CropScience) has high root systemicity and insecti-cidal activity against a wide range of economicallyimportantinsectpestsofsugarbeet(
L.; and oilseed crops (Ohkawara et al.2002, Schwarz et al. 2002). However, there are con-cerns of potential adverse impacts of this class of insecticides on nontarget organisms. Of particularconcern are potential risks to pollinators that may beexposed to chloronicotinyl residues in plant pollenand nectar, and the past decade has witnessed con-siderable debate over this impact on bees. Severalreviews indicate that chronic exposure to concentra-tionsofimidaclopridequivalenttothosefoundinseedtreatments pose negligible risks to honey bees(Schmucketal.2001,Mausetal.2003,Schmuck2004,Faucon et al. 2005). In laboratory studies, Kirchner(1999) found bees rejected imidacloprid-contami-nated food at 20 ppb, and Decourtye et al. (2001)reported compromised learning in bees after expo-sures as low as 12
48 ppb. Although the majority of studies report no acute or chronic toxicity at imida-clopridexposures
20ppb(forreview,seeMausetal.2003), a controversial study by Suchail et al. (2001)found high chronic toxicity in honey bees fed lowconcentrations of imidacloprid.Some laboratory and
eld investigations have eval-uated potential effects of clothianidin on pollinators(C.D.S.-D.andM.S.Spivak,unpublisheddata;Frank-linetal.2004;Baileyetal.2005;King2005).However,no experiments have monitored honey bee coloniesduring and after exposure in the
eld. Here, wepresent results of a long-term study ascertaining im-pactsonhoneybeecoloniesduringandafterexposuretocanolagrownfromseedtreatedwiththemaximumrecommended rate of clothianidin.
Materials and Methods
With exception of data collected during the springcolony assessment, this research was conducted inaccordance with the Organization for Economic Co-operation Development Principles of Good Labora-tory Practice (EPA 1989, OECD 1999).
Nongenetically transformed canola,
variety Hyola 420 (Interstate PaycoSeed Company, West Fargo, ND), was subjected toone of the following treatments: 1) clothianidin-treated seed prepared with a seed-treatment slurryconsistingofProsperFL(9.64%clothianidin,plusfun-gicides thiram, carboxin, and metalaxyl, Bayer Crop-Science)at1,375.0ml/100kgseed,andPoncho600FS(48.96%clothianidin,BayerCropScience)at458.7ml/100kgseed.Thisslurrydeliveredclothianidinat400gactive ingredient (AI)/100 kg seed, the highest rec-ommended rate for canola in Canada; or 2) controlseed treated with a specially prepared Prosper FL
formulation, delivering the same rate of fun-gicides and inert ingredients as the clothianidin-treatedseed,withoutclothianidin,andaspeciallypre-pared Poncho 600 FS blank formulation that alsocontained the inert ingredients but lacked clothiani-din. Samples of clothianidin-treated and control seedweresenttoALSEnvironmental(Edmonton,Alberta,Canada) for veri
cation of clothianidin content.
eldswereestablishedon the University of Guelph Elora Research Station,Elora, Ontario, Canada (two sites, E1 and E2) and ontwo farms owned by grower cooperators (two sites,W3 and W4) in proximity to the Elora Research Sta-tion. Each site consisted of two 1-ha
eldplanted with clothianidin-treated seed and the other
eldstotal.Fields at each site were separated by at least 295 m,determined by global positioning system. Canolaseedswereplantedon20
rm seedbed at the highest recommendedrate(8.0kg/ha)toensureahighnumberofplantsandample forage for bees. Therefore, clothianidin wasappliedat32g(AI)/ha.Asectionof
eldE1Cwasnotseeded properly during the initial planting and wasreplanted on 6 June. All
elds received a preplanttreatment with Treßan EC (Dow AgroSciences Can-ada Inc., Calgary, Alberta, Canada) and fertilizer ac-cording to Ontario canola production recommenda-tions(OMAF2004).Comprehensiveground-truthingdetermined that availability of alternative foragewithin 1 km of the colonies situated in canola
eldswas negligible. To our knowledge, no other ßoweringcrops or corn grown from seed treated with clothia-nidin were planted within a 1-km radius of our test
elds.Canola seedling emergence rates were determinedon 3 and 7
8 June. On each date, the number of emerged plants per 1-m row in 10 randomly selectedlocations in each
eld was counted. Plant growth ineach meter (all plants combined) was assessed andassigned a rank: 1, cotyledon; 2, two true leaves; 3,three true leaves; 4, four true leaves; and 5,
ve trueleaves. Crucifer ßea beetle,
Goeze, and striped ßea beetle,
(F.),damageineachmeterofplantsalsowasassessedby rank: 0, no damage; 1, up to 25% damage; 2, up to50%damage;3,upto75%damage;and4,100%damage.
Colony Establishment, Maintenance, and Trans-port.
Before placement in canola, test colonies (
32) were held at a spring apiary near the TownsendHouse Bee Research Facility (THBRF) at the Uni-versityofGuelph.Testcolonieshadsuccessfullyover-wintered, and they were headed by naturally matedqueens of the same lineage and approximately thesame age. Throughout the study, queen cells weredestroyed whenever found to prevent swarming. Atinitiation of the experiment each colony consisted of asinglebroodchamber(24cmindepth,10framespersuper), below a shallow honey super (16.5 cm indepth, nine frames per super). An Ontario Agricul-tural College (OAC) pollen trap (Smith 1963) was
tted above the bottom board of each colony. Allmoveable components of each colony (e.g., broodchambers, supers, covers) were labeled to ensure ac-curate cross-referencing. In spring, all colonies re-ceived three 30-g dustings of a 5:1 icing sugar/oxytet-racycline(Terramycin25,P
zerInc.,NewYork,NY)mixture at 1-wk intervals to protect against American766 J
Vol. 100, no. 3