Ovipositional Behavior of Lesser Peachtree Borer in Presence of Host-Plant Volatiles 1

Journal of Chemical Ecology, Vol. 14, No.1, 1988

OVIPOSITIONAL BEHAVIOR OF LESSER PEACHTREE BORER IN PRESENCE OF HOST-PLANT VOLATILES 1

D.K. REED,2,3 K.L. MIKOLAJCZAK,4 and C.R..KRAUSE5

3Fruit and Vegetable Insects Research Laboratory ARS, USDA, Vincennes, Indiana 47591 4Northern Region Research Center ARS, USDA, Peoria, Illinois 61604 5Nursery Crops Research Laboratory ARS, USDA, Delaware, Ohio 43015

(Received September 24, 1986; accepted January 22, 1987)

Abstract-Reactions of lesser peachtree borer [Synanthedon pictipes (G&R)] to volatiles of peach wood, either natural or chemically fractionated, were observed. Mated females were stimulated by and responsive to such materials and deposited significantly more eggs on substrates, including unnatural hosts, that had been treated with aqueous mixtures of bark-canker materials. Stim ulation to oviposit occurred even when the female was blinded, indicating the presence of chemical cues. Natural canker-bark extracts immediately stimulated ovipostion and for a few hours significantly increased the number of eggs laid. However, average fecundity was not increased. Antennectomy did not significantly decrease response to volatiles by gravid females, and alternate sites of such chemoreception were not located. Complex mixtures derived by solvent extraction, steam distillation, and volatiles trapping from bark, canker, and gum all had activity. Observations of insect behavior in outdoor cages and also in the laboratory indicated that visual, chemosensory, and mechanosensory receptors are involved in host finding and oviposition.

Key Words-Attractants, extracted volatile compounds, Synanthedon pic tipes, Lepidoptera, Sesiidae, chemosensory, insect behavior, oviposition, mechanosensory, host-insect interactions, olfactory receptors, oviposition behavior, apple, peach, pear, plant odors, Prunus spp.

I Mention of firm or product names does not imply recommendation or endorsement by USDA over others not mentioned. 2 Present address: Asian Parasite Laboratory, USDA, ARS, c/o American Embassy, Seoul, Korea, APO San Francisco 9630I.

237

I TRODUCTIO

In recent years, research on elucidation of insect-host interactions has included ovipositional responses of gravid females to their hosts. This relationship is crucial in most insect-plant interactions, but unfortunately it is sometimes dif ficult to establish and is the least understood relationship involved in such in teractions (Stadler, 1983). It is generally accepted that chemical cues may in voke both long- and short-range orientation by the ovipositing female to the host. The distinction between feeding and ovipositional stimulation is often ob scure, as exemplified by the dipteran pests (Liriomyea spp.). Fortunately, in studying nonfeeding lepidopterous females this problem is not encountered. However, with such females, moisture may become extremely important in their host-seeking (Saxena and Goyal, 1978). Another issue that needs clarifi cation is the distinction between ovipositional stimulation and ovipositional at traction. Although these functions are not necessarily mutually exclusive (Brown et al., 1970), a chemical attractant may not stimulate oviposition, and, con versely, a stimulant may not attract. The lesser peachtree borer, Synanthedon pictipes (G&R) (LPTB), is an important pest of peaches and cherries in most production areas. The host range of LPTB consists of the genus Prunus; it may therefore be considered a re stricted feeder, thus requiring fairly specific host-finding mechanisms. Females tend to oviposit on roughened bark of healthy trees near wounds where fresh gum is present (Bobb, 1959). Neonate larvae utilize light, chemotaxes, and thigmotaxes as cues in an active search for entry into the bark (Wiener and Norris, 1982). The reproductive behavior of adult Sesiidae females has been reported by Girault (1907), King (1917), Bobb (1959), and Cleveland et al. (1968), but the chemical relationships between the mated female and her host plants have not been defined. Observations have generally concluded that fe males are markedly attracted to the gummy deposits on peach wood that orig inate from cankers formed by cold damage, Cytospora sp. fungi, or mechanical injuries. A chemical isolated from peach bark, larval frass, and gum mixtures was reported by Gentry and Wells (1982) to be attractive to gravid females of peachtree borer, S. exitiosa (Say). Since S. exitiosa and S. pictipes are conge neric species, some commonality in such behavior might be expected. Our research was conducted to quantify the attractancy of gravid LPTB females to components of peach bark and to determine if such components stim ulate oviposition or increase fecundity. This paper presents the results of our attempts to isolate and identify such components and to record the ovipositional behavior of LPTB in the laboratory and field at Vincennes, Indiana, Peoria, Illinois, and Delaware, Ohio. OVIPOSITIO AL BEHAVIOR OF PEACHTREE BORER 239

METHODS AD MATERIALS

LPTB was reared using the methods of Cleveland et al. (1968). Newly mated females were maintained in screen cages on damp sand at 10 DC until used in experiments, either the same or the next day. Various parameters of response were determined in preliminary experiments. For experiments utiliz ing natural canker and gum, the cankerous areas on either Cresthaven or Red skin peach trees were removed by cutting both sides and ends of the canker and scraping bark, canker, gum and debris from the entire area. Gum was isolated by carefully selecting debris-free material. Healthy bark used as check material was peeled from 10- to 15-cm-diam. limbs from the same trees. All test mate rials were maintained in sealed plastic bags at 4 ± 1DC until used. Materials for scanning electron microscopy were mounted on aluminum stubs with con ductive cement and sputter-coated (model Hummer V, Technics, Springfield, Virginia) with 500 A of gold. SEM observations were performed with a Hitachi model S-500 (Mountainview, California) at 20 KV accelerating voltage. Chemical Cues Experiment. Small branches (2.5 x 25 cm), cut from pear trees (a LPTB nonhost) and from peach trees (a host) were presented to LPTB adult females in 50 x 50 x 50-cm screen cages in an air-conditioned green house. In test 1, branch ends were impaled by nails to a 35 x 35-cm board and exposed to 10 gravid females. Pear branches were coated with aqueous canker gum mixtures and tested in comparison to nontreated (clean) pear branches and water wicks. In test 2, peach wood was similarily treated and exposed to LPTB. The board was rotated periodically to ensure equal environmental exposure of each treatment. The test was replicated six times and data were submitted to analysis of variance. Visual Cues Experiments. In test 1, to determine the role that sight may play in host response, latex paint was applied to both compound eyes of 10 mated females. No adverse effects of the paint treatment upon the insects was noted. These treated insects were placed into a 30 x 30 x 30-cm screen cage which contained one peach branch (10 x 2 cm), treated with an aqueous canker gum-bark mixture from Redskin peach trees, and one pear branch the same size. A duplicate cage was used with 10 control (nonpainted) females. Behavior and egg deposition were observed in both cages for 6 hr. In Test 2, females, either normal or blinded with latex paint, were caged with apple, pear, and peach branches (10 x 2 cm). The canker-gum-bark mix ture was layered onto a moist cotton pad placed beneath the branches but in accessible to the insects. The cages consisted of 3. 78-liter ice-cream cartons (20 x 18 cm) with the middle of the carton removed and replaced with screen wire. A screen bottom separated the cotton pad 5 cm from the rest of the cage. 240 REED ET AL.

Treated cages as well as controls without the canker material consisted of three replications of three insects. After log transformation, 24-hr egg counts were submitted to analysis of variance and Duncan's (1955) multiple-range test. Ovipositional Stimulation Experiments. Two other types of tests were con ducted to determine if peachwood canker or gum might actively stimulate ovi position or increase fecundity. In the first experiment, air from the exhaust of an air compressor was passed through flasks containing aqueous suspensions of either canker material, gum, healthy bark, or distilled water. The volatile-laden air immediately flowed into four 13.5 x 10.5 x 3.5-cm plastic sweater boxes containing exhaust ports in the opposite ends. The boxes were layered with 2 cm-thick cotton batting saturated with distilled water. Twelve females, eclosed and mated the same day, were introduced into each box, and the eggs that were deposited on either the cotton or the box were counted after 24 hr in one series of tests and 48 hr in a second series of tests. Another experiment was conducted in individual plastic IOO-ml cups (6.5 x 6 cm) with copper screen covers. Aqueous canker material (0.5 ml) was applied to the center of a I-cm layer of moist cotton either on top of the cotton so that the insects could contact it or inverted so that the material was not accessible either tactilely or visually to the insects although they were exposed to volatiles. Each treatment was replicated 25 times; moist cotton pads were used as controls. Surgically altered LPTB females were used as subtreatments consisting of five replicate Latin-square arrangements as follows: (1) normal mated female, (2) mated female with one antenna removed at base of pedicel, (3) mated female with both antennae removed, (4) mated female with both an tennae three-fourths removed, and (5) mated female with tip of hypopharnyx removed. Cups were maintained in an air-conditioned greenhouse at 85 ± 5°C and 75 ± 10% relative humidity. Eggs were counted at 24-hr intervals for 72 hr. The entire test was replicated on each of five dates, and the mean number of eggs per test was submitted to analysis of variance after log transformation using the individual tests as replicates. A similar test was conducted and 80% confidence levels were computed for egg production at 5-, 24-, and 48-hr in tervals to determine if the canker extracts might cause more immediate effects. Behavior in Outdoor Cages. A 1.8 x 1.8 x 2.4-m screen cage was con structed for outdoor observations of LPTB behavior. Peach branches of various sizes and shapes, with and without foliage, were suspended at a height of 1.5 m in the four quadrants of the cage into which 10 mated females were released. Observations were made only on sunny, warm days since LPTB females were found to be unresponsive under other conditions. Females were observed for ovipositional movements and egg deposition. A large 6 x 2.5 x 2.5-m cage was constructed over two peach trees (var. Cresthaven ) in a peach orchard located at Southwestern Indiana Purdue Agri- OVIPOSITIO AL BEHAVIOR OF PEACHTREE BORER 241 cultural Center, north of Vincennes. One tree was spot treated with 2-chloro ethylphosphonic acid (Union Carbide Corp.) to promote gum formation. After two weeks, copious gum production around the treated portions of the trunk and limbs and a general weakening of the tree was apparent with some shedding of foliage. Varying numbers of gravid LPTB females were released from the middle of the cage, and the females were observed for periods varying from 1 to 6 hr. Behavior and egg deposition were observed. A replicated experiment was conducted in which a 2-cm-thick peach branch was divided into four 8-cm sections and placed in the four quadrants of the small cage containing 12 females which were replaced on three consecutive days. Treatments consisted of (1) a smooth branch section, (2) an identical section with canker-gum mixture applied, (3) a section with one small (0.5 cm) new canker, and (4) a section with one large (1.5 x 0.5 cm) older canker. After log transformation, total egg numbers were submitted to analysis of variance and Duncan's (1955) multiple-range test. Olfactometer Experiments with Natural Canker Materials. To quantify LPTB ovipositional behavior, we devised an olfactometer from a 50 x 5.5-cm glass tube. One end was closed with a cork containing a length of 0.5-cm latex tubing leading to a vacuum pump which was exhausted outside. The other end of the glass tube was corked with a rubber stopper modified to hold a screen cage (8.5 x 4 cm) in which the test insect was confined inside the tube during testing. Air from outside the laboratory was drawn through a charcoal filter into a glass sample chamber and finally into the olfactometer. The sample chamber contained a filter paper strip (2 x 0.5 cm) on which a drop (generally 0.01 0.02 ml) of test material was placed. Air flow was adjusted to 0.1-0.3 m/sec and the temperature was maintained at 18 ± 2°C. A fluorescent desk lamp was positioned directly over the insect holding cage of the olfactometer. A single insect was placed in the cage which was then attached to the front rubber stopper and inserted into the chamber. The insect was allowed to acclimate to the air flow for 30 sec prior to inserting the treated filter paper strip into the air flow. Reactions were observed and recorded for at least 1 min. Strong ovipositional thrusts with the ovipositor were considered positive responses. Five insects were used for each observation. Insects were maintained overnight on damp sand at 10°C and were used only one or two times during one days' observations. In one experiment, females, surgically altered in a manner similar to those in the ovipositional stimulation experiment, were exposed to natural volatiles from canker material in the olfactometer in an attempt to determine the location of chemoreceptors. Response was tabulated according to abdominal movement, and subsequent egg deposition in plastic cups layered in cotton (Cleveland et al., 1968) was recorded for each insect. Olfactometer Experiments with Materials Isolated from Canker and Bark. 242 REED ET AL.

To collect volatiles from bark, peachwood bark (var. Redskin and Bisco) was collected in the spring of 1982 and frozen dry at -18°C. These materials (ca. 1.0 kg from each variety) were ground in a Wiley mill; the ground bark was placed in a 40 X 60-cm plastic bag that had been outfitted for collection of volatiles. Nitrogen, at a flow rate of 1 liter/min, was used to sweep volatiles onto 1 x 8-cm Tenax columns. Collections proceeded for 4.5 days, but were interrupted each night, when the bark was refrozen. A new trap with fresh Tenax was used each day. The traps were eluted with 10 ml of acetone to remove condensed moisture and then with 50 ml of diethyl ether. Washings were com bined, and the solvent was concentrated by distillation through a 1 x 30-cm Vigreux column. Residual acetone was codistilled with pentane, leaving the sample in 10 ml of pentane (fraction 1). Portions (200 g) of the ground Redskin and Bisco peach bark after volatile collection were stirred for 2 hr each with three 500- to 600-ml portions of distilled water at room temperature. After a brief settling period, the water was decanted through glass wool to remove most suspended materials. The fine particles remaining were allowed to settle out overnight under refrigeration. The decanted aqueous extract was subjected to a continuous extraction with diethyl ether for 18 hr, and the ether extract was concentrated by distillation through a 1 x 30-cm Vigreux column. Residual ether was codistilled with pentane to provide 10 ml of fraction 2. A portion of the ether-extracted material was insoluble in pentane, so it was dissolved in 10 ml of acetone (fraction 3). For fractionation of gum and canker, samples (ca. 3-4 kg total) were col lected from Redskin peach trees on July 29, 1982, at Vincennes, placed together in a carboy and immediately covered with dichloromethane. The carboy and contents were transported to Peoria and stored at 4 °C. After a month, the con tents of the carboy were mixed thoroughly, and a 5-ml sample of the dischlo romethane solution was removed and served as fraction 4. Gum and cankers (21 g) were cut into 2- to 5-mm sections and then Soxhlet extracted for 3 hr with acetone. The residue was air-dried, ground finely in a mortar, and reex tracted with acetone for 7 hr. The combined extracts were concentrated to ca. 20 ml on a rotary evaporator at 40°C, and the concentrate was divided into two equal portions (one portion = fraction 5). The remaining portion was subjected to azeotropic distillation with absolute ethanol on a rotary evaporator at 40°C to remove water, then the ethanol was removed with dichloromethane, and the sample (fraction 6) was redissolved in 10 ml of acetone. A second portion of gum and canker material (20 g) was chopped and then stirred with 500 ml of distilled water at room temperature for 4 hr. The water extract was decanted and subjected to a 3-hr simultaneous steam distillation extraction using a Likens-Nickerson apparatus with dichloromethane as the ex tracting solvent at ambient pressure (Likens and Nickerson, 1964). Dichloro methane was codistilled with pentane, thus giving fraction 7 for assay. OVIPOSlTIO AL BEHAVlOR OF PEACHTREE BORER 243

Finally, a third portion of gum and canker material (20 g) was chopped and then stirred with 200 ml of distilled water at room temperature for 4 hr. The total slurry was extracted in a separatory funnel with diethyl ether-ethyl acetate (1 : 1) to give fraction 8 after distillation of the solvent. Samples were sealed in ampules until assayed against gravid LPTB females. Cytospora Experiment. Cankerous lesions on peach trees, which are known to be attractive to ovipositing LPTB, are usually associated with fungal patho gens. Swift (1986) reported that cankers associated with Cytospora were in fested with LPTB earlier and to a greater extent than uninfected wounds. He suggested that the Cytospora may have caused a physical alteration of the wounds making them more suitable for egg deposition and larval invasion. The most important organism causing such cankers, Cytospora cincta Sacc., was cultured on potato-dextrose-agar, and aqueous washes of the cultures were ex posed to LPTB females in the olfactometer to determine whether it played a role in host specificity of LPTB.

RESULTS AD DISCUSSIO

Chemical Cue Experiment. Observations of LPTB behavior in cages showed that more eggs were deposited on branches that were coated with ex tracts of gummy, cankerous peachwood bark. Pearwood, a nonhost of LPTB, became attractive when peach canker material was applied to it. In a 6-hr ob servation of branches impaled on nails exposed to LPTB females (test 1), sig nificant differences (P = 0.01) were observed. Pearwood alone received 11 % (SE = 1.6) of total eggs, while identical pearwood with canker material on it received 82 % (SE = 10.7), and a water wick control had 7 % (SE = 1. 2) of total eggs. When peach wood was exposed in a similar test (test 2), wood with canker material received 82 % (SE = 16.1), while wood without such material had only 14% (SE = 5.9) of total eggs, and a Nater wick check had 4% (SE = 1.3). In these experiments, some eggs were always deposited on unnatural hosts such as the pear wood or moist cotton. This may ha Ie been due to a concentration of volatile molecules from the treated ar _J causing stimulation of females resting on these sites. However, some indiscriminate egg-laying may be expected from laboratory-reared LPTB, since they nC'rrnally oviposit on moist cotton without the presence of host stimuli. Visual Cue Experiments. In test 1, blinded females immediately responded and began ovipositional movements in the presence of peach and pear wood. They moved about the cage and, upon encountering the branches, would ovi posit. After 6 hr, an average of 13.9 eggs per blind female had been laid, with 70% on peach and 30% on pear. Normal females had laid 21.0 eggs per female with 98 % on peach and only 2 % on pear. 244 REED ET AL.

Results of test 2, in which normal and blind females were exposed to three different types of wood with and without stimulation by canker extracts, are shown in Table 1. When no canker was present, normal females oviposited almost exclusively on peach wood. However, in the presence of strongly at tractive canker material, peach was not preferred over apple but was preferred over pear. In this experiment, the numbers of eggs deposited by normal females were greater where additional canker material was not present, although the difference was not significant. It is not known whether the greater egg deposi tion was due to a few particularly fecund individuals or to the presence of in hibitory volatiles in high concentrations of canker material. Blind females in canker-saturated environments indiscriminately desposited eggs on all three substrates, but apple apparently corresponded more closely to peach than pear in stimulating tactile response from LPTB. It was observed that blind females were strongly responsive to the canker material. When blind females were stim ulated, then placed upon an apple branch, they immediately began probing with their ovipositor. Unstimulated blind females ordinarily remained docile. Since eggs are desposited on such nonhosts by blind insects, it would seem logical that the sensory hairs present on the ovipositor (Figure 1) were mechanorecep tors rather than chemoreceptors. Since no transmission electron microscopic examinations were made, it is unknown how either the long (±100 jLm) or the short (1.0-1.5 jLm) structures are innervated. In these observations, there was never any evidence of chemoreception by the ovipositor. However, in the pres ence of a strong enough stimulus, even a normal female will oviposit on apple or pear. Normal females lay a much higher percentage of eggs on the wood than on cotton (75: 25), than blind females do (16: 84). This is an indication that sight does have some significance, as well as that blind females may ovi-

TABLE 1. HOST SELECTIO BY GRAVID LPTB FEMALES WITH OR WITHOUT VISIO I PRESE CE OR ABSE CE OF STIMULATIO BY CA KER MATERIAL

Percentage of total eggsa deposited when

Canker material Canker material absent present

Offered host Normal female Blind female ormal female Blind female

Peach wood 92.3ab 41.0a 62.7a 42.3a Apple wood 5.0b 20.3a 32.7a 38.7a Pear wood 2.7b 38.7a 4.7b 19.0a

a Three females used in each test. b Means within columns followed by the same letters are not significantly different at the 5 % level by Duncan's multiple-range test. o < ~ o Ul ~ (3 Z :J> r OJ tTl ::t :J> < (3 ::0 o 'Tl "0 tTl :J> () ::t...., ::0 tTl tTl oOJ ::0 tTl ::0

FIG. 1. Scanning electron micrograph of the tip of the ovipositor of LPTB: (A) bar = 50 Ilm; (B) bar = 10 Ilm. N..,. Vl 246 REED ET AL. posit indiscriminately in the presence of stimuli. Another indication of the im portance of sight is the observation that while normal females are stimulated and will follow a canker source in still air, blind females, although stimulated, will not follow such a source. Ovipositional Stimulation Experiments. Preliminary observations showed that females placed in closed Petri dishes were inactive and did not oviposit in the presence of peach canker material. However, after removal of the tops and replacement with screen covers, abdominal responses began immediately. It is unknown whether the observed lack of response in' closed dishes was due to habituation, to a repellent in the chemical, or to other causes. After exposure to air, females in the treated dish laid 166 eggs compared with only 72 in the untreated dish (18 hr). This led to the hypothesis that fecundity of LPTB in creased in the presence of canker material. This hypothesis, however, was re jected after examination of the data from the experiments in which ovipositing LPTB females were exposed to volatiles from canker, gum, and bark of peach. In the first experiment, in which air was bubbled through aqueous substrates into a plastic oviposition chamber, mean numbers of eggs per female were low est in the control chamber, both after 24- and 48-hr observations. However, no significant differences existed among the treatment means. Plastic cups with screen tops were used to provide a free exchange of air in the second test. A complicating factor in using bioassays such as these with LPTB is the propensity of the colonized female to oviposit on cotton or other substrates where sufficient moisture is present. For this reason, the canker was placed on the under side of the cotton in one set of tests to detect quantitative differences caused by volatile fractions. Results of this test also indicated that no significant differences in egg production occurred either in the presence or absence of canker material or its volatiles. This was true whether the female had been surgically altered or was normal. Another experiment, again in plastic cups, in which counts of eggs from five females were made at shorter intervals (Figure 2), indicated that a signifi cant effect was produced by the canker material but also that it was short-lived. The canker apparently stimulated early egg production, but total fecundity was not statistically increased. This observation is in line with other observations in the laboratory and field and indicates that the canker material not only attracts mated female LPTB, but also stimulates oviposition to some degree. However, the fact that unstimulated insects will oviposit in the presence of moisture com plicates interpretation of these data. Behavior in Outdoor Cages. The behavior of gravid LPTB females was observed inside a small cage in the presence of various peachwood substrates. Several interesting observations were made which may aid in understanding ovipositional behavior in the field: (1) There was no attraction to peach foliage even on hot, sunny days. (2) Most oviposition occurred on rough areas of the OVIPOSITIONAL BEHAVIOR OF PEACHTREE BORER 247

195_----."r-----r-----.------r------. X • CHECK o • CANKER i I 166 I I I I I T I ~ 138· I I I T I I I I I I I I I I I X I +I I T I I I I I I I I I I 1 I I .1 81· I FEMALE I *I I .1 T 52· I I ,I I I 24· I .1. i

I *.I 5 24 48

LENGTH OF EGG DEPOSITION PERIOD

FIG. 2. Egg production of mated LPTB in the presence of host volatiles over time after mating. bark. (3) Branches less than ca. 1 cm, even with canker extract on them, were not attractive for oviposition. (4) Branches with natural cankers were extremely attractive. (5) Although excised canker material would attract females to a sub strate, oviposition upon that substrate seemed to depend on the texture. Thus, in the field, chemosensory and mechanosensory receptors as well as vision, function in the act of oviposition. 248 REED ET AL.

Observations in the large field cage were made on warm, sunny days be cause females would rarely fly on overcast days, and on windy days, they be came lost in the grass. Females exhibited a high degree of attraction to sunny areas of the cage, preferring to rest on the side of the cage or tops of the trees in direct sunlight. When released in the center of the cage between the treated and untreated tree, they would usually orient toward the treated tree with the copious gum deposits, although the attraction to light seemed to be greater than attraction to either tree. When greater numbers of females (> 50) were released, some could readily be located in the treated tree and, more rarely, in the un treated tree. The untreated tree had more foliage, which may have provided more hiding places, but ovipositing females were found much more often on the treated tree. Once the females alighted on the tree, they were immediately stimulated and walked around on the branches (usually the larger branches and trunk) with their abdomens dragging until they located a site that was suffi ciently rough to oviposit. The rougher bark and canker areas invariably received more eggs than smoother areas. Much exploring of the cracks and crevices was made with the ovipositor prior to oviposition. The setae present on this organ (Figure 1) must function to locate the most strategic position for eggs after chemoreceptors have led the female to the proper site. These observations in dicate that olfaction, as with many other insect species, is important for close range identification of a suitable host by LPTB. Results of the replicated experiment using a segment branch are presented in Table 2. The branch section containing a large, older canker received sig nificantly more eggs (P = 0.05) than one with canker material applied, and also had significantly more eggs (P = 0.01) deposited on it than sections with no added material or with only a small canker. Olfactometer Experiments with Natural Canker Materials. LPBT females exhibit a definite typical response pattern when stimulated by volatiles from natural canker material. This includes antennal movement, followed by a down ward arching of the posterior abdomen with associated ovipositional thrusts of

TABLE 2. EGG DEPOSITIO BY LPTB FEMALES I OUTDOOR SCREE CAGE (12 FEMALES REPLACED DAILY FOR 3 DAYS)

Peachwood with Mean o. of eggs per branchG

o canker 44a Small canker 68a Added canker material 109c Large canker l63d

G Means within the column followed by the same letters are not significantly different at the 5 % level by Duncan' multiple-range test. OVIPOSITIO AL BEHAVIOR OF PEACHTREE BORER 249 the ovipositor. This behavior does not occur when either virgin females or males are exposed to such volatiles. Results of the test to attempt to determine the location of chemosensory structures on the LPTB are shown in Table 3. Un fortunately, we were unable to determine sensory sites since olfactory-mediated excitement was obtained no matter which type of surgery was applied. Even complete antennectomy failed to suppress ovipositional movements in the pres ence of stimulatory materials. Since the antennectomy retained the scape of the antennae, we thought that enough sensory receptors may have been present there to invoke response. However, coating these structures with fingernail pol ish did not cause further suppression of response. Using another lepidopterous insect, Plodia interpunctella HBN., Deseo (1976) found that, following 50% antennectomy, olfactory stimuli which ordinarily increased fecundity lost their effect, and fecundity became the same as if there were no stimulation. This indicated that the chemoreceptors involved in oviposition were located on the distal half of the antennae. Similarly, Yamamoto and Fraenkel (1960) found that oviposition in the tobacco hornworm appeared to be initiated by olfactory stimuli which were completely suppressed by amputation of the antennae. It is evident from our research that the sensory receptors mediating ovi position only function after mating. However, experiments involving ablation of various parts of the LPTB body reported here, as well as other observations, including tarsal amputation and ablation of tarsal sensilla with acids, failed to determine the sites of such receptors. Stadler (1976) stated that negative results in ablation experiments may not only be the effect of the removed sensory input, but could also be a kind of irritation or reduction in general activity. In our observations, ablation obviously had a detrimental effect upon the insect, but, contrary to expectations, positive rather than negative results were obtained. Whether this may be a result of elimination of inhibitory influences has yet to be determined. In any event, the location of chemoreceptors functioning in ovi-

TABLE 3. RESPO SE OF LPTB WITH VARIOUS DEGREES OF A TE ECTOMY TO ATURAL PEACH CA KER VOLATILES I OLFACTOMETER A D SUBSEQUE T FECU DITY

Treatment Response (%t Fecunditl

one, control. mated 100 42.2(24.6) Single complete antennectomy 100 33.2(19.6) Double complete antennectomy 85 31.1(52.1) Double 3/4 antennectomy 100 36.8(42.1) Virgin female o o (0) Virgin male o

a Response as indicated by ovipositional movement. b Mean number of eggs per female per day (standard deviation of mean of total eggs). 250 REED ET AL.

positional response of LPTB is still unknown, and although the majority of the receptors may reside on the antennae, those present on other parts of the body apparently function in the absence of antennal receptors. Olfactometer Experiments with Materials Isolated from Canker and Bark. Assays of mixtures obtained by solvent extraction, steam distillation, and vol atiles trapping from peach wood and canker-gum materials indicated the ex treme difficulty of isolation of the component(s) responsible for ovipositional response. Of the three fractions obtained by steam distillation of peach bark, there was little difference in response to Redskin or Bisco fractions (Table 4). Fractions 1 and 3 both elicited a good ovipositional response, while fraction 2 elicited only a 10% response to Redskin and 20% to Bisco; this was the only variation between the two. Apparently, most of the activity was concentrated in the pentane-insoluble fractions. The Soxhlet-extracted fractions from canker and gum were all active, except fraction 4 which elicited no response. The dichloromethane-extracted fractions showed the least activity, although enough active volatiles were present for stimulation. Other fractionation procedures utilizing various solvents, collection col umns, and distillation procedures were attempted, but all failed to isolate the activity into a recognizable fraction. One problem is that the active molecules apparently excite the LPTB at extremely low concentrations. Behavioral re sponses such as those employed in these experiments are apparently not suitable for quantitatively distinguishing between fractions when all or most of the frac tions contain volatiles that elicit response. For this reason, some of these same materials were assayed with the electroantennogram (EAG) method. Results of these studies (Anderson et al., 1987) indicate that this technique may be more discriminating and probably should be used for primary screening prior to un-

TABLE 4. OVIPOSITIO AL RESPO SE TO MIXTURES OBTAI ED BY SOLVE T EXTRACTIO ,STEAM DISTILLATIO ,A D VOLATILES TRAPPI G FROM PEACH WOOD AD CA KER-GUM MATERIAL

Fraction o. N Response (%t

1 10 70 2 10 10 3 10 90 4 5 0 5 45 44 6 40 75 7 15 33 8 10 60

a 0 response was elicited by females exposed to solvent alone (pentane). OVIPOSITIO AL BEHAVIOR OF PEACHTREE BORER 251

dertaking behavioral assays. Habituation by such active volatiles, as well as repellent or inhibitory effects due to hypemonnal concentrations within the bioassay arena, are possibilities, among others, that need exploring. Cytospora Experiment. When extracts of Cytospora cincta cultures were presented to gravid LPTB females in the olfactometer, no response was evident. Although LPTB larvae aid in invasion by this organism by producing new open ings in host tissue previously sealed off by callus tissue (Hildebrand, 1947), Cytospora apparently has no chemical function in attraction of females to host plants or to stimulation of oviposition. In finding and consuming host plants, insects may draw upon olfaction, vision, mechanoreception, gustation, and perhaps additional signals. The lesser peachtree borer, according to our observations, uses at least the first three sig nals in varying degrees. The LPTB has a survival advantage in that it may oviposit in the same host tree whence it encloses. However, earlier authors (Girault, 1907; King, 1917) reported that mated females flew from tree to tree, and subsequently observations have confinned this, but reports of distances flown are sketchy. To some degree, either sight or olfaction or both must be involved in host-finding. Wind-tunnel observations (Reed, unpublished data) show that mated female LPTB will orient upwind and fly 3 m to a source of canker material, indicating that olfaction may be important at some distance. Observations in the greenhouse (Reed, unpublished data) indicate that the fe male in flight can follow a hand-held source of canker volatiles, attempt to land on it, and oviposit. These observations plus results of the studies reported here indicate that LPBT females rely upon olfaction in host feeding. More research is needed on the aspect of discrimination between olfaction and sight and on detennination of distances involved in such relationships. Electrophysiologic procedures may be helpful in elucidation of these relationships. However, it must be recognized that chemoreception of plant chemicals is far more complex than the perception of pheromones (Stadler, 1980).

Acknowledgments-The authors thank Steve Ries, Department of Plant Pathology, University of lllinois, for furnishing the culture of Cytospora cincta. We also thank aomi Tromley, Paula Moriarty, Diana Marchino, and Bruce Zilkowski for their excellent technical assistance throughout all phases of the research. Appreciation is extended to Drs. Erich Stadler, Dale Norris, John Neal, Jr., and Frank Slansky, Jr. for their critical reviews of the manuscript.

REFERE CES

A DERSE . J.F., MIKOLAJCZAK. K.L., and REED. D.K., 1987. Analysis of peach bark volatiles and their electroantennogram activity with lesser peachtree borer, Synanthedon pictipes (Grote and Robinson). J. Chern. Ecol. 13:2103-2114. BOBB, M. L. 1959. The biology of the lesser peachtree borer in Virginia. 1. Econ. Entomol. 53 :634 636. 252 REED ET AL.

BROW, W.L., JR., EISNER, T., and WHITTAKER, R.H. 1970. Allomones and kairomones: Tran specific chemical messengers. BioScience 20:21-22. CLEVELA D, M.L., Wo G, T.T.Y., and LAMA SKY, K.W. 1968. Rearing methods and biology of the lesser peachtree borer, Synanlhedon piclipes, in the laboratory. Ann. Entomol. Soc. Am. 61:809-814. DESEO, K.V. 1976. The oviposition of the Indian meal moth (Plodia interpunctella HB ., Lep., Phyticidae) influenced by olfactory stimuli and antennectomy. Symp. Bioi. Hung. 16:61-65. Du CA , D.B. 1955. Multiple range and multiple F tests. Biometrics 11:1-41. GE TRY, C.R., and WELLS, J.M., 1982. Evidence of an ovipositional stimulant for peachtree borer. J. Chem. Ecol. 8: 1125-1132. GIRAULT, A.A. 1907. The lesser peachtree borer. U.S.D.A. Bur. Enlomoi. Bull. 68(part IV): 31 48. HILDEBRA D, E. M. 1947. Perennial peach canker and the canker complex in ew York, with methods of control. Cornell Univ. Agric. Sm. Mem. 276: 1-61. KI G, J.L. 1917. The lesser peachtree borer. Ohio Agric. Exp. Bull. 307:399-448. LIKE s, S. T., and NICKERSO , G. B. 1964. Detection of certain hop constituents in brewing prod ucts. Proc. Am. Soc. Brew. Chem. 5-13. SAXE A, K.., and GOYAL, S. 1978. Host-plant relations of the citrus butterfly Papilio demoleus L.: Orientation and ovipositional responses. Enlomol. Exp. Appi. 24: 1-10. STADLER, E. 1976. Sensory aspects of insect plant interactions, pp. 228-248, in Proceedings of XV International Congress of Entomologists, August 19-27, Washington, D.C. STADLER, E. 1980. Chemoreception on arthropods: Sensory physiology and ecological chemistry, pp. 223-241, in R. Gilles (ed.). Animal and Environmental Fitness. Pergamon Press, ew York. STADLER, E. 1983. Attractants, arrestants, feeding and ovipositional stimulants in insect-plant re lationships: Application for pest control, pp. 243-258, in D.L. Whitehead and W.S. Bowers (eds.). atural Products for Innovative Pest Management. Pergamon Press, New York. SWIFT, F.e. 1986. Cytospora canker and the establishment of lesser peachtree borer (Lepidoptera: Sessiidae) in peach trees. J. Econ. Entomol. 79:537-540. WEI ER, L.F., and ORRIS, D.M. 1982. Light and other stimullii in the location of feeding sites by lesser peachtree borer larvae. Environ. Entomol. 11 :399-402. YAMAMOTO, R.T., and FRAE KEL, G.S. 1960. The specificity of the tobacco hornwonn, Protoparce sexta, to solanaceous plants. Ann. Entomoi. Soc. Am. 53:503-507.