© Det här verket är upphovrättskyddat enligt Lagen (1960:729) om upphovsrätt till litterära och konstnärliga verk. Det har digitaliserats med stöd av Kap. 1, 16 § första stycket p 1, för forsk- ningsändamål, och får inte spridas vidare till allmänheten utan upphovsrättsinehavarens medgivande. Alla tryckta texter är OCR-tolkade till maskinläsbar text. Det betyder att du kan söka och kopie- ra texten från dokumentet. Vissa äldre dokument med dåligt tryck kan vara svåra att OCR-tolka korrekt vilket medför att den OCR-tolkade texten kan innehålla fel och därför bör man visuellt jämföra med verkets bilder för att avgöra vad som är riktigt. This work is protected by Swedish Copyright Law (Lagen (1960:729) om upphovsrätt till litterära och konstnärliga verk). It has been digitized with support of Kap. 1, 16 § första stycket p 1, for scientific purpose, and may no be dissiminated to the public without consent of the copyright holder. All printed texts have been OCR-processed and converted to machine readable text. This means that you can search and copy text from the document. Some early printed books are hard to OCR-process correctly and the text may contain errors, so one should always visually compare it with the images to determine what is correct. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 C M 0 1 2 3 4 5 6 7 8 9 10 11 12 IN C H VITAL VOLATILES - HOST LOCATION IN PARASITIC WASPS ATTACKING BARK BEETLES EVA M. PETTERSSON Chemical Ecology Göteborg University 2000 Cover photography: A par asitic wa sp (hym enopte ran parasi toid , natu ral size ca 4 mm) R hopalicus tutela ovipo siting on a bark beetle larva concealed under the bark of a No rway sp ruce. No te the ovipositor ( close to the mid dle leg) p enetrating the bark ( pho tograp hy by Göran Birgersson). Göteborg University Fac ulty of S cience 2000 Diss ertation VITAL VOLATILES - HOST LOCATION I N P ARASITIC WASPS ATTACKING B ARK B EETLES Eva M. Pettersson Chemical Ecology, Göteborg University Box 461 , SE-405 30 G öteborg Swe den Avhandl ing för filosofie dokt orsexam en i Kemisk Eko logi som enligt beslu t av Naturve tenskap liga Fak ulte tsnä mnd en kom mer att förs varas offen tligt fred ag den 15 dece mber 2000 kl 10:0 0 i För eläs ningssalen, Botaniska Institut ionen, Carl Sko ttsbergs Gata 22 B, Göteborg. Fak ulte ts­ oppo nent: Pro fess or Ted C . J. Tur lings, University of Neu chât el, Schwei z. Göteborg 2000 Vasastadens Bok binderi AB ISBN 91- 888 9 6- 19- 6 Dis sertation fo r the D egree of D octor of Ph ilosoph y in Chemical Ecology pre sented at Göteborg Universty, 2000. VITAL VOLATILES - HOST LOCATION IN PARASITIC WASPS ATTACKING B ARK B EETLES Eva M. Pettersson Chemical Ecology, Göteborg University, Box 461 , SE-405 30 Göteborg, SWEDEN E-mail: Eva.Pettersson@chemeco l.gu.se ABSTRACT - Parasitoid s are insects that spe nd their immatu re stages fee din g on other arthrop ods and eventu ally kill their hosts. Parasitic wa sps (Hy menop teran par asitoids ) of bark beetles (Co leopte ra: Scolytidae ) have a remark able ability to locate and ovipos it on hosts wh ich are concealed und er the bark of conifer ous trees. The trees are dam aged or kill ed by bark beetles, whic h introdu ce path ogenic microorganisms into the phlo em wh ere they breed. De spi te the lack of visual cue s, the para sitoids dist inguish bark beetle attacke d trees containing sus cep tible hosts (ma ture larvae p resent in pup al chambers). The aim of this thesis was to ide ntif y the host location cue s used by fou r spec ies of bark beetle para sitoids. Synthetic attractants for these par asitoid s we re suc cessfu lly dev elop ed. Electron microscop y (SE M and TEM) revealed several type s of odo ur per cep tive sensilla on the par asitoid antennae. Volatile chemical sample s collected from conifer s wit h, or wit hou t, bark beetle larvae wer e analysed chemically (GC -MS) and electrop hysiologically (GC -EAD). Monoterp ene hyd rocarbons wer e the pre dom inant volatile compou nds of healthy conifer s, whi le the pre sence of oxy genated monoterpe nes ind icated bark beetle infe sted trees. A limited num ber of compou nds , pri marily oxy genated mono­ terp enes, elicited dete ctable neu ral (EA D) activity in the par asitoid antennae. EAD- active compou nds wer e used to pre par e synthetic baits wh ich attracted p arasitoid fem ales in behavioural tests, includ ing both Y- olfa ctometer and win d tunn el bioassays. The attractive compou nds seemed not to arise from the para sitoid hosts, the bark beetle larvae, but from the host-p lant comp lex, involving microorganisms. In add ition to having ecological and evoluti onary interest, increased kno wle dge of the host location mechanisms in these par asitoid s is pot entially use ful in dev elop ing environmentally saf e control method s against ep ide mic ou tbreaks of b ark beetles breedi ng in conifers . KEY WORD S: Parasitoids, Hymenop tera, Pteromalid ae, Rhopalicus tutela, Roptrocerus mi rus, Roptrocerus xylo phagorum , Braconid ae, Coeloides bostr ichorum, host location, odo urs , oxy genated monoterpen es, synthetic attractants, antennal morph ology, chemical analysis, electroph ysiology, bioassay, bark b eetles, Sc olytida e, c onifers Chemical Ecology, Göteborg University ISBN 91 -88 8 9 6-1 9-6 • Wïm - ' ::.;V;>' /; • ' mmmm - »"*-'•'••• 'AiAAAtAi. • • : '•••/;• , : ' •••" W-:.^r'-C ............. '*4- ; : •~f t .S'«« "" ïWV:"* ;- S ••• ;'•• ifs®! - • '•••• V " : :: " ' '" : 'a:"a, . ' • " " ' • • ; ; • ; ; ÉtwÄ : " ' s; ' : - '•'-- • •••••-Ätsi« :• .-. • • > > . . •: ' • • '• . • : .. . : •;ï • • . V ; '. ^ ù4>4"W^JI •. •. . • . .; , :. ^ « s&1.2îl:c'«?ï. • , : - /V.' V.;:;' 'Éakj '• A. •••• :. ' »a«* ' :• •: • «v® •sKvS.ir-'Ä's r^ ;•"• :•/• - :: :; iV • : v - i ? *;• y ^ rS. : v ; S S i ' - : . •: " ' " - . ; ; - , : ; : ;;•• - .•;,/ k Si:«: > - v m~i: aaa o > , ; v T - - . s w : : : ; : : iiiiSteiap ^V-' '•> 'V-x ; •• : : ' : : ' ? '. ..' . . '• »•• •' .V-:";. •;• •;• ••• ( • •• ? .« •• & ? , • : : ' / \ Contents List of Papers 1 Objectives 2 Introduction 3 Parasitoid s 3 What is a Parasitoid? 3 Parasitoid Spe cies Add ressed 3 Attack S trategies 4 The Life of a Bark Be etle Parasitoid 5 Bark B eetles 7 Bark B eetle Spec ies Addre ssed 7 The Life of a Bark Beetle 7 Parasitoid- Host Interactions 8 Host Location Mechanisms 8 Mul titrophic Interactions 8 Method ological Conside rations 9 Exp erimental Insects & Host Trees 9 Sum mary of Techniqu es 10 Samplin g of Volatile Comp oun ds 11 Analysis of O dour Samples 12 Electrop hysiology (G C-EAD ) 12 Chemical Analysis (GC -MS) 14 Main Results and Implications 15 The Role of the Antenna 15 Antennal Morpho logy 15 Host Location Cues 16 Chemical Cues of H idde n Hosts 16 Antennal Activity 16 Volatiles from Host Inf ested Logs 17 Behavioural Attractants 20 Baits in Bioassays: N atural Baits', Sy nth etic B aits 20 Assaying Attractants: Y -tub e W alkin g Bio assay 21 Win d Tu nne l F light B ioassay Possible Sourc es of Parasitoid A ttractants 23 Concluding Remarks 25 Svensk S ammanfattning (a Sw edi sh Sum mary) 26 Acknowledgements 29 References 31 1 List of pap ers This thesis is based on the follo win g pape rs, refe rred to by their Roman num bers: I PETTERSSON E.M., HALLBERG E., and BIRGERSSON G. (200 1) Eviden ce for the imp ortance of odo r-pe rcep tion in the para sitoid Rhopalicus tutela (Wa lker ) (Hy menopte ra: Pteromalida e). Jour nal of A pplied Ent om ology, acc epte d for pub lication II PETTERSSON E.M. Volatile attractants for three Pteromalid para sitic was ps attackin g concealed spru ce bark beetles. Manus crip t sub mitted t o Chemo ecology III PETTERSSON E.M., SULLIVAN B.T., AND ERSON P., BE RISFO RD C.W., and BIRGERSSON G. (20 00) Odo r p erception in bark beetle p arasitoid Roptrocerus xylo phagorum (Ra tzeb urg) (Hy menopt era: Ptero­ malida e) exp osed to host associated volatiles. Jour nal of Chemic al Eco logy 26 ( 11 ) : in p ress IV SULLIVAN B.T., PETTERSSON E.M., SE LTMANN K.C., AND BERISFO RD C.W. (20 00) Attraction of the bark beetle para sitoid Roptrocerus xylo phagorum (Hy menopte ra: Pteromalida e) to host-associated olfa ctory cu es. E nvi ronm enta l En tomo logy 29 ( 6) : in p ress V PETTERSSON E.M., BIRGERSSON G., and WITZG ALL P. Synthetic attractants for the bark beetle para sitoid Coeloid es bos trichorum Giraud (Hy menopte ra: Braconid ae). Manuscr ipt sub mitted to Nat urw issensch afte n VI PETTERSSON E.M. Volatiles collected in short-range fro m pote ntial hosts of Rhopalicus tutela (Hy m.: Pteromalida e), a bark beetle par asitoid . Manusc ript Papers I and I II are based on researc h ideas by all authors. I compiled Paper I, and am responsible for the experimental design, field and laboratory work (except the electronmicroscopy), and statistical analysis. In Papers III, I am responsible for the compilation of the manuscript, the electrophysiology and associated chemical analyses. I took part in the planning and the compilation of Paper IV. I am responsible for the experimental design, field and laboratory work, data analysis, and the compilation of Paper V. I am solely responsible for Papers II and V I. A doctoral thesis at a University in Sweden is produced either as a monograph or as a collection of papers. At the Faculty of Science it is usually done in the latter manner, where the introductory part constitutes the formal thesis which summarises the attached papers. These have already been published or are manuscripts in press, submitt ed or in preparation . Objectives AA/ vva Mynyvy VVAv^ A fe male parasiti c w asp (hymenop teran parasit oid) is attracted to a bark beetle infested spruce tree from a distanc e and accurately locates a host under the bark, desp ite the lack of obvious visua l cues on the bark surface. The w asp then penetrates the bark with her ovipositor. An egg is laid on, or near, the bark beetle larva, whi ch is kille d dur ing the developmen t of the p arasitoid larva. GC-EAD analysis shows that only a limited number of volatile comp ound s, in the comple x odour bouque t surround ing the bark beetle larvae, elicit neural activity in the pa rasitoid antennae (illustr ation inspired by Eva Meiners). This illu stration gives a brief introdu ction to the life -history of parasitic wasp s attackin g bark beetle larvae concealed under the bark of conifers. The aim of this thesis was to deter mine if host location by larval parasitoids of bark beet les is mediated by volatiles, and then to determ ine the source and identity of sem iochemicals requi red for host lo cation. 3 - IN T ROD U CT ION - Parasitoids What is a Parasitoid ? Parasitoids are para sitic end opte rygote insects (und ergoing full meta­ morpho sis - egg, larva, pupa , adu lt - with win gs dev elop ing internally) whe re the adu lt is free living whi le the larvae dev elop by fee din g on or with in arthropo d ho sts. A single host indiv idual sus tains the d evelopm ent of o ne or more para sitoids, and the host is almost invariably kill ed by the interaction (M ills 199 4 ; Qui cke 1997 ). A par asite, on the other hand, is defi ned as an organism that for all or some pa rt of it s life der ives its food from a living organism of another spec ies, and the host is generally harmed to some ex tent but i s usu ally not k illed by the association. In the old er literatur e, para sitoids are ofte n refe rred to simply as "par asites", wh ich coul d cau se conf usion . Several alternative termino­ logies have been pro pose d, and in recent years "par asitoid" has become the generally accepte d term. This grou ping is not a monoph yletic entity since para sitoids are kno wn from seven insect ord ers (rev iew ed by Qu ick e 199 7 ). Howe ver, the vast majo rity of al l p arasitoid spe cies belong to the orde r Hymenopter a w hich comp rise a highly succ essf ul grou p w ith prob ably more than a million spec ies (Qui cke 199 7 ) , or about 20% of all insect spec ies (La Salle and Gau ld 199 1 ). "Par asitic hymenopter an" does , with few exc ept ions, imp ly the same thing as "pa rasitic was p", and both terms are use d wide ly in recent literature. Both terms are use d in this thesis. Herein, "par asitoid" is used synonymousl y wit h "Hy menopte ran para sitoid", a s this is the order addr essed in m y research. Parasitoid Species Add ressed This thesis concerns four of the most common para sitoid spe cies associated wit h the bark beetle Ips typographus , althoug h these para sitoids are poly pha gou s and attack o ther bark beetle spec ies as well. The para sitoids originated from Swe den and sou theastern USA. Their pote ntial as control agents against Ips and Den droc tonu s beetles has been evaluate d by Mills and Schlup (19 8 9 ). The fam ily Braconidae , of the supe rfam ily Ichneum onoidea (He dqvi st 199 8 ) , contains the largest para sitoids (ca 10 mm), whil e the smaller spec ies (ca 3 -6 mm) b elong to the Pteromalida e, of the supe rfam ily Chalcid oide a (Bo uce k and Heyd on 199 7 ; Grissell and Schauff 199 7 ). • Coeloide s bos trichorum Giraud (Hy menopte ra: Braconidae ) (fro m Swe den: Pap er V) is mainly distr ibut ed in northern and central Euro pe, eastern Euro pea n Rus sia, wes tern Siberia, and Mongolia (To bias et al. 1986 ; Krüger and Mills 1990 ). Oviposi tion is cond ucte d th rou gh the bark. The ovip ositor of C. bost richorum is longer than those of the other stud ied p arasitoid s, w hich mak es it poss ible for this spec ies to reach hosts at greater dep th. Coeloid es bos trichorum has been repo rted as the most effic ient bark beetle para sitoid of I . typographies (Kr üger and Mills 1990 ). This spe cies seems to be entirely confi ned to bark b eetles breeding in sp ruce (He dqv ist 199 8 ). • Rhopalicus tutela (Wa lker ) (Hy menopte ra: Pteromalida e) (fro m Swe den : Pap ers I, II, and VI) is wide ly dist ribu ted in Eur ope (Gra ham 196 9 ) and North America ( Bur ks 197 9 ). It is one of the more aggressive spe cies, fighting off other para sitoids at ovip osition sites on the bark surf ace (Kr üger and Mills 1990 ). Oviposi tion is conduc ted through the bark . This spe cies has been repo rted as an imp ortant para sitoid of the bark beetles Ips acumin atus (Gy llenhall) and Tom icus pini perd a (L.) (Her ard and M ercadi er 199 6 ). • Roptrocerus mir us (Wa lker ) (Hy menopte ra: Pteromalida e) (fro m Swe den : P ape r II) is the smallest of the stud ied para sitoids. This spec ies is repo rted fro m Eur ope (Gr aham 196 9 ; M. Kenis, pers onal commun ica­ tion) and Asia (Ya ng 198 7 ). Roptrocerus spp . enters the bark beetle galleries to ovip osit. • Roptrocerus xyl ophagorum (Ra tzeb urg ) (Hy menopte ra: Pteromalid ae) (fro m Swed en: Pap er II, and from sout heastern USA: Pap ers III and IV) has been fou nd in par ts of Euro pe, America, and Asia (Gr aham 1969 ; Bur ks 1979 ) whe re it is native. It has been succ essf ully introduc ed against exo tic Ips grandi collis bark beetles in Au stralia. Roptrocerus xylo phagorum has been repo rted as an imp ortant para sitoid of the bark beetles I . grandic ollis and Den droc tonu s fron talis (Mo ore 197 2; Mills 198 3 ). Several other economically importa nt scolytids are also pa rasitised suc cessfu lly by this spe cies (M ills 198 3). Roptrocerus xyl ophagorum resp ond s quic kly to changes in host num bers (Da hlsten 198 2), hence it may be an ef fec tive regu lator of bark beetle po pul ations. Attack Strategies Parasitoids of bark beetles can attack various dev elop mental stages of their host. How ever, all of the para sitoids stud ied in this thesis are solitary para sitoids, whi ch prim arily attack late instar bark beetle larvae. This is the pref erred host stage for most para sitoids pre ying on bark beetles (Da hlsten 198 2). These late instars are pres ent in larval mines or 5 pu pa l chambers und er the bark of conifer ous host trees, wh ere the bark beetles repr odu ce. In addit ion, R. tutela (Krü ger and Mills 199 0) and R. xylo phagorum (Sa mson 198 4 ) have been repo rted to occasionally ovip osit on youn g pu pae . In add ition to attackin g diffe rent hosts or host stages, hymenopter an para sitoids also emp loy a variety of attack strategies. The idio biont strategy is an attack s trategy that compris es several para sitoid l ife- history characteristics, some are given in Table 1. The para sitoids disc usse d in this thesis are all classified as id iobionts. Idiobiont characte ristics Generalists attackin g several host species Host concealed hosts typicall y embedd ed in p lant tissu e, s uch as bark beetle larvae presen t in pupal chambers beneath the bark of the host tree Ectopar asitism laying their eggs onto or close besid e the host, parasito id larva e feed ex ternally on the host Long-term host para lysis female p arasitoids par alyse the host befo re ovip ositing Host stage attacked l arger than the wasp Relatively long adul t life s pan (from a bout 20 days to several months, Krüger and M ills, 1990 ) Table 1. Some traits associated with idiobio nt parasi toid strategy (Quic ke 1997 ). The Life of a Bark Beetle Parasitoid Althoug h bark beetle para sitoids are free -living dur ing most of their life, they requ ire a bark beetle host (e.g . a mature bark beetle larva) to complet e their repr odu ctive cycle. Parasitoid fem ales locate sus cep tible hosts, whic h are para lysed with a venom inje ction prio r to ovipos ition (egg laying) (Qu ick e 199 7 ). To avoid sup erpa rasitism, the fem ale para ­ sitoid may mark the host/ovi posi tion site (Le nteren 198 1 ). Du ring its deve lopm ent the para sitoid l arva k ills the bark beetle larva. The dev elop ment of p arasitoid s is very tempera ture dep end ent (Krü ger and Mills 199 0). Parasitoids can overw inter beneath the bark as immatur e stages. Some spe cies are repo rted to enter diap aus e ( mainly below 15° C) whi le others do not have a true dia pau se and become active again with wa rm tempera ture s (Krü ger and M ills 199 0). At tempera ture s above 18°C para sitoids typ ically emerge as adul ts 3 -4 we eks after ovipos ition. Male par asitoids emerge ahead of fem ales, and aggressively compete for fem ales that app ear later on the bark surf ace. Rhopalicus tutela fem ales have a num ber of mature eggs in their ovaries already whe n they emerge, and can start host searching immedi ately afte r mating. Howev er, C. bos trichorum fem ales req uire s a pre- ovipos ition peri od of 8- 1 0 day s, at 20° C, a fter m ating to de velop m atu re eggs (Kr üger and Mills 199 0). Hymenop terans have hap lod iplo id sex det ermination, in whi ch unm ated fem ales lay unfe rtilised (hap loid ) eggs that dev elop into males, wh ile mated fem ales lay fert ilised (dip loid ) eggs that dev elop into fem ales (Qu ick e 199 7). Other fact ors affe cting the sex ratio in hymenopter an para sitoids are host age, maternal age and nutr itional status , host den sity, and para sitoid den sity (W aage 198 2; Waage et al. 198 5; King 198 7). In add ition, many par asitoid s are able to dist inguish hosts of d iffe rent size laying unf ertilised, m ale, eggs on smaller hosts and fert ilised, fem ale, eggs on larger hosts (Ke nis 199 6). A convincing exp lanation f or this strategy is that para sitoid fitn ess should increase wit h increased par asitoid size , wh ich in turn increases wit h host siz e, and that fem ales gain more fro m being large than males do (Ch arnov et al. 198 1). It is not yet kno wn wha t cu es the fem ale p arasitoids use to d etermine host sui tability. Many hymenopte ran para sitoid spec ies pla y a valu able role in pest control, and mu ch of th e interest in para sitoid b iology is d ue to an interest in imp roving biological control (i.e . imp ortation, aug mentation and conservation) through a better und erstand ing of the control agents (Vi nson et al. 199 8). The first doc ume nted use of a par asitoid in biological control involves the imp ortation and release of the dipt eran Cryptochetum iceryae (W illiston) from Aus tralia in 188 8 against cottony cus hion scale in Califor nia. This para sitoid established suc cessfu lly and still contribute s in controlling the host pop ulat ions (M ills 199 4). How ever, th e effic acy of para sitoids in regu lating host p opu lation den sity is influ enced by many fact ors (as mentioned above) , th erefor e k now ledg e on their biology is valuab le f or biological control pro grams. F urth ermore, par asitoid s are essential in maintaining the div ersity of natu ral commu nities (Qu ick e 199 7). Parasitoid s' h igh pos ition in the foo d chains sug gests that they are par ticu larly liable to ext inction and also like ly to includ e many key stone spec ies (La Salle and G auld 199 1). Bark Beetles Bark Beetle Species Addre ssed There are at least 6.000 bark beetle species from 181 genera worldwid e (Byers 1995). In Swe den t here are 82 species, a nd ab out 10 breed in trees of economic signific ance (Spessiv tseff 1922; M. Knize k, personal commu nication). All of the three bark beetle speci es (Coleop tera: Scolytid ae) examin ed in this study breed in conife rs (Pinopsi da: Pinaceae) . • Ips typographus L. (the eight spined spruce bark beetle, from Sweden : Papers I, II, V, and VI) breeding in Norway spruce (Picea abies L. Karst.) . Ips typographus belongs to one of the most aggressive bark beetle genera (Mills 1991), killing healthy trees during epidem ic outbreak s. This beetle causes damage to spruc e forest s and is of economical impo rtance in Swedis h forestry . • Ips grandi collis Eichhoff (the southern pine engraver, from southeas tern USA: Papers III and IV) breeding in loblolly pine (Pinus taeda L.). Althou gh not as aggressive as the other two species, I. grandico llis has caused damage to Australian pine forest s, where it was accidentally introdu ced and lacked native natural enemies (Sams on and Smibert 1986; Berisford and Dahlsten 1989; Abbott 1993). • Dendro ctonus frontalis Zimmerm ann (the southe rn pine beetle, from southeas tern USA: Papers III and IV) breeding in loblolly pine. This is the most destru ctive forest pest in the southern United States and an obligate tree kil ling spe cies (Pay ne 1980). The Lif e of a Bark B eetle Semiochemicals, both from trees and bark beetles, are crucial for bark beetle reprod uction . The resin of c onifers is both a physica l (sti cky) and a chemical (toxi c) ba rrier, protectin g the tree against bark an d w ood b oring insects. To overcome this defence of their host trees, the first colonising bark beetles co-ord inate a massive attack by emitting aggregation pheromo nes that attract conspec ifics of both sexes . The beetles excavate galleries in the host tree phloem, where they reprod uce. During this proces s variou s associated microorganisms, particu larly yeasts and fungi (Leufv én and Birgersson 1987, Kroken e and Solheim 1996), are introduced into the phloem. The trees are killed by beetle mining that girdles the phlo em, and the grow th of a ssociated microorganisms, that cut off th e water a nd nutr ient transporta tion in the vascular t issues of the tree. Microorganisms seem to hasten tree dea th although D. fro ntalis is able to kill trees in the absence of microorganisms (Br idg es et al. 198 5). Bark beetle f emales lay eggs, h atching larvae generate galleries, a nd late larval stages exc avate pup al chambers whe re they pup ate. About 40 day s after infe station (by I. typographus und er laboratory conditi ons, at 20° C) the new generation of bark beetles emerge from the host tree. Parasitoid-H ost Interactions Host Location Mechanisms Host location by para sitoids is a complex pro cess that typ ically can be div ided into five main step s: host habitat location (i.e . long-range host location), foll ow ed by short-range host location that in tur n is divi ded into host location, host accepta nce, host sui tability and host regulat ion (Vin son et al. 199 8). It is generally recognised that host location like ly involves mu ltip le cue s and several sensory modal ities (D e Moraes and Mescher 199 9). Concerning bark beetle par asitoid s, how ever, cue s such as infra red rad iation (hea t), a nd h ost generated v ibrations have been rule d out (Ri cherson and Bord en 197 2; Mills et al. 199 1; Pape r I) . Semio- chemical* stimuli are commonly use d by host for aging hymenopte ran par asitoids (V inson 198 5; Rut ledg e 199 6; Vinson et al. 199 8). Multitrophic Inte ractions Parasitoid s attacki ng larvae commonly func tion with in a tritroph ic context (pa rasite id -host-host habitat). Host fora ging par asitoids have to deal with the so-called dete ctability/rel iability pro blem (Ve t et al. 1995 ). This means that althou gh the signals emitted by the hosts are very reliable indi cators of a tru e host, such signals are diff icul t to det ect as the hosts are und er selection to avoid being foun d and par asitised . On the other hand, signals fro m the host sub strate (e.g. a spr uce tree), on whi ch the hosts feed or breed, should be easy to dete ct but might not necessarily ind icate the pre sence of sus cep tible hosts. Conseq uen tly, par asitoids that attack bark beetle larvae are und er continu ous selection pre ssu re to opti mise their host fora ging strategy, since effi cient location of well hid den , an d o fte n scarce, h osts is critical for su ccessfu l repro duc tion. * "Semi ochemical", o r the synonymous "inf ochemical", is t he general term for any chemical that is emitted by one organism and invo king a physio logical or behavioura l respons e in another organism. For term inology see Nord lund and Lewis (197 6 ) or Dicke et al. (1990 ). To op timise their host for aging, the pa rasitoids of ten us e cu es both from the host and its habitat (e.g . Vet et al. 19 9 5 ; Ru tledg e 19 96 ; De Moraes and Mescher 19 99 ). Parasitoids wi th pr ior ex pe rience of host related cu es are able to increase their host f oraging effi cacy. F ur thermore, they are also able to learn ad dit ional cu es ind irectly associated wi th the pr esence of su scep tible hosts (V et et al. 19 95 ). There are nu merou s ex ample s of mu ltitroph ic commun ication systems, wh ere host fo raging pa rasitoids us e volatile cu es from various origins and troph ic levels (re viewe d by Ru tledg e 19 96 ). Fo r bark b eetle pa rasitoids , s uc h cu es may be emitted by the host insect itself , its pr od uc ts (e .g. boring du st or fae ces), the host tree, or host-associated microorganisms (D ix on and Payne 19 80 ; Krok ene and Solheim 19 96 ). Thus , pa rasitoids of bark beetles fu nction wi thin a mu ltitrop hic context. Besid es the pa rasitoid itself , this mu ltitroph ic interaction involves the conif erous tree, the bark beetle larvae, and the bark beetle associated microorganisms that are introd uc ed into the host tree p hloem. Natur al selection is exp ected to fav ou r pa rasitoids that are most ef fic ient in host f oraging, a nd bark beetles are un de r selection pr essu re to avoid being fo un d and par asitised . Thus , the majo r cu es us ed in host fo raging are ex pe cted to originate not dir ectly from the host organism itself bu t from other host associated sourc es. There are nu merou s ex ampl es show ing how the fee din g by ph ytoph agou s insects indu ces qu alitative and qu antitative changes in the volatile compo sition of pla nts, wh ich attract pa rasitoids of (e. g. Tu rlings et al. 19 9 1 ; Ru tledg e 19 96 ). Alborn et al. (1 99 7 ) show ed that a sp ecifi c compo un d in the oral secretions of a ph ytoph agou s insect ind uc es su ch pla nt emissions. In the bark beetle pa rasitoid interaction, the tree is dy ing or is already de ad . Howe ver, de ad host habitat material, wi th the hosts removed , may still emit volatiles attracting p arasitoids (T ak åc s et al. 19 9 7 ). Methodologi cal Consider ations Experimental Insect s & Host Trees Norw ay sp ru ce logs cu t fro m 5 0-9 0 years old trees infe sted wi th pa rasitised bark beetle larvae we re collected in fo rests of sout hern Sw ed en du ring late su mmers (m ainly in the pr ovince of Småla nd ). The logs w ere treated an d stor ed for later u se as des cribed in deta il in Paper I . Parasitoids and b ark b eetles to be u sed i n the ex pe riments we re reared ou t from these logs. Un-infes ted logs we re also collected, to be us ed as su ch (fresh lo gs) or to be infested by bark beetles in the laboratory. Parasitoids from southea stern USA (Papers III and IV) were obtained from a laboratory colony. These insects were reared continuo usly on freshly cut logs of loblolly pine infested in the laboratory with I. grand icollis (Sullivan et al. 1999). Sum mary of Techniq ues The analytical techniqu es used in this thesis are listed below , with references to the paper s that give further de tails: • The antennal morpholog y of the parasit oids was examin ed by Scanning and Tr ansmission Electron Microscopy (SEM and TE M) (Pape r I). • The possibl e p resence of "hot spots" on the bark surface above pupal chambers containing bark beetle larvae was in vestigated with an Infrared Radiatio n Scanner (Pape r I). • Volatiles were collected from logs or pieces of bark containing susceptib le bark beetle hosts, in order to isolate potential host location cues. Host associated volatiles were collected from various bark beetle and host tree species . Both whole logs, isolated larvae, and pupal chambers were sampl ed. Three collection techniq ues were used: Dynamic Heads pace Adsorp tions (Papers II, IV, V, and VI) ; Static Headspace Solid Phase Micro-Extr action (HS-SP ME) (Paper VI); and Steam-D istillation (Pap ers III and IV) . • Electrophysiol ogical analysis by combined gas chromatograph y- electroantennographic detectio n (GC-EAD ) was perform ed to screen for compound s in the volatile collections that elicited neural activity in the parasitoid antennae, thus being potentia l host location cues (Papers I, II, III, V, and VI). Some of these GC-EAD active compo unds were then used to pre pare sy nthetic baits for b ehavioura l assays. • Compo unds present in the odou r samp les w ere analysed and identified by combined GC-mass spectro metry (GC-MS) . Identific ation of GC- EAD active compo unds received special attention, since these are potentia l host location cues (Pap ers I-VI) . • The attraction of various specie s of female and male parasi toids to natural odour source s (bark or logs with or without bark beetle larvae) and synthetic baits was examined with two types of bioassays: Short range attraction of walking parasito ids was tested in a Y-tube olfactome ter (Pape rs II, III, and IV) , and long range attraction of flying parasitoid s was as sayed in a wind tunnel (Pa per V). 1 1 Samplin g of V olatile Compounds Chemical Ecology research requires method s for sampli ng volatile chemicals (odour s) from biological material. Sampli ng of volatile compou nds in the headspace (the air surroun ding the sample d material) can be perform ed in several ways. A variety of adsorp tion filters (adsorbe nts) are commercially available for s amplin g of di fferent types of volatile compo unds. Since compound s of interest often are released in minute amounts into the surround ing air, sampli ng generally requires some sort of enrichment. Such enrichment is typically done by first enclosing the material that is to be sample d, with in a limited vo lume, thus increasing the concentration of compound s in the headsp ace and minimising the introduc tion of contaminants. Samplin g can then be achieved by introd ucing an adsorben t, such as a Solid Phase Microextractio n fiber (SPME, Pawliszyn 1997), into the fixed volum e of headspa ce (stat ic head space). Alternatively, very clean air can be allowed to pass over the sample and through an adsorbe nt (dynam ic headsp ace adsorpt ion). The adsorben ts can then either be desorbe d thermally directly in the GC injector, or extract ed with a solvent. The latter desorp tion method has the advantag e that samp les can be analysed repeated ly and can be concentrated by partially evapora ting the solvent. The different sampl ing method s that are available have recently been reviewed and discu ssed b y Millar and Sims (1998). Techniqu es used in this thesis to sample v olatile compound s: • Dynam ic headsp ace adsorption was used to collect volatiles from whole spruce logs, with or without bark beetle hosts (Paper s II and V). Single logs were enclosed in a glass desiccat or, Figure 1, and air was drawn through the enclosu re onto a Porapak Q (divinyl benzene - ethylvinylbenz ene, a slightly polar copoly mer) adsorben t (Alltec h). Porapa k Q was chosen as adsorbe nt due to its trapping capacit y of semipol ar compo unds, s uch as o xygenated monoterpenes ( the compound s of main interest in this thesis) and other conifer related compo unds. The sample d volati le comp ounds w ere eluted from the ad sorbents with d iethyl ether. Volatiles were also collected from pieces o f infe sted or u n-infeste d bark (Pap er IV), and at close-range from pupal c hambers, wi th or w ithout matur e bark beetle larvae, as well as from larvae outside the pupal chambers (P aper VI). 12 Cha r c o a l- fil t e r e d inl e t a ir Spr u c e log Air pum p e d out of th e des i c c a t o r thr o u g h the ad s o r b e n t Sam p l e s wer e elu t e d fro m the ads o r b e n t to a spe c i f i e d volum e of solven t. Fig ure 1 . The equ ip­ ment use d fo r dyn a­ mic headsp ace ad­ sorp tion of volatiles from spr uc e logs wit h or wit hou t bark beetle larvae (illu stration by Claes Joh ansson). • Solid Phase Microextra ction (SPME), with carbow ax-divi nylbenzene as the adsorb ing phase, was used for sampli ng the headsp ace, in an enclosed vol ume, over pu pal ch ambers with or without bark beetle larvae (Paper VI). This particular fiber phase was c hosen since it is efficien t in trapping semip olar compoun ds (e.g . oxyg enated monoterpene s). So me of the advan tages and disadvanta ges of sampli ng by SPME from this material are discu ssed in Paper V I. • Steam-Distil lations of pie ces of bark co ntaining suscep tible hosts were performed as describe d by Sullivan (1997) and Sulliva n et al. (1997) to collect odours associated with the host species from the USA ( Papers III and IV). Such odours were also collected by dynam ic headsp ace adsorpti on (Pa per IV ). Analy sis of Od our S a m p l e s Electrophysiology Combined gas chromatography-el ectroantennographi c detecti on, GC- EAD, was used to screen for compound s eliciting neural activity in the parasitoi d antenna. Electroantennographic method s (GC-E AD and electroantennograph y, EAG) measur e the summ ed neural activity of all sensilla (smal l sensory structure s) present on the antenna. In the schematic GC-EAD recording in Figure 2, two out of five compo unds elicit neural activity in a parasitoid antenna, which is mounted betwee n a grounded and a recording electrode measuring changes in potenti al over the antenna (Arn et al. 1975). Each insect was recorded only once. To ensure that low er response magnitude s at the end of the recordi ng period 1 3 wer e not due to weak ened antennae, the conditi on of the antennae was tested by EAG befo re and after each GC-EAD record ing (Fig . 2* in Pap er III) . In each EAG test, the antenna was exp osed to a standard blend of volatiles that was not sepa rated by GC. The blend in a Pasteur pipe tte wa s pu ffed into the glass interfac e betwe en the GC exit and insect antenna. GC-EAD Sam ple injection Ante nna Column 1 _ L i l _ L Fi gu re 2. GC-EAD analysis, with simulta neous record ings of signals from a GC equipp ed w ith a flame ionisation detect or (FID ), and an insect antenna connected to an electroantennographic detector (EA D). The injecte d sample is split (1: 1 ) betwee n the two detect ors (modif ied from Bäckm an 19 99 ). Che mo- Electrical Signal Trans ducti on The minute ope nings (por es) in the cutic ular wa lls of chemoreceptiv e sensilla ( Fig ure 4) allow c hemicals to d iffus e into the sensillar lum en. In the lum en, there are odo uran t bind ing prot eins, wh ich may faci litate the transfer of hyd rop hobic odou rant molecul es to the sensory den drite althoug h a def inite pro of for the func tion is lack ing (Bre er 199 7 ). There are spe cifi c odo uran t binding sites, receptor s, residin g on the den drit ic membrane of olfa ctory sensory neu rones. Several similar type s of molecu les can bind to the same site, althoug h certain molecu les can generate a stronger respo nse than others do. The binding of odo ura nts to the recepto rs trigger a chain-reaction, resu lting in the activation of ion channels (Ca 2+ or K + ) in the dend ritic membrane. This, in turn , g enerates an electrical imp ulse (dep olarisation) over the membrane, whi ch is pro pag ated to the central nervous system and registered as a resp onse (rev iew ed by Breer 1997 ; Krieger and Breer 199 9 ). Electrical imp ulse s can be det ected by electrop hysiological techniqu es, enabling a screening of p otentially behaviour ally active compou nds . * A Figure in a Paper is referre d to as "Fig.", while a figure in the thesis is referre d to a s "Fig ure". 14 Che mica l an alys is Combined gas chromatograph y-mass spectro metry (GC-M S) identified the compo unds present in the collected volatile sample s. Semip olar, polyethy lene glycol (PEG) coated columns (Cp-Wax 58 CB, or HP Innowa x) were used both in GC-EAD and GC-MS analyses, facilitat ing identifica tion of antennal active compo unds. These column phase s were used since they efficiently separa te semipol ar compoun ds, such as oxygena ted mon oterpe nes associated with da maged conifers . Comp ounds were identifie d by their mass spectra l matches to library spectra and by retention time matches to known standar ds. Fig ure 3. Chemical analysis of vola tile sample s and ide ntification of GC - EAD activ e compo unds we re done by GC-MS. I d e n t i f i c a t i o n of a GC-EA D ac t ive com p o u n d by G C-MS Mas s p e c t r o m e t r y (M S ) reveal s th e ma s s p e c t r u m of Ca m p h o r (a n o x y g e n a t e d mo n o t e r p e n e ), the st r u c t u r e of wh i c h is sh o w n. 1 5 - MAIN RESULTS AND IMPLICATIONS - My results imply that odours are the most import ant host location cues for parasito ids that attack bark beetles breeding in conif ers. Synthetic attractants for the addresse d parasito ids were develo ped. A limited numbe r of key compo unds are required for host location althou gh the volatile compo sitions of host infested materials is comple x. Specific oxygena ted monoterpenes have been identified as essential host foragin g cues, both in Swed ish and North American parasitoid s, irrespe ctive of bark beetle host and conifero us tree species. Some of these compou nds increase in concentration as bark beetle larvae develo p. Compo unds attracting parasi toids do not arise directly from bark beetle larvae but from the int eraction of th e bark be etle, its a ssociated micro organisms, and the infes ted con iferou s host tree tissue. The Role of th e Antennae Antenn a I Morphology Electron microscopy (SEM and TEM) were used to eluci date which sensillar types (sensor y modaliti es) were present and most abunda nt on the antennae of R. tutela (Paper I). Olfacto ry placoid , tricoid , and coeloconic sensilla were found to be abundan t on the antennae of both female s and males, imply ing the impo rtance of od our percept ion. Placoid sensilla were the most abundant type of sensilla. A view of a schematic placoid s ensillum is sho wn in F igure 4, to be compared w ith the SEM and TEM photog raphs of this sensillum from R. tutela in Paper I (Figs . 2 and 3). Amputa tion experime nts have shown that parasit oid femal es f ailed in locating bark beetle larvae if less than half of the parasito id antennae were left (Riche rson and Borden 1972). This indica tes that the antennae are necessary in host finding, although sensilla can be found on several other parts of the insect body (e.g. mouthp arts, tarsi, and ovipos itor) (Slifer 1970). Based on structural evidence , it is possib le to state if a sensillu m is, or is not, a chemoreceptor (Slifer 1970). 1 6 d e n d r i t e s se n s o r y * * ne u r o n e s receptor l umen /(c on tain in g lymph) H j- c utic le Figure 4. Transverse sec­ tion throug h a schematic Sensill um placod ium with porou s walls, which is a typical feature of a chemo- recep tor. The space beneath the p orous cuticle (lume n) is filled with dendr ites, among which two are show n in the figure (mod ified from Gul­ lan and Cranston 1994) . i m L l — ep ide r m a l ce l l f J® I \ ^ s h e a t h ce l l s se n s o r y ^ g • ^ a x o n s ne u r o n e s ( n e r ve fib r e s ) (n e rve ce l l s ) Host Locatio n Cues Parasitoids fa ce a serious challenge in de tecting their concealed bark beetle hosts, wi th no obviou s visua l cu es available. It is pla us ible that cu es other than od ou rs are comp lementary in host fo raging. Previou s stu die s on the antennal morph ology of the bark beetle pa rasitoid Coeloide s br un ne ri (B raconida e) sug gested inf rared -p ercep tive abilities of sensilla p lacod ea, s ince no cu ticula r po res we re de tected (R icherson et al. 19 72 ). Howe ver, in Pape r I cu ticula r p ores we re fo un d in this typ e of sensilla, and no "h ot sp ots" (in creased IR-radia tion) could be de tected above pu pal chambers (P ap er I) , su pp orting the result s by Mills et al. (1 99 1 ) that reje ct inf rared rad iation as a host location cu e. Nu merous stu di es show that pa rasitoids are attracted to volatiles associated wi th their hosts (re view ed by Ru tledg e 19 96 ) , althoug h inf ormation on host location cu es in bark beetle pa rasitoids is scarce. The resu lts from Pape r I, pro vid e evide nce fo r the impo rtance of olfac tory cu es in bark beetle pa rasitoids , and thus constitute the basis for the rest of the wo rk in this thesis. Chemica! Cues of Hid den Hosts Antenn al A ctivity There is no pr eviou s rep ort on the us e of GC-EAD techniqu e on bark beetle pa rasitoids . Therefo re, the wo rk of this thesis started by investigating the od ou r pe rcepti ve abilities of the pa rasitoid antennae by assaying a synthetic stand ard blend wi th GC-EAD (P ap er Ï) . Both fem ale and male R. tutela we re analysed (15 ind ividu als of each gend er). To 17 measure the reliability of the GC-EAD resp onse to a certain compou nd, this was exp ressed as the mean ± standard error. By measuri ng several ind ivid uals exp osed to the same stimuli, it was pos sible to qua ntitatively evaluat e the resp onse elicited by a defi ned amount of a certain compou nd. Pap er I showed that certain host related compou nds elicit neur al activity of r epr odu cible magnitude in antennae from diff erent indivi dua ls. There were no diff erences in resp onsiveness betw een fem ale and male R. tutela. Mon oterpen e hydro carbons (con stitu tive compou nds of fres h host- trees) did not elicit dete ctable EAD -respon ses. All antennal active compou nds we re either bark beetle phe romone compon ents or oxy genated monoterpen es. The conclus ion of P ape r I is that the antennae of R. tutela are well adap ted to odo ur perc epti on, and that certain host related o dou rs elicit neur al activity in the antennae. Volatiles fro m Host Infest ed Logs Based on the results from Pap er I, the thesis work pro ceeded by investigating volatiles sampled from conifer logs containing par asitism- susc epti ble bark beetle larvae (Fig ure 1). GC-EAD analyses (Fig ure 2) were perf ormed to isolate compou nds eliciting neu ral activity, t hus being pote ntial host location cues. A sum mary of all GC-EAD active compou nds foun d in the tested odo ur samples and standard mix ture s are given in Table 2. Sw edi sh R. xylo phagorum has been f oun d to resp ond similarly as the tested Sw edis h para sitoids whe n exp osed to odo urs collected from who le logs (as in Pap er II) or pup al chambers (as in Pape r VI) (E.M . Pettersson, unp ubli shed dat a). The GC-EAD active compou nds we re mainly oxy genated monoterpen es (ind icating dam aged conifer s) in all investigated spe cies, irrespect ive of bark beetle host and conifer ous tree spe cies. Fem ale Rhopalicus tutela and Roptrocerus miru s revealed similar trends in resp onsiveness to host associated volatiles. Most monoterpen e hyd rocarbons (tre e compou nds ) wer e not antennal active, in contrast to several oxy genated monoterpene s. Campho r (co- elut ing wit h minor qua ntities of pino campho ne) and isop inocamp hone elicited the strongest antennal activities (Pap er II, Fig. 2). Similar resu lts were obtained from Coeloid es bos trichorum (Pap er V, Fig. 1). Roptrocerus xyl ophagorum , originating from sou theastern USA and expo sed to volatiles associated with bark beetles breeding in loblolly pine, revealed slightly diffe rent trends in resp onsiveness (Pa per III, Fig . 3). Howev er, the diff erences in resp onsiveness are most prob ably due to diffe rent relative abu nda nce of 1 8 comp ou nd s in sampl es collected from Norw ay sp ru ce (S we de n) or from loblolly pin e ( USA) . Tab le 2. Comp ounds eliciting EAD-activi ty in pa rasitoids atta cking bark beetle larvae. Highlighted compo unds were used in prepari ng various synthetic baits tested in bioassays. (Th e footn ote ind icates the compoun ds used i n behaviou ral tests with each parasi toid species. The specific compo sitions of the various baits are given in the spe cified papers.) Compound Species addressed papers Parasitoid Bar k Tr ee beetle Acetic acid Cb It Pa V 4-Allyl anisole Cb; Rt; R m; Rx Df; Ig; It Pa; Pt II; II I; V Anethol 1 Rx III Benzal deyde Cb; Rt It Pa V; VI Borneole Cb; Rt; R m; Rx 1 It Pa; I; II; II I; V Bornyl acetate Rt2 ; Rx 1 I; I II C3-benzene Cb It Pa V C 4-benze ne Cb It Pa V C4-styren e Rx Df; Ig Pt III S-Cadinene Rt It Pa VI Camph or b ' c ' d ' e Cb; Rt; Rm; Rx Df; Ig; It Pa; Pt I; II; II I; V ; VI czs-Chalcogran 2 Rt I ("r a/z .s'- Chalcogran 2 Rt I />Cym en-8- ol Rt It Pa VI p-Cym ene Rt; Rm It Pa II; VI Fenchol Rt; Rx Df; Ig; It Pa; Pt III; VI Fencho ne b ' c ' d ' e Cb; Rt; Rm; Rx Df; Ig; It Pa; Pt I; II; I II; V; V I Heptanol Cb It Pa V Ips die nol2 Rt I Ips enol2 Rt I lS0| _ . Cb; Rt; R m; Rx 1 It Pa I; II; II I; V; VI Limonene Rt It Pa VI Linalool Rx Df; Ig Pt III Myrtenale Cb; Rt; Rm; Rx Df; Ig; It Pa; Pt II; I II; V; V I 1 9 Table 2. (continuing) Compou nd Parasitoid Bark beetle Tree Pap ers Myrtenol d Cb; Rt ; Rx Df ; It Pa; Pt I; II; III ; V; VI Nonanal Cb It Pa V / ^P heliandre ne Rt It Pa VI Pinocampho ne * Rt; Rm It Pa II; VI trans- - - v b ' c, d Cb; Rt; R m; R x Df ; Ig; It Pa; Pt I; II; II I; V; VI Pinocarvone Rt; Rm; Rx Df ; I t Pa; Pt II; I II; VI T * .: v. 4 ? m a - b - c - d - c Cb; Rt; Rm; Rx Df ; Ig; It Pa; Pt I; II; II I, V; VI ar-Tcrpineol dl e Cb; Rt; Rm; Rx Df ; Ig; It Pa; Pt II; I II; V; VI Terpino lene Cb; Rt; Rm It Pa II; V; VI u.i 83 ; 1 52 Rt; Rm It Pa II; VI u.i. 79 ; 9 3 ; 1 3 7 Rx Df ; Ig Pt III u.i. 81 ; 9 3 ; 1 07 Cb It Pa V u.i. 83 ; 5 5 ; 9 5 Cb; Rt; Rm It Pa II; V u.i. Sq T Cb; Rt; Rx Ig; It Pa; Pt III; V; VI trans- Verbenol2 Rt I Verbenone d Cb; Rt; Rm; Rx Df ; It Pa; Pt I; II, III; V; VI u.i. = un- iden tifie d c ompou nd, pred ominant mass spe ctral f ragments are given; Sq T = sesqui terpen es; Cb = Coeloid es bos trichorum (Sw ede n); Rt = Rhopalicus tutela (Sw ede n); Rm = Roptrocerus mir us (Sw ede n); Rx = Roptrocerus xyl ophagorum (US A); Df = Den dro cton us fro nta lis (US A); Ig = Ips grand icollis (U SA); It = Ips typographus (Sw ed en) ; Pa = Pic ea abie s (Sw ede n); Pt = Pin us taed a (U SA) 1 Constitu ent of the synthetic blend analysed in Pap er III, wh ere the origin of compou nds is given. 2 Constitu ent of the synthetic blend analysed in Pap er I, wh ere the origin of compou nds is given. * Pinocamp hone co-elu ted with campho r, an d w as not measu red sep arately. " ~ e A compon ent of at least one of th e synthetic baits u sed in behavioura l tests wit h a spe cif ied para sitoid spe cies. The spe cifi c compos itions of the various s ynthetic baits are given in the resp ective p ape rs: " Sw edis h Coeloide s b ostrichorum (Pa per V). b Sw edis h Rhopalicus tutela (Pa per I I). c Sw edis h Roptrocerus mir us (Pap er II). d Sw edis h Roptrocerus xyl ophagorum (Pa per II). e North American Roptrocerus xyl ophagorum (Pa per III). The addresse d parasito id species seem to have the same odour perceptive abilities, since the over all response s of differe nt parasito id species to host related compou nds a re very similar. Howev er, the natural baits used in the different GC-EAD analyses were collected from different host materials, therefore there were some qualitati ve and quantitative differe nces between these baits. Differe nces in the bait composi tions are the most probable reason why compo unds occasionally elicited r esponse s in some parasi toids but not in others. Behavioural Attractants Electrop hysiological method s isolate compou nds that elicit neural activity, however it does not disclose the behavioural signifi cance of these compoun ds. This issue require s behavioural assays (bioass ays), which were used to test attraction of walking or flying parasit oids to various volatile baits. The attraction of parasitoid s to "natura l baits" (i.e. logs or pieces of bark, with or without susceptibl e hosts) and "synthet ic baits" (compos ed of compou nds associated with suscept ible hosts) were assayed in behavioural tests. A limited number of oxygena ted monoterp enes w as requir ed to at tract fem ale parasi toids. Baits in Bioassays • Natural baits tested in bioassays were either whole logs (Paper II) , dynamic headsp ace adsorpti ons fr om such logs ( Paper V), piece s of bark, isolated hosts ( Paper IV), or steam distillates from pieces of b ark (Paper s III and IV). The chemical composi tions of the natural baits used were analysed, and are presen ted in th e respecti ve paper. • Synthetic baits ( Papers II, III, and V) were compo sed of compo unds in the natura l baits that elicited GC-EAD activ ity (Ta ble 2), as identifi ed by GC-MS (Figure 3). In most synthetic baits the propor tions among the compoun ds reflec ted their abundanc e in the natural baits. The composi tions of the various synthetic baits are given in the respectiv e papers. Some of the synthetic baits (Papers II and V) were partly based on chemical analysis of changes in the volatile compo sitions of six Norway spruce logs during bark beetle infestati on, until young bark beetles emerged (E.M . Pettersson, unpublis hed data). Correspondin g referenc e analyses were done of un-infested logs. Some GC-EAD a ctive compoun ds, as campho r and isopinoc amphone , were found to increase quantitat ively as bark be etle larvae developed (Figure 5). Verbenone was not found to increase signific antly after three weeks but was used in synthetic baits (Papers II and V) based on its relatively strong GC-EAD activity. 20 - _ - J -Camphor Isopin ocam phon e Terp inen e-4-o l Fench one Verben one 10 20 30 Day s A fter Bark B eetl e A ttack 40 Fig ure 5. The abu nd­ ance of certain oxy ­ genated monoterpen es increase as bark beetle larvae dev elop (ca 20 day s afte r a bark beetle attack) . In nature, parasi toids are attracted to conifer s containing suscep tible bark beetle larvae. These differ chemically from un-inf ested trees by having higher concentrations of oxygena ted monoterpene s. Howeve r, there were no compo unds detecte d in un-infe sted trees that are not also presen t in infe sted tr ees. One aim wh en bioassaying synthetic baits was to find the least numbe r of compoun ds require d to elicit attraction. Theref ore, in the behaviou ral tests of Paper II (Fig. 1) and Paper V (Figure 6), odour s from an un-infe sted spruc e log were used in combination with various synthetic baits (simul ating an infest ed tree). Synthetic baits we re tested a lone in Papers I II and V. Assaying Attractants The first step in bioassays was to observe the attraction to natur al baits obtained from bark beetle infested trees, versus natura l baits from un- infeste d trees or solvents. Parasitoid females signif icantly preferred volatiles from bark beetle infested trees (Papers II, III, IV, and V). In contrast, male parasit oids did not show any such prefer ence (Papers II 2 2 and III). The fem ale resp onses wer e pres uma bly related to host location, whi le the absence of male resp onses was exp ected since males der ive no obviou s benef it fr om attraction to bark beetle larvae. Synthetic baits, compos ed of 3 -8 compou nds , wer e fou nd to be attractive wh en tested against a solvent blank or un- infe sted conifer s in Y-tub e wal king bioassays with fema le R. tutela, R. mir us , and R. xylo phagorum (Pa per s II and III) . In the win d tunn el (Fi gur e 6), fem ale C. bos trichorum oriented tow ards synthetic baits compos ed of at least three compou nds (Pa per V, Fig. 2). Synthetic baits were not as attractive as bark (Pap er III) or spru ce logs ( Pap er V) containing sus cep tible hosts. How ever, wh en certain synthetic baits were combined wit h odo urs from un- infe sted spr uce logs (wh ich were not attractive alone) , this combination wa s as attractive as infe sted logs (Pa per s II and V). These resu lts show that a limited num ber of key compou nds are req uir ed to attract par asitoid s. The oxy genated monoterpen es campho r, isop inocamp hone, and terp inen-4-o l seem to be among the most imp ortant compou nds . E x h a u s t (ch a r c o a l fil t e r ) Ins e c t •~ 7 or i e n t a t i o n fli g h t Lam i n a r air flo w Air inl e t (ch a r c o a l- I R e l e a s i n g j pa r a s i t o i d s Odo u r sou r c e (a spra yerl Fi gu re 6. Exper imental equip ment f or wind t unnel bioassays (ill ustrat ion by Claes Johan sson) . Possible Sour ces of Parasitoid Attractants The aims of this thesis were not only to isolate and iden tify pot ential host location cue s, but also to isolate the sour ce emitting these para sitoid attractants. The results show that the compou nds attracting p arasitoids do not prim arily arise from the bark beetle hosts but rather fro m the interaction of the bark beetle, its associated microorganisms, and the phl oem and cambiu m tissu e of the host plan t (Pap ers IV and VI). Host location cues that originate direc tly from the bod ies of the host insect is pro bably qu ite rare in natu re. As mentioned in the "Mu ltitroph ic Interactions" section, there are instead num erou s exa mpl es of par asitoid s usin g host location cue s indir ectly associated wit h the pre sence of their hosts. Bark beetle boring dus t or infes ted tree tissu e, with or wit hou t bark beetles ( I . gran dico llis), were highly attractive to fema le R. xylo- phagorum (Pa per IV). In contrast, isolated bark beetle larvae, or un- infe sted or artifi cially dam aged tree tissu e we re not attractive. Qua litative and qua ntitative chemical diffe rences were foun d in dyn amic headsp ace samples , or steam dist illates, from bark beetle dam aged tree tissu e (ph loem and cambiu m of loblolly pin e) versus non-da maged tree tissu e (Pa per IV) . Pupa l chambers wer e most complex in chemical compos ition and emitted the greatest quan tities of GC-EAD active compou nds whe n analysing R. tutela (Pap er VI). In contrast, only few compou nds in minute amount s we re d etected i n sampl es fr om isolated b ark b eetle larvae ( I . typographus) . Odo ur samples from pup al chambers wit h or wit hou t bark beetle larvae contained larger quan tities of these compou nds , imp lying that they originate from the tree tissue rather than the bark beetle larvae. In effo rts to ide ntif y the volatiles of pote ntial use in ovipos ition site location, odo ur samples were collected with in a few millimetres of susc epti ble hosts (clo se-range) (Pap er VI). The GC-EAD active compou nds det ected in these samples wer e the same as the ones prev ious ly fou nd in odo ur samples from who le logs (Pa pers II and V) or larger piece s of bark (Pap er III and IV). Thu s, no qua litative diff erence in volatile compos ition at ovip osition sites cou ld b e de tected a s compar ed to prev ious analyses of long-range host location cues (ho st-habitat location cue s). How ever, the thinner bark layer above pup al chambers most likely result in spat ial variability in the compou nds emitted , whi ch wou ld guid e par asitoid s in close-range host fora ging. The pos sible consequ ences of the diffe rence in bark thick ness are disc usse d in Pap er VI. The thinner bark above pupa l chambers may either resu lt in larger 24 release rates of volatiles since compou nds pro duc ed in the phl oem wou ld pas s through more easily. Alternatively, the thinner bark layer might resu lt in low er release rates since a thinner phl oem tissue cou ld prod uce smaller qua ntities of compou nds that attract pa rasitoid s. The GC-EAD active and attractive oxy genated monoterpen es, whi ch arise from conifer s containing susc epti ble bark beetle larvae, do not originate dire ctly from the host organism but from the sur rou ndi ng conifer tissu e (ph loem and cambium ). Oxy genated monoterpen es prob ably arise through spon taneous oxid ation of m onoterp ene pre curs ors exp osed to the air, metabolic p rocessing of monoterpen e hydro carbons by invadi ng insects and microorganisms, and other proc esses (LeufVén and Birgersson 198 7 ; Leu fvén et al. 198 8 ; Hun t et al. 198 9 ; Leu fvén 199 1 ; Smithetal. 1 993 ). Microorganisms seem to be imp ortant in this mul titrophic interaction. The blue -stain fung i spe cies fou nd in association wit h I . t ypographus are diff erent from the fimgi associated with non-aggressive bark beetle spe cies, indi cating that path ogenic fung i are imp ortant for tree killi ng bark beetle spe cies (Kr oke ne and Solheim 199 6 ). In contrast, the aggressive D . fron talis are able to kill trees (by mining that gird les the phl oem) e ven with out associated blue -stain fun gi. Where there wer e less blue -stain fun gi fou nd in attacked trees, the D . fron talis out breaks wer e most severe (Br idge s et al. 198 5 ). Thus , the interaction betwe en bark beetles and b lue -stain fung i may be di ffere nt f or d iffer ent bark be etle/tre e spe cies. Laboratory infe stations of axe nic I . grandi collis (rea red with out microorganisms) still attract para sitoids, althoug h "wi ld" beetle infe stations (wi th associated microorganisms) are pref erred over the form er when given a choice of the two (B.T. Sul livan, unp ubl ished d ata) . Bridge s and Moser (198 6 ) fou nd a posi tive correlation betw een the pre sence of blu e-stain fung i and pre dat ory mites attacki ng D. fro nta lis. The pres ence of bark beetle associated microorganisms thu s seems to faci litate host fo raging in para sitoids a nd p red ators of bark beetles. This imp lies that microorganisms, direc tly or indi rectly, enhance the prod ucti on of volatile host location cues . The observed increased pro duc tion of oxy genated monoterpene s as bark beetle larvae dev elop (E.M . Pettersson, unp ubli shed dat a) coincide s interestingly with a rep orted change in the tree phl oem po lyph enols par enchyma cells (Fra nceschi et al. 2000). This phy siological change has been sugg ested a s an indu cible host tree defe nce against bark beetles and microorganisms. Thus, the volatile host location cue s used by these par asitoid s seem to arise from some interaction betw een hosts, m icroorganisms, and the host tree. - CONCLUD ING R EMARKS - This thesis show s that chemical cues are of vital import ance to parasiti c wasps attacking bark be etle larvae. The odour percept ive abilities of the addres sed parasit oids were investigated and synthetic parasitoi d attractants were develo ped. A limited numbe r of compoun ds, primarily oxygena ted monoterpenes, are crucia l in host foragin g among these parasito ids (see "Main Resul ts and Implic ations" ). Interesting tasks for future research efforts inclu de trying to unrav el the chemical origin of these c ompoun ds, eluc idating the role of enantiomers, and conduc ting a thorough screening of chemical and other cues that elicit oviposi tion. The sophist icated interactions between parasito ids, bark beetles, associated microorganisms and host trees are truly fascina ting, and a most interesting field of research. Increased knowl edge about this compl ex chemical commu nication system is not only of utmost ecological and evoluti onary interest, but is also of potent ial use in develop ing more efficie nt, environmentally safe, biological control techniqu es. ~ En sven sk samma nfatt ning ~ Livsvikt iga do fte r » värd lokal isering hos p arasitstekla r som angripe r barkb orrar Vissa par asitstek lar (pa rasitoider ) angripe r och död ar barkbo rrelarver som ligger do lda und er bark en på barrträd . Stekla rna påv erk ar såle des bark borrens pop ula tionstä thet. Syfte t med det här fors kni ngspro jek tet var att u tröna hu r stek larna finner sin värd trots att det inte tycks finnas någ ra synliga ledt råda r (s e illustr ation s. 2 "Ob ject ives"). I den na avhand ling har ett fata l nyckels ubs tanser (do fte r) ide ntif ierats som nöd vän dig a för vär dlo kali seringen hos barkbo rre-par asitoid er. Dess a sub stanser är främst oxy generad e monoterp ener och kom mer inte i förs ta hand från barkbo rrelarven uta n från des s vär dträ d. De mik roorganismer som (i de fles ta fal l) introdu ceras i träde t genom ett barkbo rreangrepp tyck s ha betyde lse för pro duk tionen av sub stanser som attraherar para sitoider . De und ersökta barkbo rreangripn a barrträ dsa rterna, från Sverige och USA, avgav vissa kem isk a sub stanser som attraherad e par asitoid er. Oavsett barkbo rre och barrträ dsa rt, var des sa sub stanser i stort sett de samma. Eft er kem isk analys av dof tpro ver från naturl igt material tillverka des syntetiska beten, med sub stanser som lockad e par asitoid - honor i olik a beteende test. Vissa av des sa par asitoid arter har tidi gare använt s, eller utv ärd erats, för biologisk kon troll av olik a bark borrearter. Ök ad kun ska p om hur par asitoide rna finner sin värd gör det möj ligt att vid areutv eck la den type n av biologisk kon trollmetod . Det inveck lad e samspe let mellan para sitoider , barkbo rrar, mik roorganismer och vär dträ d är des sut om av stort ek ologisk t och evolutio när t intresse. ~ En svensk sammanfattning ~ Parasitoid er & Barkbor rar Parasitoid er dö dar per defi nition sin värd , vi lket skil jer dom från par asiter som i allmänh et inte död ar sin vär d. Vid are lever par asitoid er främ st på egen hand men kräv er en vär dor ganism en begrän sad per iod av livscyk eln, und er sina tidig a levnad sstadier . Parasitoid er ka n tillhöra må nga olik a insek tsordni ngar men par asitstek larna ( Hymenop tera) ä r d en överläg set största gru ppe n och utgö r ca 20% av alla insekt sarter, vilk et motsvarar över 1 miljon arter. De par asitoide r som avhand las här är alla stek lar och förekommer i stora dela r av vär lde n, der as vetenskap liga namn ä r: Familj Art Äggläggning Braconidae Coeloid.es bostr ichorum genom barken Pteromalidae Rhopalicus tutela genom barken — Roptrocerus miru s i barkborrarnas gångsystem — Roptrocerus xylo phagorum i barkborrarnas gångsystem Möjlig en är de n ringa storleke n (3-1 0 mm) orsaken till att des sa stek lar inte fått nå gra svensk a namn. Samtliga angripe r fler a olika arter av bark borrar som föryngrar sig i barrträ d, vilk et i Sverige frä mst innebä r den åtta tandad e granbark borren (Ip s typographies, Coleopte ra: Scolytidae ) som här föry ngrar sig i gran ( Pice a abi es). I und ersökn ingar av par asitoid er frå n USA var vär dba rkb orrarna Ips grandi collis och Den droc tonu s fron talis (vilk a föry ngrade sig i en amerika nsk tall, Pinu s taeda) . De ssa tre bark borrearter har eko nomisk betyde lse genom att de, när de för eko mmer i stor män gd, kan orsaka betyda nde vär def örlu ster i barrskog ar. Vär dlo kali sering Parasitoid erna läg ger helst ägg på bark borrens sena larvstadi e, som finns i pup pka mmare, dol da und er granbarke n. Den nyk läc kta par asitoid larven äte r seda n up p bark borrelarven. De t är emellertid inte helt enk elt för par asitoid honan att hitta ett bra stäl le för ägg läg gning. I en sko g full av granar mås te hon först hitta ett barkbo rreangripet träd som innehå ller värd dju r i r ätt larvstad ium , sed an åte rstå r att lokal isera exak t rätt stä lle att läg ga sitt äg g på. Trots att det inte finns någ ra synliga led trå dar på bark ytan lyckas båd e R. tutela och C. bos trichorum att läg ga ägg på larven rakt igenom granbark en. De båd a Roprocerus-a rterna går in i barkbo rregång arna och lokali serar larven dä r. ~ En svens k samm anf attning ~ Do fter är mycke t betyde lsef ulla i mån ga and ra par asitoid -vä rd förhål­ land en. När det gäl ler värd loka lisering hos para sitoider på barkbo rrar är dock inte mycke t kän t seda n tidig are. Flera hyp oteser har lagts fra m om vad som väg led er par asitoid en till den dol da värd en: vibrationer, vär mestrå lning (IR ), och dof t. Med und antag av dof t, har övriga hyp oteser f örk astats. Doftforsknings metodik & Resu ltat Stud ier (ele ktr onmikr osk opi ) av yttre och inre antennstru ktu rer visade en mån gfal d av kem orecep torer Figu re* 4 (Fi g. 1-7, Pap er* I) . Det ta indi kera r att do fte r har stor betyd else för para sitoider na. Dof tpro ver insamlad es frå n stockar och barkbi tar med eller utan barkbo rrelarver. Olika artk ombinationer av pa rasitoid- bark borre- barrträd und ersökt es (Pa per s II-V). Ett antal dof tsamlingsmetod er anvä ndes varav den mest använd a beskriv s i Fig ure 1. Do ftpr over insamlades äve n från larver samt pup pka mmare med eller uta n larver (Pap er VI). Insamlade dof tpro ver innehåll er en mån gfal d a v sub stanser. Med hjäl p av elek trof ysiologi (GC -EAD, Fig ure 2) kan man ursk ilja de sub stanser som utlö ser nervak tivitet i insektsa ntennen, vilka därm ed är pot entiella vär dlok aliseringssubs tanser (Ta ble 2). Dof tpro verna analyserad es äve n kem iskt (GC -MS, Fig ure 3). De fles ta GC-EAD ak tiva sub stanserna ide ntif ierades som oxy generade monoterpen er, vilk a ind ike rar ska dad e barrträ d. Kontinuer lig analys av granstockar und er ett bark borreangrep p, visade att avgivningen av vissa GC-EAD akt iva sub stanser öka r när bark borrelarver av "rät t" stadie böljar utve ckla s u nde r barken (ca 3 ve cko r ef ter barkbo rreangrepp ), n ågr a av d essa su bstanser visas i Fig ure 5. Ett antal syntetiska par asitoid beten tillverk ade s, baserat på elek trof ysiologisk och kem isk analys av insamlade dof tpr over. De ssa beten testade s i beteend etest för att und ersöka om de dof ter som insekte n kan kän na är tilld ragand e, från stötand e, eller neu trala för den . Gåe nde par asitoide r und ersökt es i "Y- tub e walk ing bioassay" (Fig . 1, Pap er II), med an vindtu nnel (Fig ure 6) använd es för flyg and e par asitoid er. Vissa syntetiska beten var mycket attrakti va för par asitoid honorna (Pa per s II, III och V). Minst tre sub stanser visad e sig nöd vän dig a för att attrahera par asitoid honor (Pa per V). Parasitoid honor lockas starkas t till stockar med barkbo rrelarver. Oangrip na granstock ar var inte attraherande , men i kom bination med vissa syntetiska beten blev oangrip na stocka r lik a attraherande som granstockar med b ark borrelarver (Pa pers II och V). * "Figure" hänvisar till figur i själva avhandlingen, medan "Fig." hänvisar till figur i en bifogad artikel ("Paper"). "It was th e best of times, it was the w orst of t ime s... " Charles D icke ns ~ Acknow ledgeme nts ~ I have many people to thank for inspiratio n, help, and valuable discussio ns during these years. The first to be mentioned is my family. Thanks for all of your support ! It is a p ity that I stop working with bark beetle parasitoi ds, now that we ha ve become so good at collecting them. Most of all, I thank Claes Johanss on for always being there, sharing disappo intment and suc cess with me, I lo ve you. I wish to expre ss my gratitud e to Hans Alborn, Peter Anders on, Susanna Anders son, Gö ran Birgersson, and Brian Sullivan , who provided useful suggesti ons of improve ment of this thesis. Göran and Peter were also my enthus iastic and encoura ging scientific advisor s. I thank my co­ authors and review ers for v aluable help an d comme nts on the paper s. The whole grou p of Chemical Ecology in Göteborg has contribu ted: Göran Birgersson, I do apprecia te your impres sive scientific skills but you have also become a true friend. Pia Bergquist, my fo rmer room mate, thanks fo r discu ssions, laught er, and Pastis. Stefan Pettersson, my cu rrent room mate, showin g me how to fish piranha and climb the trees. I wish you the bes t of lu ck with your ba ts. Susan na Anderss on, keep up the g ood work! T hanks for numer ous enco uraging discussi ons and for drag ging me along to the p hysical training. Ann-Britt Wassgren, thanks for advice on GC-colu mns and for sharing your great knowled ge in chemistry. Jette Knudsen , thank s for showing me the Ecuadorian rainfore st - and some really good restau rants in Quito. Bertil Ståhl, the advice from a true botanist is sometimes very useful since we are situated at a Botanical Departm ent. Gunna r Bergström never failed in sendin g cheerfu l post cards to the research group. Former members of the research group should not be forgott en: Many warm thanks to Monica Appelg ren, Inga Groth, Per Ivarsson, and Peter Bergman who introdu ced me to chemical ecology. I would also like to thank Hans Alborn. The distance between Göteborg and Gainesville is not always that long, yo u h ave p rovide d lots of input throu gh e-mail. I really appreciate this, and your sense of humour ! All my f riend s and colleges at the Botanical departm ent, includi ng the library staff, are acknow ledged for their support and good spirits . I am also grateful to all my friends outsid e the d epartme nt. Many w arm thanks to all of yo u! Working on this project, I have become very dependent on well functioni ng compu ters and would like to specific ally expre ss my gratitude to Roger Eriksson and Pär Johansso n who taught me a lot about these machines. 30 I thank the frie ndly peo ple of the Chemical Ecology group at SLU, Alnarp. I really enj oyed s taying wit h you all! I d o not dar e to think what this thesis wo uld have been with out Peter And erson. Thanks for cheer and sup por t, helpin g me getting things into pers pect ive "if you hop e for better GC-EADs than these, then you only have you rself to blame". Spe cial thanks also to Ylva Hillbur for sharing your exp erience in electrop hysiology and unex pect ed skil ls in f ootball. Brian Sul livan, USD A Fore st Service in Pineville, Lou isiana (wit h fam ily) has become a near and dea r frien d. You have been one of my main sup port s dur ing these years. I hope we will be able to do more collaborative wor k in the fixture , and hope to get to hear the Lou isiana version of "S wa mp feve r" p laying sometime. Scientif ic meetings have been a tremendo us sou rce of insp iration, prov idin g contacts wit h lots of plea sant and help ful peo ple . Exp licitly I wo uld like to thank Mark Kenis, CABI, Swi tzerl and, for spe ciation, and for l end ing me the gui de to San Fran cisco. Many other peo ple, from diff erent plac es and cou ntries, contributed both scientif ically and socially to this thesis. Althoug h all of you are not ack now ledg ed exp licitly, plea se bear with me. I greatly app reciate you r sup por t and hop e you rea lise how muc h you meant f or this w ork. A doctoral positi on for this thesis was provid ed by the Swed ish Counc il for Forest ry and Agricultura l Research (SJFR ); and the Faculty of Science at Göteborg University. The thesis work was financi ally suppo rted by grants to me from Adle rbertska Forskn ingsfon den; Anna Ahrenbergs Fond; Helge Ax:son Johns ons Stiftel se; Fonden för Skogsv etenska plig Forsk ning; Wilhelm & Martina Lund grens Vetenskapsf ond; C.F. Lunds tröms Stifte lse; Kungl iga och Hvitfe ldtska Stipen diestif telsen; Studen t Travel Award s from the Faculty of Sc ience at Göteborg University; Stiftelse n fru Mary von Sydow (född Wijk) Donatio nsfond ; T he Royal Society of Arts and S ciences in Göteborg (KVV S); The Swedis h Royal Acade my of Science (KV A); a nd A nna & G unna r Vidfel ts Fond. -Tack alla som varit till € " ar insp iration, stöd och gläd je unde r mitt dok toranda rbete! 3 1 Refere nces Abbott I (199 3 ) Review of the ecology and control of the introd uced bark beetle Ips grand icollis (Eicho ff) (Coleo ptera: Scolytidae) in western Austral ia, 195 2-1 990 . Calm S cience 1 : 35 -46 Alborn H.T., Turli ngs T.C.J ., Jones T.H., Stenhagen G., Lough rin J.H., and Tumli nson J.H. (199 7 ) An elicitor of plant volatiles from beet armyw orm oral secretion. Scien ce 276: 9 45- 949 Arn H., Städle r E., an d Rauch er S. (197 5 ) The electroantennograph ic dete ctor - a selective and sensitive tool in the gas chromatographic analysis of insect pherom ones. Z . Naturf orsch. 30c: 722 -72 5 Berisf ord C.W. and Dahlste n, D.L. (198 9 ) Biological control of Ips grandi collis (Eichh off) (Coleop tera: Scolytidae) in Austr alia - a pre liminary evalua tion, pp 81 -9 3, in D.L. Kulhav y and M.C. Miller (eds.) Potential for Biological Control of Dendro ctonus and Ips Bark Beetles. Univ. of Texas Press, Austi n Bouc ek Z. and Heyd on S.L. (199 7 ) Pteromalidae, pp. 541 -69 2, in G.A.P. Gibson, J.T. Hube r, a nd J.B. Woolley (eds.) Annotated Keys to the Genera of Nea rctic Chalcid oidea ( Hymenopter a). NRC Research Press, Ot tawa Breer H. (199 7 ) Molecular mechanisms of phero mone recept ion in insect antennae, pp. 11 5-1 33 , in R.T. Cardé and A.K. Mink s (eds. ) Insect Pheromone Research, New Directi ons. Chapman & Hall, L ondon Bridge s J.R., Nettieton W.A., and Connor M.D. (19 8 5 ) South ern pine beetle {Dendr octonu s front alis) (Coleo ptera: Scolytidae) infesta tions withou t the bluestai n fun gus, Ceratocystis mino r. J. Econ. Entom ol. 7 8 : 3 25 -32 7 Bridge s J.R. and Moser J.C. (198 6 ) Relationship of phore tic mites (Acari: Tarsonemid ae) to the bluesta ining fungus Ceratocystis min or, in trees infeste d by southe rn pine beetle ( Dendr octonu s fronta lis) (Coleo ptera: Scolytidae). E nvir on. En tomol. 1 5 : 95 1-9 53 Burks B.D. (19 7 9 ) Pteromalidae, pp 768 -85 3 , in K.V. Krombein, P.D. Jr. Hurd, D.R. Smith, and B.D. Burks (eds.) Catalogue of Hymenopter a in America North of Mexic o I. Smithsonian Inst. Press, W ashington D .C. Byers J. (19 9 5 ) Host-tree chemistry affectin g colonizatio n of bark beetles, pp 154 -21 3, in R .T. Cardé and W .J. B ell (eds.) Chemical Ecology of In sects 2. Chapm an & Hall, New Y ork Bäckm an A.-C. (199 9 ) Olfacto ry Communicati on in the Codl ing Moth, Cydia pomone lla L. PhD disserta tion. Depar tment of Plant Protection Sciences, Swedis h University of A gricultu ral Sciences, A lnarp Charnov E.L., H artogh los d en R.L., Jone s W.T., an d A ssem van den J. (19 8 1 ) Sex r atio evolutio n in a variable environment. Natu re 289: 27 -3 3 Dahlsten D.L . (1982 ) Relationships betwee n bark beetles and their natural enemies, pp 14 0-1 82 , in J.B. Mitton and K.B. Sturg eon (eds.) Bark Beetles in North American Conife rs - a System for the Study of Evolu tionary Biology. Univ. of Texas Press, A ustin De Moraes C.M. and Mescher M.C. (199 9 ) Interactions in entomology: Plant- parasit oid int eraction in tritroph ic systems. J . Entom ol. Sei. 3 4 : 3 1-3 9 32 Dicke M., Sabelis M.W., T akabay ashi J., Bruin J. , and Posthu mus M .A. (1990) Plant strategies of manipu lating predato r-prey interactions throug h allelochemicals: pr ospect s for applica tion in pest control. J . Chem . Ec ol. 1 6 : 309 1-3 11 8 Dixon W.N. and Pa yne T.L. (1980) Attraction of ent omop hagou s and assoc iate insects of the southe rn pine beetle to beetle- and host-tree produce d volatiles. J. Ga. Entom ol. Soc . 1 5 : 37 8-3 89 Francesc hi V.R., Kroke ne P., Krekli ng T., and Christiansen E. (2000 ) Phloem parench yma cells are involved in local and distan t defens e respon ses to fungal inocu lation or bark-b eetle attack in N orway spru ce (P inaceae) . Am . J. Bot. 8 7 : 314 — 3 26 Graham M.W.R. de V. (196 9 ) The Pteromalidae of North-Western Europe (Hyme noptera : Chalcidoide a). Bull, of the Brittisk Muse um (Natura l History) , Lond on, Su ppl. 16: 412-^427 Grissell E.E. and Schauff" M.E. (199 7 ) Superfam ily Chalcidoide a, pp. 45 -11 6 , in G.A.P. Gibson, J.T. Huber, and J.B. Woolley (eds.) Annotated Keys to the Genera of Nearctic Chalcidoidea (Hyme noptera ). NRC Research Press, Ottawa Gullan P.J. and Cranston P.S. (199 4 ) The Insects: An outline of entomology. Chapm an & Hall, Londo n Hedqvist K.-J. (19 9 8 ) Bark beetle enemies in Swede n II. Braconidae (Hyme nopter a). En otom ol. S cand. Supp l. 52: 1- 87 , illustr. Herard F. and Mercadier G. (199 6 ) Natural enemies of T omicu s piniper da and Ips acumi natus (Col .; Scolytidae) on Pinus sylvestris near Orleans, France : Tempora l occu rrence and relative abunda nce, and notes on eight predato ry species . E ntom ophaga 4 1 : 1 8 3- 210 Hunt D.W.A., Bord en J.H., Lindg ren B.S., and Gries G. (198 9 ) The role of autoo xidat ion of «r-pinene in the pro ductio n of phero mones of De ndroc tonus pondero sae (Co leopt era: Sco lytida e). Can. J . For. Res. 1 9 : 12 75- 128 2 Kenis M. (199 6) Facto rs affect ing sex ratio in rearing of Coeloide s sordid ator (Hym. : Braco nidae ). Ento mopha ga 4 1 :2 17-2 24 King B.H. (198 7 ) Offsp ring sex ratios in parasito id wasps . Q . Rev. Biol. 62: 367 -39 6 Krieger I. and Breer H. (199 9 ) Olfacto ry receptio n in invertebrates. Scien ce 286: 7 20-723 Krokene P. and Solheim H. (199 6 ) Funga l associates of five ba rk be etle spe cies colonizi ng Norw ay sp ruce. C an. J . For . Res. 26: 21 15-21 22 Krüger K. and Mills N.J. (19 90) Observations on the biology of three parasitoid s of the spruc e bark beetle, Ips typographus (Col ., Scolytidae): Coeloid es bost rychorum , Dendro soter midde ndorff ii (Hym ., Braconidae) and Rh opalicus tutela (Hym. , Ptero malida e). J . Ap pl. En t. 1 10: 2 81 -29 1 LaSalle J. and Gauld I.D. (19 9 1 ) Parasitic Hymenopter a and the biodive rsity crisis. Redia 7 4 : 3 1 5 -33 4 Lenteren van J.C. (198 1 ) Host discrim ination by paras itoid s, pp 15 3-1 79, in D.A. Nordlu nd, R.L. Jones, and W.J. Lewis (eds. ) Semiochemicals, their role in p est control. Wiley & sons, New Yo rk 3 3 LeufVén A. (199 1 ) Role of microorganisms in spruc e bark beetle-conif er interactions, pp. 467 -48 3 , in P. Barbosa, V.A. Krischik, and C.G. Jones (eds.) Microbial Mediat ion of Plant-Herbivore Interactions. Wiley & Sons, New York LeufVén A. and Birgersson G. (198 7 ) Quantit ative variation of differe nt monoterpen es aroun d galleries of Ips typographies (Co leopte ra: Scolytidae) attacki ng Norway spruc e. Can. J. Bot. 6 5 : 10 38- 104 4 LeufVén A ., Bergström G., and Falsen E. (198 8 ) Oxyg enated monoterpen es produce d b y yeasts, isolated from Ips typographies (Col eopter a: Scolytidae) and g rown in phloe m medium. J . Chem . Ecol. 1 4 : 35 3-3 62 Millar J.G. and Sims J.J. (19 9 8 ) Prepara tion, cleanup and prelim inary fractio nation of extrac ts, pp. 1- 31 , in J.G. Millar and K.F. Haynes (eds.) Method s in Chemical Ecology. Chemical Methods. Klu wer, N orwel l Mills N.J. (19 8 3 ) The natural enemies of scolytids infesti ng conifer bark in Europe in relation to the biological control of De ndroct onus spp. in Canad a. Biocontr ol N ews Inf . 4: 305 -32 8 Mills N.J. ( 19 9 1 ) Searching strategies and attack rates of pa rasitoids o f th e ash bark beetle ( Lepe risinu s varius) and its relevance to biological control. E col. Entomo l. 1 6 : 461-^170 Mills N.J. (19 9 4 ) Parasitoid guilds: Defini ng the struct ure of the parasit oid communitie s of endop terygote insect hosts. E nviro n. Entom ol. 23: 1066 1083 Mills N.J. and Schlup J. (19 8 9 ) The natural enemies of Ips tyographus in Central Euro pe: Impac t and potential use in biological control, pp . 131 -14 6 , in D.L. Kulh avy and M.C. Miller (eds.) Potential for Biological Control of Dendr octon us and Ips B ark Be etles. Univ. of Te xas P ress, A ustin Mills N.J., Krüger K., and Schlup J. (19 9 1 ) Short-range host location mechanisms of bark beetle p arasitoids. J . Ap pl. Ent . 1 1 1 : 3 3- 43 Moore G.E. (197 2) South ern p ine beetle mortality in North Carolina caused by parasites and preda tors. E nviron. Entomo l. 1 : 58 -65 Nordlu nd D.A. and Lewis W.J. (19 7 6 ) Terminology of chemical releasing stimuli in intraspecif ic and inte rspecifi c interactions. J . Chem . Eco l. 2: 2 11 - 220 Payne T.L. (198 0) Life history and habits, pp 7- 28 , in R.C. Thatcher, J.L. Searcy, J.E. Coster, an d G.D . Hertel (eds.) The Southern Pine Beetle. USDA Forest Se rvice Science Ed ucation A dminist ration Technical Bulletin 1631 Pawlisz yn J. (19 9 7 ) Solid Phase Microextractio n, theory and practi ce. Wiley- VCH, Inc. N ew York Quick e D.L.J. (19 9 7 ) Parasitic Wasp s. Chap man & Hall, L ondon Richerson J.V. and Borde n J.H. (1972) Host findin g behavior of Coeloides brunne ri (Hym enop tera: B raconida e). Can. Entom ol. 104: 1 235 -12 50 Richerson J.V. , Bord en J.H., and Hollingdale J. (19 7 2) Morph ology of a unique sensillum placodeu m on the antennae of Coeloides brunn eri (Hym enopt era: B raconidae). C an. J. Z ool. 50: 90 9-9 13 34 Rutled ge C.E. (199 6 ) A survey of identifi ed kairomo nes and synomones used by insect paras itoid s to locate and a ccept their hosts. Chem oecology 7 : 12 1— 13 1 Samson P.R. (198 4 ) The biology of Roptrocerus xylop hagorum (Hym .: Torymidae), with a note on its taxo nomic status. E ntomoph aga 29: 2 87- 298 Samson P.R. and Smibert J. (19 8 6 ) Preliminary studies on the efficac y and establishment of Roptrocerus xylopha gorum (Hym .: Torymidae), a parasit oid of Ips grand icollis (Col .: Scolytidae) , in Austr alia. Entom ophaga 3 1 : 17 3-1 82 Slifer E.H. (1970 ) The structu re of ar thropo d ch emoreceptors, pp. 121 -14 2, in R.F. Smith and T.E. Mittler (eds.) Annual Review of Entomology. Ann. Rev. Inc., Palo Alto Smith M.T., S alom S.M., and Payne T.L. (199 3 ) The southern pine bark beetle guild: an historical review of the research on the semiochemical-based communica tion system of the five princip al specie s. Virginia Agric ultural Experim ent Station Bulletin 9 3 -4 . Virginia Polytechnic Institu te and State University. Blacksbu rg Spessiv tseff P. (1922 ) Bestäm ningstabell över svensk a barkbo rrar. Medde landen från Statens Skogsf örsöks anstalt, häfte 19, nr 6. Centraltrycke riet, Sto ckhol m Sulliva n B.T. (199 7 ) The chemical ecology of host habitat location by larval parasi toids of the southe rn pine beetle, Dend roctonus front alis Zimm ermann: Olfac tory cues and their possibl e source s. PhD dissert ation. Departm ent of Entomology, Univer sity of Georgia, Athe ns Sulliva n B.T., Berisford C.W., and Dalusky M J . (199 7 ) Field respo nse of southe rn pin e beetle parasi toids to some natural attractants. J. Chem . Ecol . 23: 83 7-8 56 Sullivan B.T., Seitmann K.C., a nd Beri sford C .W. (199 9 ) A simp le continuou s- rearing technique for the bark beetle parasi toid Roptrocerus xylop hagorum (Ratze burg). J . Ent omol . Sei. 3 4 : 2 60 -26 4 Takäcs S., Gries G., and Gries R. (199 7 ) Semiochemical-med iated location of host habitat by Apant eles carpatus (Say) (Hym enopte ra: Braconidae) , a parasitoid of clo thes moth larvae. J . Chem. Ecol. 23: 4 59- 472 Tobias V.l., Belokoby lskij S.A., and Kotenk o A.G. (198 6 ) Guid e to the insects of Europea n part of the USSR Hymenopte ra. Opre delite li po faune SSSR 3 ( 4 ) : 1 -50 0, illustr . Turling s T.C.J ., Tuml inson J.H., Eller F.J., and Lewis W.J. (19 9 1 ) Larval- damag ed plant s: sourc es of volatile synomones that guide the parasit oid Cotesia marg inive ntris to the micro-habitat of its host. Ento mol. Exp. Appl . 5 8 : 75 -82 Waage J.K. (198 2) Sex ratio and popula tion dynam ics in natura l enemies - some possib le interactions. A n n . A ppl. Bi ol. 101: 15 9-1 64 , illus tr. Waage J.K., Carl K., Mills N.J., and Greathead D.J. (19 8 5 ) Rearing entomophag ous insects, pp 45 -6 6, in P. Singh and R.F. Moore (eds.) Handbo ok of Insect Rearing, Vol . 1. Elsevier, Amst erdam Vet L.E.M., Lewi s W.J., and Cardé R.T. (199 5 ) Parasitoid forag ing and learning, p p. 65 -10 1, in R.T. Cardé and W.J. Bell (eds.) Chemical Ecology of Insec ts 2. Chapm an & Hall, N ew York Vinson S.B. (198 5 ) The behavior of p arasitoids, pp. 417 -46 9 , in G.A. Kerkut and L .I. Gilbert (eds.) Compr ehensive Insect Physiology, Biochemistry, an d Pharmacology, Vo l. 9. Pergamon Press, New Y ork Vinson S.B., B in F., and Vet L.E.M. (199 8 ) Critical issu es in host selection by insect p arasitoids. Biol. C ontro l 1 1 : 7 7- 78 Yang Z. (19 8 7 ) A prelim inary surve y of paras itic wasp s of Dend roctonus arman di Tsai et Li (Coleo ptera, Scolytid ae) in Qinlin g Moun tains (China) with descri ption of three new specie s and a new Chinese record (Hym enopt era, Pt eromalidae). E ntomota xonomia 9 : 17 5-1 84 , illus tr. På grund av upphovsrättsliga skäl kan vissa ingående delarbeten ej publiceras här. För en fullständig lista av ingående delarbeten, se avhandlingens början. Due to copyright law limitations, certain papers may not be published here. For a complete list of papers, see the beginning of the dissertation. *Dc$&xlfc efC&t & "Su tuiett- 'Vctea Atade Mi' f3o&&ù td&U s4 "S 2 0 0 0 GÖT EBORG UNIVERSITY Fac ulty of Science