cocoon formation. Presumably the effects on progeny would have been different if Comparison of data from tests with A. O. cornifrons had the chance to forage on mellifera and B. terrestris proved these two treated flowers and provision their cells with species were affected in opposite ways by contaminated pollen. Narita (1988) did not the application of the same IGR. Therefore, find any damage in adult O. cornifrons risk assessment for one test species can not released on trees 3 days after spraying the be extended to others without appropriate plants with solutions at 25 and 50 mg.kg–1 of additional investigations. chlorfluazuron and flufenoxuron respectively. De Oliveira Campos (1978) tested the Résumé – Effets des régulateurs de crois- juvenile hormone analogue Altozar® on lar- sance des insectes sur les abeilles domes- vae of the social bee Melipona quadrifas- tiques et les abeilles sauvages. Les régu- ciata. Topical applications of 18.3 µg.larva–1 lateurs de croissance des insectes ou RCI to larvae spinning their cocoon promoted ou encore insecticides de la troisième géné- ration agissent comme des ecdysones, des the transformation of female worker larvae hormones juvéniles ou des inhibiteurs de into queen pupae and male pupae into chitine. Ce sont des composés synthétiques female adults. ou des substances naturelles comme l’Aza- dirachtin, qui interfèrent dans l’équilibre naturel des hormones de mue. De ce fait, 4. CONCLUSION ils risquent peu de produire des dommages chez les adultes, par contre ils peuvent Insect growth regulators, used for pest engendrer des troubles dans le couvain. control management will cause no damage L’application d’analogues d’hormones à des to adult honey bees and probably other adult abeilles adultes a modifié le comportement pollinators, and can be considered as safer de butinage, la production de phéromone for foragers than second generation insecti- d’alarme, l’hémolymphe, le gain de poids, la cides. Oral and contact laboratory tests and synthèse de la vitellogénine, la longévité et field observations or trials proved adults tol- a inhibé les glandes hypopharyngiennes erated these compounds, in particular at the (Redfern et Knox, 1974 ; Jaycox et al., registered doses. 1974 ; Rutz et al., 1974 ; Gerig, 1975 ; Atkins This safety for pollinators is only appar- et al., 1976 ; Robinson, 1985 ; Usha et ent, since serious damage to brood has been Kadasamy, 1986 ; Ceparano et Job, 1989 ; reported in honey bees and bumble bees. Gupta et Chandel, 1995 ; Pinto et al., 2000). Abnormal mortality in eggs, larvae or pupae Chez les larves l’intoxication expérimentale 542 J.-N. Tasei par contact ou par ingestion provoque la mort de sévères pertes chez les larves et les œufs des œufs, des larves, ou des malformations à la dose de 300 g.ha–1. Ces résultats on été des larves, des nymphes ou des adultes. Les corroborés par des tests sur larves en labo- nymphes ont alors des yeux atypiques portant ratoire qui ont permis d’évaluer les DL 50 un cercle coloré. Il y a aussi possibilité d’in- des 2 insecticides sur plusieurs stades lar- hibition des glandes hypopharyngiennes et vaires (Gretenkord et Drescher, 1996). d’apparition de caractères royaux. Larves et nymphes intoxiquées sont éjectées des cel- régulateur croissance insecte / Apis lules par les ouvrières (Zdarek et Haragsim, mellifera / Bombus / abeille sauvage / 1974 ; Beetsma et Ten Houten, 1975 ; Gerig, évaluation risque 1975 ; Hrdy et Skrobal, 1976 ; Hussein et Abdel-Aal, 1978 ; Rembold et al., 1980, 1982 ; El Din et al., 1990 ; Gerig, 1990 ; Van Zusammenfassung – Auswirkungen von der Steen et de Ruijter, 1990 ; Marletto et al., Wachstumsregulatoren der Insekten auf 1992 ; Chandel et Gupta, 1992 ; Nitsch et al., Honigbienen und Wildbienen. Die Wachs- 1994 ; Gromisz et Gromisz, 1996). tumsregulatoren der Insekten, auch als IGR oder Insektizide der 3. Generation bezeich- Plusieurs méthodes ont été publiées pour net, wirken wie Ecdyson, Juvenilhormon tester la toxicité des RCI sur les larves oder greifen in die Chitinsynthese ein. Es (Barker et Taber, 1977 ; Wittman, 1982 ; handelt sich dabei sowohl um synthetische Atkins et Kellum, 1986 ; Engels, 1990 ; Verbindungen als auch um natürliche Sub- Czoppelt, 1990 ; Van der Steen et De Ruijter, stanzen wie z.B. Azadirachtin (vom Neem 1990 ; Naumann et Isman, 1996). La plu- Baum), die in das Gleichgewicht des Hor- part des données proviennent d’expérimen- monhaushalts während des Wachstums ein- tations sur deux produits le fenoxycarb et greifen. Dadurch sind sie bei adulten Tieren le diflubenzuron, testés sur l’abeille et les weniger gefährlich, im Gegensatz dazu kön- bourdons Bombus terrestris. Chez l’abeille, nen sie bei der Brut Schäden hervorrufen. selon les auteurs les concentrations dans la Die Applikation der Hormonanaloge bei nourriture, sans effets par ingestion, varient adulten Bienen hat Einfluss auf das Sam- de 1 à 50 mg.kg–1 de diflubenzuron (Barker melverhalten, die Produktion des Alarm- et Taber, 1977 ; Barker et Waller, 1978 ; pheromons, die Hämolymphe, die Gewichts- Czoppelt et Rembold, 1981 ; Wittman, zunahme, die Vitellogeninsynthese, die 1982 ; Stoner et Wilson, 1982 ; Tomic et al., Lebenserwartung und hemmt die Entwick- 1985 ; Nation et al., 1986 ; Chandel et Gupta, lung der Hypopharynxdrüsen (Redfern 1992). Le fenoxycarbe provoque la mort de und Knox, 1974; Jaycox et al., 1974; Rutz 100 % des nymphes à 1a concentration de et al., 1974; Gerig, 1975, Atkins et al., 1976; 0,5 mg.kg–1 de substance active (Czoppelt, Robinson, 1985; Usha und Kadasamy, 1991). En champ, aux doses expérimentales 1986; Ceparano und Job 1989; Gupta und (35 à 400 g.ha–1) le diflubenzuron est estimé Chandel, 1995; Pinto et al., 2000). Bei der sans danger pour les colonies par la plupart Brut führt die experimentelle Giftapplika- des auteurs, au contraire du fenoxycarbe qui tion zum Absterben der Eier und Larven cause des dommages au couvain à 140 g.ha–1 oder zu Verkrüppelungen der Larven, Pup- (de Ruijter et Egger 1977 ; Robinson et pen oder erwachsenen Tiere. Die Puppen Johansen, 1978 ; Robinson, 1979 ; Emmet et haben demzufolge atypische Augen, die Archer, 1980 ; Schroeder et al., 1980 ; einen gefärbten Kreis aufweisen. Auβerdem Bauml, 1982 ; Van der Steen, 1987 ; Arzone entstehen Hemmungen der Hypopharynx- et al., 1989). Chez B. terrestris on a trouvé drüsen und es kommt zur Ausbildung von que les traitements à 1200 g.ha–1 de fenoxy- Königinnenmerkmalen. Vergiftete Larven carbe n’avaient pas d’effet négatif sur les und Puppen werden von den Arbeiterinnen larves alors que le diflubenzuron entraînait aus den Zellen entfernt (Zdarek und Impact of insect growth regulators on bees 543

Haragsim, 1974; Beetsma und Ten Houten, auf verschiedene Larvenstadien erfolgte 1975; Gerig, 1975; Hrdy und Skrobal, 1976; (Gretenkord und Drescher, 1996). Hussein und Abdel-Aal, 1978; Rembold et al., 1980, 1982; El Din et al., 1990; Gerig, Insektenwachstumsregulatoren / Honig- 1990; Van der Steen und de Ruijter, 1990; bienen / Wildbienen / Toxizität / Risikoab- Marletto et al., 1992; Chandel und Gupta, schätzung 1992; Nitsch et al., 1994; Gromisz und Gromisz, 1996). Es wurden verschiedene Methoden pub- REFERENCES liziert um die Toxizität der IGRs auf Lar- A.C.T.A. (2001) Index phytosanitaire ACTA, Associ- ven zu testen (Barker und Taber, 1977; ation de Coordination Technique Agricole, 149 rue Wittman, 1982; Atkins und Kellum, 1986; de Bercy, Paris. Engels, 1990; Czoppelt, 1990; Van der Steen Anonymous (1989) L’Insegar et les abeilles. Nouvelles und De Ruijter, 1990; Naumann und Isman, prescriptions, Rev. Suisse Vitic. Hortic. 21, 64–65. Atkins E.L., Anderson L.D., Kellum D., Neuman K.W. 1996). Die meisten der Arbeiten weisen Ver- (1976) Protecting honey bees from pesticides. suche mit 2 Produkten, Fenoxycarb und Leaflet 2883, University of California. Diflubenzuron auf, die bei Honigbienen und Atkins E.L., Kellum D. (1986) Comparative mor- Hummeln Bombus terrestris geprüft wur- phogenic and toxicity studies on the effect of pes- den. Solange die Konzentrationen in der ticides on honeybe brood, J. Apic. Res. 25, 242–255. Arzone A., Dolci M., Marletto M. (1989) Detection of Nahrung bei adulten Honigbienen zwischen fenoxycarb on mulberry leaves, Apic. Mod. 80, –1 1 bis 50 mg.kg betrugen, ergaben sich 149–152 (in Italian). nach den Autoren keine Wirkung (Barker Barker R.J., Taber S. (1977) Effects of diflubenzuron und Taber, 1977; Barker und Waller, 1978; fed to caged honey bees, Environ. Entomol. 6, Czoppelt und Rembold, 1981; Wittman, 167–168. Barker R.J., Waller G.D. (1978) Effects of difluben- 1982; Stoner und Wilson, 1982; Tomic zuron wettable powder on caged honey bee et al., 1985; Nation et al., 1986; Chandel colonies, Environ. Entomol. 78, 535–574. und Gupta, 1992). Fenoxycarb bewirkt bei Bauml W. (1982) Field studies on the effects of Dim- einer Konzentration von 0,5 mg.kg–1 der ilin on the honey bee, Anz. Schaedlingskd. Pflanzen. aktiven Substanz zu 100 % den Tod von Umweltsch. 55, 97–101 (in German). Beetsma J., Ten Houten A. (1975) Effects of juvenile Puppen (Czoppelt, 1991). Bei Behandlungen hormone analogues in the food of honey bee im Feld mit einer Dosierung von 35 bis colonies (Apis mellifera L.), Z. Angew. Entomol. 77, 400 g.ha–1 gilt das Diflubenzuron bei den 292–300. meisten Autoren als für Bienenvölker unge- Ceparano G., Job S. (1996) Flucycloxuron: a new aca- fährlich, im Gegensatz zu Fenoxycarb, das ricide, Inf. Fitopatol. 46, 32–34. Chandel R.S., Gupta P.R. (1992) Toxicity of difluben- Schäden bei der Brut bei einer Konzentra- zuron and penfluron to immature stages of Apis –1 tion 140 g.ha erzeugt (de Ruijter und cerana indica F. and Apis mellifera L., Apidologie Egger, 1977; Robinson und Johansen, 1978; 23, 465–473. Robinson, 1979; Emmet und Archer, 1980; Czoppelt C. (1990) Effect of insect growth inhibitors and pesticides on honey bee larvae (Apis mellifera Schroeder et al., 1980; Bauml, 1982; Van L.) in contact poison and feeding tests in vitro, Pro- der Steen, 1987; Arzone et al., 1989). Bei ceedings of the 4th International Symposium on B. terrestris fand man, dass Behandlungen Harmonization of Methods for Testing the Toxicity mit 1200 g.ha–1 Fenoxycarb keinen nega- of Pesticides to Bees, May 15–18, Rez near Prague, Research Institute of Apiculture, Dol, Czechoslo- tiven Effekt auf die Larven hat, während vaquia 1990, pp. 76–83. Diflubenzuron bei einer Konzentration von Czoppelt C. (1991) Evaluation of toxicity of the 300 g.ha–1 schwere Verluste bei Eiern und juvenoid phenoxycarb to honeybee larvae reared Larven zur Folge hatte. Diese Ergebnisse in vitro, Apidologie 22, 457–459. wurden durch Untersuchungen mit Larven Czoppelt C., Rembold H. (1981) Effects of Dimilin on the growth of honey bee larvae reared in an incu- im Labor unterstützt, mit der die Bestim- bator, Mitt. Dtsch. Ges. Allg. Angew. Entomol. 3, mung der LD 50 dieser beiden Insektizide 1–13; 191–195 (in German). 544 J.-N. Tasei

Davis E.J., Robinson W.S., Roush C.F., Akre R.D., Gretenkord C., Drescher W. (1993) Development of a Johansen C.A., Turner W.J. (1978) Impact of chem- cage test method for the evaluation of pesticide ical control applications for the Douglas-fir Tus- hazards to the bumble bee Bombus terrestris L., sock moth on beneficial insects including biologi- Proceedings of the 5th International Symposium. on cal studies of bees, yellowjackets, and flesh flies, the Hazards of Pesticides to Bees, Oct. 26–28, Melanderia 30, 1–7. Wageningen, The Netherlands, Appendix 20. Davis A.R. (1989) The study of insecticide poisoning Gromisz Z., Gromisz M. (1996) Harmful effects of the of honeybee brood, Bee World 70, 163–174. Nomolt formulation on honey bees, Pszcz. Zesz. Davis A.R., Solomon K.R., Shuel R.W. (2000) Labo- Nauk. 40, 175–183 (in Polish). ratory studies of honeybee larval growth and devel- Grosscurt A.C. (1978) Diflubenzuron: Some aspects opment as affected by systemic insecticides at adult- of its ovicidal and larvicidal mode of action and an sublethal levels, J. Apic. Res. 27, 146–161. evaluation of its practical possibilities, Pestic. Sci. Egger A. (1977) The side effect of Dimilin on honey- 9, 373–386. bees, an investigation in relation to aircraft appli- Gupta P.R., Chandel R.S. (1995) Effects of difluben- cation in forests. Zentralblatt für Gesamte Forstwe- zuron and penfluron on workers of Apis cerana sen 94, 5–72 (in German). indica F. and Apis mellifera L., Apidologie 26, El-Din H.A., El-Samni M.A., Attia M.B. (1990) Latent 3–10. effects of some insect growth regulators on the Herbert E.W., Argauer R.J., Shimanuki H. (1986) The body weight of workers of the honey bee Apis mel- effect of an insect chitin synthesis inhibitor on lifera L. var. carnica (Hym. Apidae), Anz. honey bees, Apidologie 17, 73–76. Schaedlingskd. Pflanzen. Umweltsch. 63, 67–68. Hrdy I., Skrobal D. (1976) Effect of two juvenoids on Emmett B.J., Archer B.M. (1980) The toxicity of honey bee colonies (Apis mellifera L.), Apidolo- diflubenzuron to honey bee (Apis mellifera L.) gie 7, 39–47 (in German). colonies in apple orchards, Plant. Pathol. 29, 177–183. Hussein M.H., Abdel-Aal Y.A.I. (1978) Effect of ZR 515 on honey bee, Apis mellifera L., Z. Angew. Engels W. (1990) Testing of insect growth regulators Zool. 87, 109–111. and of varroacides by Apis-larvae test, Proceed- ings of the 4th International Symposium on Har- Kawashima K. (1989) Effect of chitin inhibitors on the monization of Methods for Testing the Toxicity of horn faced, Osmia cornifrons Radoskowski, J. Soc. Pesticides to Bees, May 15–18, Rez near Prague, Plant Prot. North Jpn. 40, 171–173 (in Japanese). Research Institute of Apiculture, Dol, Czechoslo- Jaycox E.R., Skowronek W., Guynn G. (1974) Behav- vaquia 1990, pp. 84–87. ioral changes in worker honey bees (Apis mellif- George D.A., Rincker C.M. (1985) Results and con- era) induced by injections of juvenile hormone clusions of using pesticides with the alfalfa leaf- mimic, Ann. Entomol. Soc. Am. 67, 529–534. cutting bee in the production of alfalfa seed, Marletto F., Arzone A., Dolci M. (1992) Action of J. Agric. Entomol. 2, 93–97. fenoxycarb on honey bee brood, Apic. Mod. 83, Gerig L. (1975) The effects of juvenile hormone ana- 209–218 (in Italian). logues on summer bees (Apis mellifera L.) in the Marletto F., Arzone A., Dolci M. (1997) The honey field and laboratory, Schweiz. Landwirtsch. Forsch. bee as a biological indicator of fenoxycarb pollution, 14, 355–370 (in German). Apic. Mod. 88, 107–110 (in Italian). Gerig L. (1985) Treatment of blooming orchard trees Narita T. (1988) Control of the tortrix (Archips fus- with fenoxycarb, Proceedings of the 3rd Interna- cocupreana Walsingham) on apple trees by chitin tional Symposium on Harmonization of Methods synthesis inhibitors, Kitanippon Byogaichu for Testing the Toxicity of Pesticides to Bees, 18–21 Kenkyukaiko 39, 242–245 (in Japanese). March 1985, Rothamsted, UK, Appendix 16. Nation J.L., Robinson F.A., Yu S.J., Bolten A.B. (1986) Gerig L. (1990) News about the use of Insegar (fenoxy- Influence upon honey bees of exposure to very low carb) in Switzerland, Proceedings of the 4th Inter- levels of selected insecticides in their diet, J. Apic. national Symposium on Harmonization of Meth- Res. 25, 170–177. ods for Testing the Toxicity of Pesticides to Bees, May 15–18, Rez near Prague, Research Insti- Naumann K., Isman M.B. (1996) Toxicity of a neem tute of Apiculture, Dol, Czechoslovaquia 1990, (Azadirachta indica A. Juss) insecticide to larval pp. 74–75. honey bees, Am. Bee J. 136, 518–520. Gerig L. (1991) Importance de l’Insegar pour l’api- Nitsch C., Vorwohl G. (1992) The effect of Insegar on culture et l’arboriculture, J. Suisse Apic. 88, the honey bee, Allg. Dtsch. Imkerztg. 26, 6–8 (in 235–238. German). Gretenkord C., Drescher W. (1996) Laboratory and Nitsch C., Horn H., Vorwohl G. (1994) Effect of Dim- cage test methods for the evaluation of the effects ilin on queen rearing, Dtsch. Bienen. J. 2, 16–19 of growth regulators (Insegar, Dimilin) on the brood (in German). of Bombus terrestris L., Proceedings of the 6th Oliveira Campos L.A. de (1978) Sex determination in International Symposium on Hazards of Pesticides bees. VI. Effect of juvenile hormone analogue in to Bees, Sept. 17–19, 1996, Braunschweig, Ger- males and females of Melipona quadrifasciata many, Appendix 29. (Apidae), J. Kans. Entomol. Soc. 51, 228–234. Impact of insect growth regulators on bees 545

Oomen P.A., Ruijter A., Steen J. van der (1992) Method ing the Toxicity of Pesticides to Bees, May 15–18, for honey bee brood feeding tests with insect Rez near Prague, Research Institute of Apiculture, growth-regulating insecticides, EPPO Bull. 22, Dol, Czechoslovaquia 1990, pp. 88–91. 613–616. Stoner A., Wilson W.T. (1982) Diflubenzuron (Dim- Pinto L.Z., Bitondi M.M., Simoes Z.L.P. (2000) Inhi- ilin): effect of long-term feeding of low doses in bition of vitellogenin synthesis in Apis mellifera sugar cake or sucrose syrup on honey bees in stan- workers by juvenile hormone analogue pyriprox- dard-size field colonies, Am. Bee J. 122, 579–582. yfen, J. Insect Physiol. 46, 153–160. Tasei J.N., Carré C. (1987) Effects of the pyrethroid Redfern R.E., Knox D.A. (1974) Adult honey bee rel- insecticide WL85871 and phosalone on adults and ative toxicity of two juvenile hormone analogues, progeny of the leaf-cutting bee Megachile rotun- Am. Bee J. 456–458. data F., pollinator of lucerne, Pestic. Sci. 21, Rembold H., Sharma G.K., Czoppelt C., Schmutterer 119–128. H. (1980) Evidence of growth disruption in insects Tasei J.N., Carré C., Moscatelli B., Grondeau C. (1988) without feeding inhibition by neem seed fractions, Recherche de la DL50 de la deltaméthrine (Decis) Z. Pflanzenkd. Pflanzenschutz. 87, 290–297. chez : Megachile rotundata F., abeille pollinisa- Rembold H., Czoppelt C. (1981) Evaluation of insect trice de la luzerne et des effets de doses infralé- growth regulators from Azadirachta indica (neem) thales sur les adultes et les larves, Apidologie 19, using rearing tests on honey bee larvae, Mitt. Dtsch. 291–306. Ges. Allg. Angew. Entomol. 3, 196–198 (in Ger- Thompson H.M., Barrett K.A. (1990) Assessing the man). effects of glasshouse application of a novel insect Rembold H., Lackner B. (1981) Rearing of honeybee growth regulator on bumble bee colonies, in: larvae in vitro: effect of yeast extract on queen dif- Belzunces L.P., Pélissier C., Lewis G.B. (Eds.), ferentiation, J. Apic. Res. 20, 165–171. Hazards of pesticides to bees, Proc. 7th Int. Symp. Rembold H., Sharma G.K., Czoppelt C., Schmutterer on Hazards of Pesticides to Bees 7–9 Sept 1999, H. (1982) Azadirachtin: a potent insect growth reg- Avignon, France, INRA Paris, pp. 227–228. ulator of plant origin, Z. Angew. Entomol. 93, Tomic N., Popescovic D., Dujin N. (1985) Some larval 12–17. and post-larval effects of diflubenzuron on honey Robinson F.A. (1979) The effects of repeated spray bee (Apis mellifera L.), Proceedings of the 29th applications of Dimilin W-25 on honey bee colonies International Congress of Apiculture, Budapest, in cotton fields, Am. Bee J. 119, 193–194; 197. 1983, Apimondia Publishing House, Bucharest, Robinson F.A. (1985) Effects of a juvenile hormone Romania, pp. 275–278. analogue on bee foraging behaviour and alarm Tornier I. (2001) Side effects of an Insect growth reg- pheromone production, J. Insect Physiol. 31, ulator on bumble bees and honey bees, in: 277–282. Belzunces L.P., Pélissier C., Lewis G.B. (Eds.), Robinson F.A., Johansen C.A. (1978) Effects of control Hazards of pesticides to bees, Proc. 7th Int. Symp. chemicals for douglas fir tussock moth on forest on Hazards of Pesticides to Bees, 7–9 Sept. 1999, pollination, Melanderia 30, 9–56. Avignon, INRA Editions, Paris, les Colloques n° 98, 299. Ruijter A. de, Steen J. van der (1987) A field study on the effect on honey bee brood of Insegar (fenoxy- Usha K., Kandasamy C. (1986) Comparative toxicity carb) applied on blooming apple orchard, Api- of certain pesticides to the honey bee Apis indica F., dologie 18, 355–357. Pest. Manage. 1, 31–35. Rutz W., Gerig L., Wille H., Lüscher M. (1974) A Wael L., de Greef M., de Laere O. Van (1995) Toxicity bioassay for juvenile hormone (JH) effects of insect of pyriproxyfen and fenoxycarb to bumble bee growth regulators (IGR) on adult worker honey brood using a new method for testing insect growth bees, Preliminary report, Mitt. Schweiz. Entomol. regulators, J. Apic. Res. 34, 3–8. Ges. 47, 307–313. Webster T.C., Peng Y.S. (1989) Short term and long- Schäfer H., Mühlen W. (1996) First experiences to test term effects of methamidophos on brood rearing side-effects of Alsystin on bumble bees (Bombus in honey bee (Hymenoptera: Apidae) colonies, terrestris L.) in the field, Proceedings of the 6th J. Econ. Entomol. 82, 69–74. International Symposium on Hazards of Pesticides Wigglesworth V.B. (1972) The principles of insect to Bees, Sept. 17–19, 1996, Braunschweig, Ger- physiology, 7th ed., Chapman and Hall, London, many, Appendix 30. UK. Schroeder W.J., Sutton R.A., Beavers J.B. (1980) Williams C.M. (1956) The juvenile hormone of insects, Diaprepes abbreviatus: Fate of diflubenzuron and Nature 178, 212–213. effect on non-target pests and beneficial species after application to Citrus for weevil control, Wittmann D. (1982) Determination of the LC50 of J. Econ. Entomol. 73, 637–638. Dimilin 25 WP for honey bee brood on free flying Steen J. van der, Ruijter A. de (1990) Test methods to colonies as an example for the use of a new Apis- determine the hazards of IGRs (insect growth reg- larvae-test, Apidologie 13, 104–107. ulators) on honey bee brood, using phenoxycarb Zdarek J., Haragsim O. (1974) Action of juvenoids on as an example, Proceedings of the 4th International metamorphosis of the honey bee, Apis mellifera, Symposium on Harmonization of Methods for Test- J. Insect Physiol. 20, 209–221.