6989

Journal of Chemical Ecology. Vol. 18. No. 10. 1992

ORIENTATION OF croceipes (: ) TO GREEN LEAF VOLATILES: DOSE-RESPONSE CURVES

D.W. WHITMANI.* and FJ. ELLER2

I Depanmem ofBiological Sciences ll/inois State Universirv Normal. Illinois 6176j 2Nationol Cemer for Agricultural Utilization. USDA-ARS Peoria, Illinois 61604

(Received February 25. 1992; accepted May 28. 1992)

Abstract-Female Microplitis craceipes wasps were tested in a wind tunnel for their ability to orient to various concentrations of eight different green leaf volatile (GLV) substances [hexanal, (£)-2-hexenal. (£)-2-hexen-I-ol. (Z)-3­ hexen-I-ol. (£)-2-hexenyl acetate, (Z)-3-hexenyl acetate, (Z)-3-hexenyl pro­ pionate, and (Z)-3-hexenyl butyrate]. Overall. the esters elicited the greatest percentage of successful orientation flights. the alcohols elicited an intenne­ diate response, and the aldehydes elicited a low response. The semilog dose­ response curves were generally hill-shaped with high responses at medium release rates and low responses at high or low release rates. For the aldehydes. positive responses occurred at all GLV release rates between 0.01 and 100 nl/min. For some alcohols and esters. positive responses occurred at release rates as low as I pi/min and as high as I ILl/min. These data show that M. craceipes wasps are strongly attracted to GLVs and are capable of orienting to GLV concentrations that would occur in nature when a caterpillar feeds on a green leaf. Hence. in nature. GLVs may be important clues. enabling M. croceipes to locate their hosts.

Key Words-Green leaf volatile. semiochemical. synomone. volatile attrac­ tant. Microplitis craceipes. Hymenoptera. Braconidae. Heliothis zea. Lepi­ doptera. NoclUidae. biological control. tritrophic. host location. parasitoid behavior.

INTRODUCfION

Many wasp parasitoids seek out and oviposit into plant-feeding caterpillars. Volatile semiochemicals originating from the caterpillars or their food plants

·To whom correspondence should be addressed.

1743

0098'{)3311921100ll-1743S06.SOIO © 1992 Plenum Publish..g Corporation 1744 WHITMAN AND ELLER serve as imponant cues for these wasps. enabling the wasps to find their hosts (Nordlund et al .. 1988: Whitman, 1988a; Lewis and Manin, 1990: McAuslane et al .• 1991; Turlings et al .• 1991a,b). Among the volatile substances released from damaged plants are the six-carbon alcohols. aldehydes, and derivative esters that are collectively known as green leaf volatiles (GLVs) (Hedin et al .. 1973; Saijo and Takeo, 1975; Wallbank and Wheatley, 1976; Visser et al., 1979: Hamilton-Kemp et al., 1988, 1989; Tollsten and Bergstrom. 1988: Her­ nandez et al., 1989; Lwande et al., 1989). Recently, we suggested that cater­ pillars feeding on green leaves cause plants to release GLVs and that parasitic wasps orient to these substances and hence to their caterpillar hosts (Whitman and Eller, 1990). In this tritrophic interaction, the GLVs would serve as plant­ to-parasitoid synomones; their release would benefit both the plant and the par­ asitoid but would be detrimental to the caterpillar. In this paper we examine the ability of female (Cres­ son) wasps to orient to various concentrations of individual GLVs in a wind tunnel. M. croceipes is a braconid wasp that attacks the corn earworm caterpillar. Helicoverpa (= Heliothis) zea (Boddie) (Lepidoptera: Noctuidae). We show that the GLVs vary in attractiveness, that wasps can successfully orient to GLV concentrations ranging over five orders of magnitude, that there is variability in the low and high response thresholds of wasps to various GLVs, and that the dose-response curves for various GLVs differ more in magnitude and capacity than in shape. We also show that wasps can orient to release rates similar to those produced naturally when a caterpillar feeds on a green plant.

METHODS AND MATERlALS

Insects. Larval M. croceipes were reared on at the Biology and Population Management Research Laboratory at Tifton, Georgia. The Heliothis were cultured on artificial diet (Bunon, 1969). Adult parasitoids were kept communally and allowed to mate in Plexiglas cages (30 x 30 x 17 cm) at 28°C, 50-70% relative humidity, a 16:8 light-dark photoperiod, and provided daily with fresh water and honey (Lewis and Bunon, 1979). Adults were isolated from plant and host materials until tested. Flight Tunnel. The ability of wasps to orient to GLVs was tested in a 75 x 75 x 200-cm laminar-flow Plexiglas wind tunnel (wind speed: 31 cm/sec; temperature: 26-27°C; relative humidity: 40-80%; wind tunnel light provided by four overhead 80-W fluorescent bulbs). An electric fan, placed outside the building, pulled air through the tunnel at a constant rate and vented it outside, eliminating contamination from recycled air. Chemicals. Various GLVs were obtained from Sigma Chemical Co. and Aldrich Chemical Co. Following the tests, chemicals were assayed by GCMS; purity'ranged from 93 to 99%. ORIENTATION OF BRACONID TO LEAF VOLATILE 1745

Preparation of Water Extract of H. zea Frass. Ten grams of fresh moist H. zea frass (collected from caterpillars raised on cowpea seedlings) was thor­ oughly blended with 100 ml of distilled water. The resultant slurry was filtered through Whatman No. 1 filter paper, producing a green-colored filtrate. This filtrate was kept frozen and served as the stock extract throughout the tests. To produce the dried extract, approximately 1 ml of stock extract was unfrozen and shaken. Then 30 ILl of extract was collected into a micropipet and sponed onto the center of a llO-cm disk of Whatrnan No. 1 filter paper. After the l-cm diameter moist area dried (-7 min.), the filter paper was used to stimulate wasps. Experimental Design. Three- to eight-day-old female M. croceipes wasps were tested for their ability to orient in a wind tunnel to various concentrations of eight different GLVs (see Figure 1 below). Eight to 10 different release rates were tested for each GLV, resulting in 73 different treatments. Thiny or more wasps were flown to each treatment; each wasp was tested only once. Eight control treatments were also employed: Sixty wasps were flown to a blank control consisting of a l-cm2 paper disk and 15-30 wasps were flown to each of seven different hexane release rates, which encompassed the full range of GLV release rates tested. Wasps were individually tested, allowing individual responses to be recorded. All trials were single choice tests and were balanced for compound, dose, day, and time of day. The odor source was placed in the flight chamber 13 cm above the chamber floor, along the longitudinal midline of the chamber, and 20 cm downwind from the up wind end of the chamber. Immediately prior to testing, a wasp was stimulated for 30 sec by contact with dried water extract of H. zea frass. The wasp was then placed into a clean 4-ml shell vial directly 100 cm downwind from the odor source. Wasps usually walked upwards to the opening of the shell vial, encountered the lower ponion of the odor plume, faced into the wind, and after 5 sec to 5 min took flight. Wasps that did not take flight in 5 min or wasps incapable of nonnal flight were discarded. Each wasp was given two successive chances to fly to the odor source and then was discarded. If a wasp flew toward and landed on the odor source or a paper target placed 3 mm from the odor source, a positive response was recorded and the trial was ended. If the wasp took flight, but failed to land on the odor source or target, a negative response was recorded. In the above trials, wasps were released 100 cm downwind from the odor source. To examine the effects of distance, a second set of trials was run with wasps released only 15 cm downwind from the odor source. Release and Calibration of Green Leaf Volatiles. Different methods were used to obtain low, medium, and high GLV release rates. For low release rates, one end of a 5-, 10-, 20-, 50-, or lOO-ILI-capacity glass micropipet (Fisher Scientific, Pittsburgh, Pennsylvania) was flame sealed, and a measured amount of GLV was placed into the open end. The pipet was then spun in a centrifuge 1746 WHITMAN AND ELLER to pull the liquid down to the sealed end ofthe tube. The resulting tube contained a measured amount of GLV at the sealed end. After a 24-hr equilibration period. a single pipet was placed venically in the wind tunnel with the open end up. providing a relatively constant release rate of GLV. By using various sized micropipets, various release rates were produced. A standardized release rate among same-sized pipets was assured by maintaining a standard distance of 110 mm between the GLV meniscus and the pipet opening. For medium release rates, 0.5- or l-,.d-eapacity glass micropipets (Drum­ mond Scientific, Broomall, Pennsylvania) or 5-, 10-, 20-, 50-, or lOO-,uI-capac­ ity micropipets (Fisher Scientific) were filled with 10 mm of a selected green leaf volatile substance. A filled pipet was suspended crosswise in the wind tunnel, but at a 45 0 angle from venical. This arrangement assured that the lower meniscus of test liquid always remained just at the lower orifice of the pipet. producing a uniform evaporative surface. If the pipet was placed venically or horizontally, uniform evaporation was not achieved; with a venical pipet, a droplet oftest liquid sometimes emerged from the tube, increasing volatilization. With a horizontal pipet, capillary action sometimes pulled the liquid into the center of the tube, decreasing volatilization. Higher release rates were achieved by insening an 8-mm2 or 30-mm2 tri­ angular filter paper wick into the open end of a lOO-,u1 micropipet containing 20 mm of GLV. Pipets were positioned at a 45 degree angle from venical, with the tube and surface of the wick perpendicular to the air flow. For all pipets, a l-em2 disk of white filter paper served as a landing target. The disk was attached to the micropipet in a venical position so that the perim­ eter of the disk was 3 mm from the releasing end of the tube. With this arrange­ ment, a wasp orienting to volatiles emerging from the end of the micropipet usually landed on the venical paper disk. Medium and high release rates were determined directly by measuring the change in volume of the internal liquid when the pipets were kept in the wind tunnel for 24 hr. The release rates from the low-release pipets were determined indirectly as follows: The eight synthetic GLVs were loaded individually into 10-, 20-, 50-, and lOO-,u1 micropipets as described above. After equilibrating in a fume hood for 24 hr, the pipets were placed venically in a volatile collection system for 24 hr. The volatile collection system consisted of a 20-cm x 2.2-cm-ID glass tube sealed on each end with a cork stopper. The bottom cork held a prefilter (7-em x 4-mm-1D glass tube) packed with 6 mm of Super Q (80-100 mesh; Alltech Associates, Inc., Deerfield, Illinois) held between a 325-mesh stainless-steel screen and a glass-wool plug. The top cork held a similar filter with 4 mm of Super Q to collect volatiles. Air was drawn up through the tube at a flow rate of 130 ml/min. Collected volatiles were extracted from the down­ wind filter with 240 ,ul methylene chloride. Ten microliters of a 250 ngl,ul ORIENTATION OF BRACONID TO LEAF VOLATILE 17~7 solution (i.e.. 2500 ng) of l-octen-3-o1 was added to each filter extract as an internal standard to quantify collected volatiles. Volatile collections were ana­ lyzed using a Hewlen Packard 5890 Series II gas chromatograph (Hewlett Pack­ ard Co., Avondale. Pennsylvania) equipped with a fused silica HP-5 column (0.17 ~m. 25 m x 0.32 mm 10) (Hewlen Packard), a flame ionization detector. and a Spectra-Physics SP4400 integrator (Spectra-Physics, Inc., San Jose. Cal­ ifornia). Injections were made in the splitless mode and after 0.60 min changed to the split mode. The temperature program was; 40°C for 3 min. then lOoC increase/min to llO°C. The injector and detectortemperatures were l70°C and 250°C, respectively. Three collections were made of each compound-pipet combination.

RESULTS

Dose-response curves for female M. croceipes wasps orienting to various individual GLVs in a wind tunnel are shown in Figure 1. The overall response ofwasps appeared to vary according to the functionality ofthe compound tested. The aldehydes [hexanal and (E)-2-hexenal] were not very anractive and elicited few successful orientations. The alcohols (E)-2-hexen-I-ol and (Z)-3-hexen-l-ol elicited moderate responses, and the esters (E)-2-hexenyl acetate, (Z)-3-hexenyl acetate, (Z)-3-hexenyl propionate, and (Z)-3-hexenyl butyrate were more anrae­ tive and elicited the greatest number of successful orientations. In contrast, no wasps oriented to the blank or hexane controls. The dose-response curves for the eight tested compounds tended to be similar in shape when displayed on a sernilog plot (Figure 1). The curves were generally hill-shaped, with high responses at medium release rates and low responses at either very high or very low release rates. Maximum responses generally fell at release rates between 0.1 and 10 nl/min. The data show that the range of anractive doses varied among the different GLVs. With aldehydes, wasps responded to release rates that ranged across four orders of magnitude; above and below this active range, wasps failed to orient. For each of the alcohols and esters, positive responses were observed for over five orders of magnitude. Note, however, that for the alcohols and esters, we did not test above or below this range (it is difficult to reliably produce such extreme release rates). Hence, wasps may be able to orient over a greater range of release rates than shown here. For example, 30% of wasps successfully oriented to the lowest (Z)-3-hexenyl butyrate release rate tested (0.5 pi/min) and 32 % of wasps oriented to the highest (Z)-3-hexen-I-ol release rate tested (1 ~I/min). Our data also indicated that the low threshold for response varied for dif­ ferent GLVs. For (Z)-3-hexenyl butyrate, some wasps oriented to release rates as low as 0.5 pi/min. For hexanal, however, this figure was only 10 pi/min. 1748 WHITMAI; AI;D ELLER

100 hexanal (E)-2-h9xenal 80

60

40 • 20

0 100 Qi (E)-2-hexen-1-01 {Z)-3-hexen-l-01 ~ 80 ell I- 60 Cii • • en • r:: 40 • •• '6 r:: 20 3 • •• • (Jl 0 • Q. (Jl 100 ell (E)-2-hexenyl acetate (Zj-3-hexenyl acetate := 80 15 Q) 60 • • en • ~ r:: 40 •• Q) u ~ 20 Q) • 0.. • 0 100 {Z)-3-hexenyl propionate (Zj-3-hexenyl butyrate 80

60

40 •• •

20 0+-....,...... -...... ,..-,...... -,...... ,,.--...,...... -1 -4 -3 ·2·' 0 2 3 4 -4 ·3 -2·' 0 3 4 Log Release Rate (nl/min)

FIG. 1. Dose-response curves for female Microplitis croceipes wasps flying to various green leaf volatiles in a wind tunnel. Each point represents the results from 30 or more wasps. Curves generated and drawn by computer as second- or third-order polynomials. Each wasp was given two chances to successfully orient. ORlENTATION OF BRACONID TO LEAF VOLATILE 1749

TABLE I. ORlENTATION RESPONSE OF FEMALE Micropliris croceipes WASPS TO NEAR AND DISTANT SOURCES OF INDIVIDUAL GREEN LEAF VOLATILES

Percentage" of wasps successfully orienting to odor source from Release rate Volatile (pliminl 100 em" 15 em'

(Z)-3-hexen-I-01 3.2 3 80 (E)-2-hexenyl acetate 4.7 3 60 (Z)-3-hexenyl butyrate 1.9 40 80 (Z)-3-hexenyl butyrate 3.0 47 87

"Each wasp given two chances to orient. "Thirty or more wasps tested to each volatile. 'Fifteen wasps tested to each volatile.

Wasps responded differently to high versus low release rates. At high release rates. wasps often oriented well when between 30 and 100 cm downwind from the odor source. However, when the wasps flew closer (between 5 and 20 cm from the odor source), they often turned sharply away and made no further attempts to enter the odor plume. In contrast, with low release rates wasps oriented poorly when 50-100 cm downwind from the odor source. but flew well when closer. When these wasps exited the odor plume, they often looped down­ wind or performed transverse zigzag flights until they reentered the plume. For example, when wasps were tested with (Z)-3-hexenyl propionate at a release rate of 10 pi/min, all 18 wasps that approached within 10 cm of the odor source landed, but at 13 nllmin, only 12 of the 22 wasps that approached within 10 cm landed. Likewise. at a release rate of 9 pllmin of (Z)-3-hexen-l-01, all 13 wasps that approached within 10 cm of the odor source landed, but at 45 nil min, only 12 of the 25 wasps that approached within 10 cm landed. Table 1 shows that wasps were better able to orient to the low release rates of certain GLVs when the wasps were released close to the odor source.

DISCUSSION

Our results demonstrate that female Microplitis croceipes wasps can detect and orient to each of the eight GLVs tested. However, individual GLVs varied in attractiveness. Six-earbon aldehydes were the least attractive, with no responses above 30%. Six-earbon alcohols were moderately attractive, with some doses eliciting up to 55% successful flights. Eight- to lO-earbon esters were the most attractive, with 60-70% successful flights at some doses. 1750 WHITMAN AND ELLER

Our results also suggest that wasps can orient to extremely low doses of certain GLVs; release rates of approximately 2 pI/min for (E)-2-hexen-l-ol. (Z)-3-hexenyl propionate, and (Z)-3-hexenyl butyrate were attractive to some wasps. The fact that some wasps oriented to these low doses suggests that even lower release rates might be attractive. The low release rates to which M. croceipes responded are comparable to GLV release rates that occur when a caterpillar feeds upon a green leaf. For example, 3-5 pllmin ofGLVs are released when a Heliothis zea caterpillar feeds upon a cowpea leaf (Whitman and Eller, 1990), suggesting that M. croceipes are capable of responding to the levels of GLV that would be produced when a caterpillar feeds on a green leaf in nature. Undamaged plants release much lower levels of GLVs (Dicke et al., 1990a; Whitman and Eller, 1990; Turlings et al., 1991a,b). This difference in GLV release rates may allow M. croceipes to distinguish between undamaged and caterpillar-attacked plants; wasps might only orient to the higher release rates that occur when a caterpillar feeds on a leaf. There is additional evidence to suggest that M. croceipes might orient to even lower GLV release rates than tested here: The cross-sectional area of an odor plume increases and the concentration of volatiles in that plume decreases with distance from the odor source. In our tests, M. croceipes were released 100 cm downwind, where the concentration of test volatile was relatively low. When wasps were released much closer to the odor source (15 cm distant), where odor concentrations were greater, wasps oriented much better (Table 1). This is an important point because in the field M. croceipes hunt by flying slowly 5-10 cm downwind from plants. At such a close range, the wasps should be able to detect very low levels of GLVs, perhaps lower than we have tested here. In our tests, we used individual GLVs, but in nature, a caterpillar feeding on a plant would produce a complex blend of volatiles (Whitman and Eller, 1990; Turlings et al., 1990). A survey of previous literature (Drost et al., 1986, 1988; Eller et al., 1988; Prevost and Lewis, 1990) shows that when M. croceipes wasps were tested in wind tunnels to the full complement of plant and caterpillar odors, 45-90% of the wasps made successful orientation flights. This compares favorably with the highest responses (60-70%) recorded from our trials using individual compounds. Because both the sensitivities and behavioral responses of to odor blends may exceed the sum of component thresholds and responses (Laska et al., 1990), we suspect that combinations of two or more individual GLVs might elicit greater responses from the wasps. Information about the sensory and neural mechanisms ofthe wasps' response to GLVs can be derived from the various GLV dose-response curves. All eight curves were fairly similar in shape, suggesting a similar stimulus-response mech~ism. Generally, the height of the curve was proponional to its length; ORlENTATION OF BRACONID TO LEAF VOLATILE 1751 compounds that elicited high responses were also active over a greater range. For each GLV, there was usually one concentration that produced a maximum flight response. Above or below that release rate, the orientation rate dropped. The decreased flight responses that were observed at low release rates probably occurred because GLV concentrations were insufficient to reach sensory or neural thresholds. Decreases in flight response at high release rates were possibly due to stimulus overload: as wasps approached the odor source, the GLV concen­ tration increased to a point where it became repellent. This was shown by the fact that, when exposed to high GLV release rates, wasps often oriented well when far from the odor source, but turned sharply away when within 5-20 em of the source. It should be noted, however, that such high release rates are anificial and would seldom occur in nature. When the dose-response curves are plotted on numerical, instead of semi­ log scales, the left-hand sides of the curves are very steep, and the maximum responses lie close to the low dose thresholds, suggesting that the maximum responses of the wasps have evolved to occur near their low sensory thresholds. The dose-response curves reponed here (Figure 1) clearly show that the response ofM. croceipes wasps to GLVs varies with the dose. This underscores the great imponance of always examining dose effects when investigating semi­ ochemical interactions. Failure to consider dose effects when comparing semi­ ochemicals could easily lead to erroneous conclusions about their absolute or relative activities. Many authors have suggested that carnivorous natural enemies use plant chemicals to locate their herbivorous hosts (Whitman, 1988a,b; Ding et aI., 1989; Navasero and Elzen, 1989; Sheehan and Shelton, 1989; Dicke et al., 1990a; Manin et al., 1990; McAuslane et al., 1991; Turlings et aI., 1991a,b). Others have hypothesized that cenain plant volatiles may have evolved to serve a tritrophic communicative function (i.e., plants release natural enemy attractive substances in direct response to herbivore attack) (Whitman 1988a,b: Dicke and Sabelis, 1989: Dicke et aI., 199Oa,b; Turlings et al., 1990, Whitman and Eller, 1990). The release of GLVs is undoubtedly advantageous for M. croceipes wasps because it allows the wasps to locate hosts. If it is shown that plants benefit when their attendant herbivores are destroyed by natural enemies that respond to these GLVs, then these chemicals should be regarded as plant-to­ parasitoid synomones. Acknowledg~ms-We thank W.1. Lewis and the staff at the IBPMRL. USDA for advice and assistance.

REFERENCES

BURTON, R.L. 1969. Mass rearing the com earworm in the laboratory. U.S. Dept. Agric. Tech. Bull. ARS Ser. 33-134. 8 pp. DICKE. M., and SABEUS. M. 1988. How plants obtain predatory mites as bodyguards. Neth. J. Zool. 38: 148-165. 1752 WHITMAN AND ELLER

DICKE. M.. VAN BEEK. T.A.. POSTHUMUS. M.A.• BEN DoM. N.. VAN BOKHOVEN. H.. and DE GROOT. A.E. 199Oa. Isolation and identification of volatile kairomone that afrects acanne predator-prey interactions. Involvement of host plant in its production. J. Chem. Em/. 16:381­ 396. DICKE. M.. SABELIS. M.W.. TAKABAYASHI. 1.. BRUIN. J.• and POSTHUMUS. A. 1990b. Plant strat· egies of manipulating predator-prey interactions through allelochemicals: prospects for appli· cation in pest control. J. Chem. Ecol. 16:3091-3118. DING. D .. SWEDENBORG. P.D.• and JONES. R.L. 1989. Chemical stimuli in host-seeking behavior of Macrocenrrus grantiii (Hymenoptera: Braconidae). Ann. Enromo/. Soc. Am. 82:232-236. DROST. Y.e.. LEWIS. WJ.. ZANEN. P.O.• and KELLER. M.A. 1986. Beneficial Anhropod behavior mediated by airborne semiochemica1s. 1. Flight behavior and influence of preflight handling of Micropliris croceipes (Cresson). J. Chem. Ecol. 12:1247-1262. DROST. Y.C.. LEWIS. WJ.• and TUMLINSON. J.M. 1988. Beneficial anhropod behavior mediated by airborne semiochemica1s. V. Influence of rearing method. host plant. and adult experience on host-searching behavior of Micropliris croceipes (Cresson). a larval parasitoid of Heliorhis. J. Chem. Ecol. 14:1607-1616. ELLER. FJ.• TUMLINSON. J.H.• and LEWIS. WJ. 1988. Beneficial anhropod behavior mediated by airborne semiochemica1s: Source of volatiles mediating the host-location flight behavior of Micropliris croceipes (Cresson) (Hymenoptera: Braconidae). a parasitoid ofHeli01his zea (Bod­ die) (Lepidoptera: Noctuidae). Environ. Enromol. 17:745-753. HAMILTON-KEMP. T.R.. ANDERSEN. R.A.• RODRIGUEZ. J.G.. loUGHRIN. J.H.. and PATIERSON. e.G. 1988. Strnwbeny foliage headspace vapor components at periods of susceptibility and resistance to Terranychus unicae Koch. J. Chem. Ecol. 14:789-796. HAMILTON-KEMP. T.R.• RODRIGUEZ. J.G.• ARCHBOLD. D.D.• ANDERSEN. R.A.• loUGHRIN. J.H.. PATIERSON. C.G.• and loWRY. S.R. 1989. Strnwbeny resistance to Terranychus unicae Koch: Effects of flower. fruit. and foliage removal-eomparisons of air- vs. nitrogen-entrained volatile compounds. J. Chem. Ecol. 15:1465-1473. HEDIN. P.A.. THOMPSON. A.C.. and GUELDNER. R.C. 1973. The boll weevil-cotlon plant complex. Toxicol. Environ. Chem. Rev. 1:291-351. HERNANDEZ. H.P.• HSIEH. T.e.-Y.. SMITH. C.M.. and FISCHER. N.H. 1989. Foliage volatiles of two rice cultivars. PhyTOchemisrry 28:2959-2962. LASKA. M.• HUDSON. R.• and DISTEL. H. 1990. Sensitivity to biologically relevant odours may exceed the sum of component thresholds. Chemoecology 1:139-141. LEWIS. WJ.• and BURTON. R.L. 1979. Rearing Micropliris croceipes in the laboratory with Heliorhis zea as hosts. J. Econ. Enromol. 63:656-658. LEWIS. WJ.. and MARTIN. W.R.. JR. 1990. Semiochemicals for use with parasitoids: Status and future. J. Chem. Ecol. 16:3067-3089. LWANDE. W .. McDoWELL. P.G.• AMlANI. H.. and AMOKE. P. 1989. Analysis of airborne volatiles of cowpea. PhyTOchemisrry 28:421-423. MARTIN. W.R.. JR .• NORDLUND. D.A.. and NETILES. W.e.. JR. 1990. Response of parasitoid Eucelaloria bryani to selected plant material in an olfactometer. J. Chem. Ecol. 16:499-508. McAuSLANE. H.J.. VINSON. S.B.• and WILLIAMS. H.J. 1991. Influence of adult experience on host microhabitat location by the genernlist parasitoid. Campoleris sonorensis (Hymenoptera: leh· neurnonidae). J. Insecr Behav. 4: 101-113. NAVASERO. R.C.• and ELZEN. G.W. 1989. Responses of Micropliris croceipes to host and nonhost plants of Heliothis virescens in a wind tunnel. Enroma/. Exp. Appl. 53:57-63. NORDLUND. D.A.• LEWIS. W.J.• and ALTIERI. M.A. 1988. Influences of plant-produced allele­ chemicals on the host/prey selection behavior of entomophagous . pp. 65-90. in P. Barbosa and D. Letourneau (eds.). Novel Aspects of Insect-Plant Internctions. John Wiley & Sons. New York. ORJENTATJON OF BRACONID TO LEAF VOLATILE 1753

PREVOST. G .. and LEWIS. WJ. 1990. Heritable differences in the response of the braconid wasp MicropJitis croceipes to volatile allelochemicals. J. Insect Behal". 3:277-287. SAIlO. R.• and TAKEO. T. 1975. Increase of cis-3-he,;en-I-QI content in tea leaves following mechan­ ical injury. Phytochemistry 14:181-182. SHEEHAN. W.. and SHELTON. A.M. 1989. The role of e,;perience in plant foraging by the aphid parasitoid Diaeretielia rapae (Hymenoptera: Aphidiidael. J. Insect Behal". 2:743-759. TOLLSTEN. L .. and BERGSTROM. G. 1988. Headspace volatiles of whole plants and macerated plant pans of Brassica sinapis. Phytochemistry 27:4013-4018. TURLINGS. T.C.. TUMLINSON. J.H.. and LEWIS. WJ. 1990. Exploitation of induced plant odors b;. host-seeking parasitic wasps. Science 250:1251-1253. TURLINGS. T.C.. TUMLINSON. J.H.. ELLER. FJ.• and LEWIS. WJ. 1991a. Larval-damaged plants: Source of volatile synomones that guide the parasitoid Cotesia marginil'enrris to the mlcrn­ habitat of its host. EnIomo/. Exp. App/. 58:75-82. TURLINGS. T.C.. TUMLINSON. J.H.. HEATH. R.R.. PROVEAUX. A.T.. and DooLlTILE. R.E. 1991b. Isolation and identification of the alle10chemicals that attract the larval parasitoid. COIesia marginivenrris (Cresson). to the microhabitat of one of its hosts. J. Chem. Eco/. 17:2235­ 2251. VISSER. J.H.• VAN STRATEN. S.. and MAARSE. H. 1979. Isolation and identification of volatiles in the foliage of potato. So/anum tuberosum. a host plant of the Colorado beetle. Leptinotarsa decemJineata. J. Chem. Ecal. 5: 13-25. WALLBANK. B.E.. and WHEATLEY. G.A. 1976. Volatile constituents from cauliflower and other crucifers. Phytochemist!}' 15:763-766. WHITMAN. D.W. 1988a. Plant natural products as parasitoid cuing agents. pp. 386-396. in H.G. Cuttler (ed.). Biologically active natural products potential use in agriculture. ASC Symposium Series 380. American Chemical Society. Washington. D.C. WHITMAN. D.W. 1988b. Allelochemical interactions among plants. herbivores. and their predators. pp. 11-64. in P. Barbosa and D. Letourneau (eds.). Novel Aspects oflnsect-Plant Inte.dctions. John Wiley & Sons. New York. WHITMAN. D. W.. and ELLER. F. 1990. Parasitic wasps orient to green leaf volatiles. Chemoec%gy 1:69-75.