Influence of caterpillar-feeding damage on the foraging ... - Springer Link

Journal of Insect Behavior, Vol. 6, No. 6, 1993

Influence of Caterpillar-Feeding Damage on the Foraging Behavior of the Paper Wasp Mischocyttarus flavitarsis (Hymenoptera: Vespidae) Mary L. Cornelius 1 Accepted July 23, 1992; revised May 17, 1993

Experiments were performed to determine the effect of caterpillar feeding damage on wasp foraging behavior and to determine the relative importance of visual and olfactory plant cues for foraging wasps. In an experiment using caterpillar-damaged leaves, wasps took significantly more larvae from the previously damaged plants compared to the controls in the experiments with tobacco plants, but wasps did not distinguish between damaged and control plants in the experiments with tomato plants. Another experiment indicated that wasps use a combination of visual and olfactory cues of plant damage in their search for prey rather than just visual or olfactory cues alone. Furthermore, these results suggest that leaf shape may affect wasp detection of caterpillar feeding damage and thus detection of prey. K E Y WORDS: foraging behavior; caterpillar-feeding damage; paper wasps; Lepidoptera.

INTRODUCTION

Evidence of caterpillar feeding may serve as a cue for predators of Lepidoptera. Heinrich (1979) suggested that palatable caterpillars have developed behavioral adaptations to avoid visually hunting predators by feeding only on the underside of leaves or only at night and by concealing the evidence of feeding damage by clipping off partially eaten leaves, abandoning a partially eaten leaf and moving to a new distant leaf, or traveling between resting and feeding sites. Lederhouse (1990) found that early instar tiger swallowtail caterpillars fed along the leaf margins, producing damage with smooth edges. He suggested that this behavior t Department of Entomology, University of Hawaii, Honolulu, Hawaii 96822. 771 0892-7553/93/1100-0771$07.00/0 9 1993PlenumPublishingCorporation

772

Cornelius

is an adaptation to minimize visual searching cues for predators. There have been relatively few studies that have actually tested how predators respond to caterpillar feeding damage. Two of these studies produced opposite results: chickadees learned to forage preferentially on trees with caterpillar-damaged foliage (Heinrich and Collins, 1983); in contrast, the presence of feeding damage did not affect predation by great tits (Bergelson and Lawton, 1988). Arthropod predators and parasitoids may also use plant damage as a cue in their search for prey. The spined soldier bug Podisus maculiventris orients to soybean plants damaged by Trichoplusia ni (Greany and Hagen, 1981). The predators, Chrysoperla carnea and Collops vittatus, are attracted to a terpenoid released by damaged cotton leaves (Flint et al., 1979). Predators of bark beetles respond to volatile terpenes that are released by pine trees when they are being attacked by bark beetles (Wood, 1982). Predatory mites are attracted to plant volatiles released from plants infested with spider-mites (Dicke et al., 1990). Turlings et al., (1990) found that a parasitoid used volatile terpenoids which were released by caterpillar-damaged corn plants as searching cues and that caterpillar-damaged plants were significantly more attractive than artificially damaged plants. Vespid wasps may use damaged leaves as cues in their search for prey. Polybia wasps searched damaged leaves more frequently than undamaged leaves (Raveret Richter, 1988). Aldrich et al. (1985) found that the eastern yellow jacket, Vespula maculifrons, was attracted to volatile compounds which are released by damaged leaves. The searching behavior of predators can also be affected by morphological characteristics of the plant such as the presence of glandular trichomes (Obrycki, 1986) and differences in plant surfaces or plant structural complexity (Rasmy, 1977; Carter et al., 1984; Kareiva and Sahakian, 1990; Grevstad and Klepetka, 1992). Predators may also respond to differences in leaf morphology. For example, Niemela and Tuomi (1987) hypothesized that natural enemies might be attracted to irregularly shaped leaves because they mimic caterpillar damage. This study examines how a vespid wasp, Mischocyttarusflavitarsis (Sauss.), uses visual and olfactory plant cues in its search for prey and how the interaction between leaf morphology and caterpillar feeding behavior affects the foraging behavior of wasps. MATERIALS AND METHODS

Mischocyttarus flavitarsis (Hymenoptera: Vespidae) is a polistine wasp which occurs in the western United States and Canada. It is common in the San Francisco Bay area, where this study was conducted. Open nests are built in protected sites such as the eaves of buildings or in dense brush. Female wasps emerge from hibernation in the spring and initiate new nests. The majority of

Influence of Caterpillar-Feeding Damage on Wasp Behavior

773

nests is founded by a single queen. Workers begin to emerge in late spring or early summer. Colonies reach their peak during the summer and begin to decline in the fall (Litte, 1979). Nests in my study were abandoned during November and early December. The experiments were conducted in a greenhouse of approximately 150 m 2 at the Division of Biological Control, University of California, Berkeley, in Albany, CA. Wasp nests were initially established in the greenhouse in the summer of 1989 by collecting solitary queens with their nests from buildings on the UC Berkeley campus and taping the pedicel of each nest to the upper comer of a wooden cage in the greenhouse. The queen was then released onto her nest. Cage doors were left open so that wasps could forage freely in the greenhouse. In the spring of 1990, founding queens built their own nests in the greenhouse. Wasps were provided with a source of sucrose from a hummingbird feeder containing sugar water and from cotton wicks immersed in beakers of sugar water. Dead potato tuber moth larvae Phthorimaea operculella (Zeller) (Lepidoptera: Gelechiidae) were set out every few days to provide a source of protein. The caterpillar prey used in this study were the tobacco hornworm Manduca sexta (L.) (Lepidoptera: Sphingidae) and the cabbage looper Trichoplusia ni (Hubner) (Lepidoptera: Noctuidae). M. sexta is an oligophagous herbivore which feeds on a number of plant species in the family Solanaceae in nature (Yamamoto and Fraenkel, 1960). T. ni is a polyphagous herbivore which feeds on plant species in several families, including members of the Solanaceae (Tietz, 1972). The solanaceous plant species used in these experiments were tomato, Lycopersicon esculentum, and tobacco, Nicotiana tabacum. M. sexta and T. ni larvae were obtained from laboratory cultures kept at the University of California, Berkeley.

ExperimentalDesign In all experiments, a damaged and an undamaged plant were set out in the greenhouse. One larva of the same size (10-20 mm) was placed in the same spatial position, i.e., third leaf from the bottom, on each plant in the pair. For each trial, a pair of plants of the same species and approximately the same height and number of leaves was set out on a bench in the greenhouse. Plants used in experiments ranged in height from approximately 20 to 60 cm. Each trial lasted until a wasp took one of the larvae. Wasps were captured after preying on a larva, then chilled in the refrigerator for several minutes, marked on the thorax with a numbered tag, and released after the experiment was over. Thus, each trial represents both a different plant pair and a different individual wasp. Marked wasps were sometimes used again in later experiments, but there was at least a week between trials using the same individual. The number of larvae

774

Cornelius

on damaged or undamaged plants taken for each experiment was compared using a sign test (Sokal and Rohlf, 1981).

Effect of Caterpillar-Feeding Damage on Wasp Foraging Behavior The effect of caterpillar feeding damage on wasp foraging behavior was examined by presenting wasps with larvae on plants with visible feeding damage and larvae on plants without any damage. This experiment was performed with both tobacco and tomato plants in 1989 and 1990. In this experiment, secondor third-instar M. sexta larvae were used. Larvae were reared upon the same plant species as they were tested on. On the undamaged plant, a larva was placed on the plant immediately before starting the trial. On the damaged plant, a larva had been placed on the plant at least 3 h before starting the trial and had already started feeding on the plant prior to the test. Leaf area removed ranged from approximately 5 to 20 mm 2. In all trials the larva remained on the same leaf as initially placed, but its location on the leaf varied between trials. Also, differences in leaf morphology between the two plant species affected caterpillar feeding behavior. Tobacco plants have large, broad leaves, whereas tomato plants have leaves composed of small leaflets with serrated edges. On tobacco plants, larvae often continued to feed in the same spot, but sometimes they left a hole and moved to a different area of the leaf. Larvae rarely fed along the leaf edge. On tomato plants, larvae were much more likely to be feeding along the edges of a leaflet and sometimes they consumed most of a leaflet and moved to a new leaflet before the experiment started. In the experiment performed in 1989, wasps were not captured and marked and each trial lasted until wasps took larvae on both the damaged and the undamaged plants. I recorded which larva was taken first and compared the searching time of wasps for caterpillars on damaged and undamaged plants. The length of time between the beginning of a trial and the time when each larva was taken in a trial was compared using a Wilcoxon signed-ranks test for matched pairs (Sokal and Rohlf, 1981). In the 1990 experiment, wasps were captured and marked and each trial lasted until a wasp took one of the larvae.

Effect of Visual and Olfactory Cues on Wasp Foraging Behavior In order to determine which plant cues wasps were using in their search for prey, I performed a series of experiments designed to separate visual and olfactory cues. These experiments were performed using artificially damaged plants. Freshly killed M. s e x t a caterpillars were used in order to test the response of wasps to damage without any other complicating factors related to caterpillar feeding such as the presence of frass or the behavior of the caterpillar. Caterpillars were killed just before running the experiment by placing them in hot water for a few seconds.


775

An experiment was performed to compare the response of wasps to freshly damaged plants and plants with no damage. This experiment was performed with both tobacco and tomato plants. In the damaged plant, an 8-mm-diameter hole was made with a cork borer in a leaf immediately before running the trial. In the tobacco plants, the hole was made in the middle of the leaf next to the midvein. In the tomato plants, the hole was made in the center of a leaflet. One freshly killed larva was placed next to the artificial hole on each damaged plant and in the same spatial position on undamaged plants. In order to test the effects of visual feeding damage in the absence of olfactory cues, this same experiment was also performed using tobacco plants which had been damaged 12 days before running the experiment. The response of wasps to plants with a 12-day-old hole was compared to undamaged plants. The response of wasps to visual cues alone was also tested by presenting wasps with two artificial plants made of green cardboard leaves attached to a wooden stick where the leaf of the damaged plant contained an 8-mm-diameter hole made with a cork borer compared to an artificial plant without a hole. One freshly killed larva was placed next to the artificial hole on the damaged plant and in the same spatial position on control plants. Because I observed that wasps seemed more likely to search holes in the leaf rather than damage located along the leaf edge and that larvae feeding along the leaf edge seemed more likely to escape detection, I performed an experiment to minimize the effect of visual searching cues by clipping the leaf along the edge. On damaged plants, one-third of a leaf was clipped off along the leaf margin with scissors immediately before running each trial. On the damaged plant, one freshly killed larva was placed on the clipped leaf and in the same spatial position on control plants. An experiment was performed to test the response of wasps to olfactory cues in the absence of visual cues. Two 8-mm-diameter holes were made just above the leaf petiole of the damaged plant immediately before running each trial and then the lower half of the leaf was covered with cheese cloth so that the holes would not be visible. The lower half of one leaf was also covered with cheese cloth in the control plants, but there were no holes. One freshly killed larva was placed on the exposed edge of the leaf for both damaged and control plants. Effect of Plant Species on Wasp Foraging Behavior

An experiment was performed to compare the foraging behavior of inexperienced wasps on tobacco and tomato plants in the absence of feeding damage. One plant of each species was set out in the greenhouse. A freshly killed T. ni larva (10-20 mm) was placed on each plant. Larvae were reared on alfalfa plants so that indirect effects of larval diet would not influence wasp behavior.

Cornelius

776

Wasps did not have any previous exposure to either plant species and were only used once in this experiment. The numbers of wasps that took larvae from tobacco and tomato were compared using a sign test. Observations were also made to determine if marked wasps returned to the same plant species during a later trial. RESULTS

Effect of Caterpillar-Feeding Damage on Wasp Foraging Behavior It was found that wasps took significantly more larvae from the previously damaged plants compared to the controls in the experiments with tobacco plants but that there was no difference between damaged and control plants in the experiments with tomato in both 1989 and 1990 (Table I). Comparisons of the time it took wasps to take larvae from damaged and undamaged plants show that wasps took significantly longer to find larvae on control plants than on damaged plants in the tobacco experiment. However, in the tomato experiment the length of time it took wasps to take larvae from damaged and control plants was not significantly different (Table II).

Effect of Visual and Olfactory Cues on Wasp Foraging Behavior Wasps took significantly more caterpillars from the freshly damaged plants compared to the control plants in both the tobacco (77 % of 39 trials) and the tomato (68% of 38 trials) experiments. In contrast, there was no significant difference in the number of larvae taken from plants with a 12-day-old hole and undamaged plants. Wasps took larvae from the damaged plants in only 37% of Table I. Number of M. sexta Larvae (Second- or Third-Instar) Taken by Wasps Searching for Larvae on Damaged and Undamaged Plants in Paired-Choice Tests in 1989 and 1990 Experiment 1989 Tobacco Tomato 1990 Tobacco Toma~

N

Damaged

Undamaged

15 12

12 5

3* 6

20 13

15 8

5* 5

(one tie) a

aOne trial was counted as a tie because the larva on the undamaged plant was taken less than 20 s before the larva on the damaged plant was taken. *Number of larvae taken was significantly different (P < 0.05, Sign test (Sokal and Rohlf, 1981).


777

38 trials. Also, wasps did not distinguish b e t w e e n leaves with holes and without holes w h e n they w e r e presented with larvae on artificial cardboard plants. Wasps t o o k larvae f r o m the artificial l e a f with a hole in 59 % o f 22 trials (Table III). H o w e v e r , it is possible that there w e r e other c o m p l i c a t i n g factors i n v o l v e d in using artificial plant structures. In the e x p e r i m e n t w h e r e leaves w e r e artificially d a m a g e d by cutting t h e m along the edge, wasps did not distinguish b e t w e e n d a m a g e d and u n d a m a g e d plants. W a s p s t o o k larvae f r o m the d a m a g e d plant in only 38% o f 13 trials (Table III). T h e r e was no significant difference b e t w e e n wasp response to d a m a g e d and u n d a m a g e d plants w h e n wasps w e r e presented with olfactory cues in the absence o f visual cues. In the cheese cloth experiment, wasps took larvae f r o m d a m a g e d plants in 57 % o f 56 trials (Table IlI).

Table II. Mean (+SE) Number of Minutes Spent by Wasps Searching for Larvae on Damaged and Undamaged Plants in Paired-Choice Tests in 1989a

Plant species

N

Damaged

Undamaged

P

Tobacco Tomato

15 12

6.8 _+ 2.9 30.6 _+ 9.3

21.0 _+ 4.7 22.8 _+ 9.3

0.02 0.56

aComparisons are made using Wilcoxon's signed-ranks test for matched pairs (Sokal and Rohlf, 1981).

Table IlL Number of Freshly Killed M. sexta Larvae (Second or Third Instar) Taken by Wasps in Paired-Choice Tests a

Experiment

N

Damaged

Undamaged

P

Fresh holeb Tobacco Tomato Old holec Cardboard d Leaf edgee Cheesecloth f

39 38 38 22 13 56

30 26 14 13 5 32

9 12 24 9 8 24

< 0.05 < 0.05 nsg ns ns ns

aComparisons are by a sign test (Sokal and Rohlf, 1981). bPlant damaged by cutting a hole in one leaf with a cork borer immediately before each trial. Experiment performed with both tobacco and tomato plants. CPlant damaged by cutting a hole in one leaf with a cork borer 12 days prior to test. Experiment performed with tobacco plants. dArtificial cardboard plant damaged by cutting a hole with a cork borer in a cardboard leaf. ePlant damaged by clipping one-third of one leaf off along the leaf margin with a scissors. fPlant damaged by cutting holes in one leaf with a cork borer and covering holes with cheesecloth. gNot significant.

778

Cornelius Effect of Plant Species on Wasp Foraging Behavior

There was no difference in wasp foraging behavior between tobacco and tomato plants in the absence of caterpillar feeding damage. Out of 42 individuals, 23 wasps took the larva on tobacco first and 19 wasps took the larva on tomato first. Of 20 marked individuals that returned to take larvae in later trials, 13 wasps took larvae from the same plant species and 7 took larvae from the other plant species in their second visit.

DISCUSSION The difference in wasp foraging behavior in experiments with tobacco and tomato indicates that leaf morphology may influence predator response to caterpillar feeding damage. Wasps with no previous exposure to tobacco or tomato plants showed no difference in their ability to find prey on the two plant species in the absence o f caterpillar feeding damage. However, wasps responded differently in experiments comparing damaged and control plants when using tobacco and tomato plants. Wasps did not distinguish between caterpillar-damaged tomato foliage and undamaged foliage; however, they did respond differently to tomato leaves that were damaged by cutting a hole in the center of a leaflet with a cork borer. In the experiments with tobacco plants, wasps did not respond differently to naturally and artificially damaged tobacco leaves, except when the leaves were cut along the margin. In that case, wasps did not distinguish between damaged and control plants. Thus, wasps may be searching for visually apparent holes in a leaf and may not be using damage along a leaf margin as a cue. Furthermore, wasps did not take more caterpillars from the damaged plants when the holes were covered with cheese cloth compared to controls, indicating that odors released from mechanically damaged leaves in the absence of visual cues do not significantly affect wasp foraging behavior. However, it is possible that the cheese cloth did not allow for good odor release. Results also indicated that wasps did not respond to visual cues in the absence of olfactory cues. When wasps were presented with tobacco plants that had a hole made 12 days earlier compared to a control, there was no difference in their searching behavior between the damaged and undamaged plants. However, when the same experiment was performed using a freshly made hole in a leaf compared to a control, wasps took 77 % of the larvae from damaged plants. The results of this study suggest that wasps respond to a combination of visual and olfactory cues but that, in isolation, neither one has a significant effect on their searching behavior. Evidence from other research shows that vespid wasps respond to both


779

visual and olfactory prey cues. Wasps are strongly attracted to the presence of caterpillar hemolymph on the substrate (Hirose and Takagi, 1980; Raveret Richter and Jeanne, 1985; Raveret Richter, 1988). Raveret Richter and Jeanne (1985) tested the response of wasps to visual and olfactory caterpillar cues. Wasps hovered over a descented visual model of a caterpillar, but they landed more often on sites where hemolymph was applied to the substrate. Observations indicate that wasps use visual cues to find prey. Jeanne (1972) observed foraging Mischocyttarus drewseni pounce on lumps of twigs and grass stems and barbs of barbed wire that may have appeared like prey. Duncan (1939) observed Dolichovespula maculata foragers pounce on nailheads on a wall where wasps were apparently searching for blowflies. Wasps use many factors in their search for prey, including caterpillar feeding damage. Wasps could potentially improve their searching behavior by using a combination of visual and olfactory plant cues to locate prey. By distinguishing between freshly damaged leaves and leaves damaged days earlier, wasps could increase the effectiveness of their search since they would be much more likely to find a caterpillar by searching freshly damaged leaves. Also, since some kinds of plant damage are much more likely to be associated with the presence of a caterpillar than others, wasp foragers could hunt more successfully by only responding to certain combinations of cues. Much of the previous work on the evolution of defensive strategies against visually hunting predators has focused on vertebrate predators such as birds (Heinrich, 1993). This study demonstrates that vespid wasps also use feeding damage as a cue in their search for prey and therefore may also be playing a role in the development of behavioral adaptations by lepidopteran larvae to reduce the evidence of feeding damage. Vespids can significantly reduce lepidopteran populations. Lawson et al. (1961) reported that Polistes wasps could kill 50 to 98 % of Manduca sexta larvae present in tobacco fields. Gould and Jeanne (1984) found that there were 62% fewer late=instar Pieris rapae larvae in plots where Polistes wasps were foraging than in control plots during one season. Hence, vespid wasps could potentially play an important role in the evolution of defensive strategies of lepidopteran larvae.

ACKNOWLEDGMENTS I thank Elizabeth Bernays, Ken Hagen, Jeremy Lee, and Clytia Montllor for helpful advice on my experimental design. I also thank Elizabeth Bernays, Clytia Montllor, and two anonymous reviewers for constructive criticism of the manuscript. This study was undertaken in partial fulfillment of the requirements for the Ph.D. degree in Entomology at the University of California, Berkeley.

780

Cornelius

REFERENCES Aldrich, J. R., Kochansky, J. P., and Sexton, J. D. (1985). Chemical attraction of the eastern yellow jacket, Vespula maculifrons (Hymenoptera: Vespidae). Experientia 41: 420-422. Bergelson, J. M., and Lawton, J. H. (1988). Does foliage damage influence predation on the insect herbivores of birch? Ecology 69: 434-445. Carter, M. C., Sutherland, D., and Dixon, A. F. G. (1984). Plant structure and the searching efficiency of coccinellid larvae. Oecologia 63: 394-397. Dicke, M., Sabelis, M. W., Takabayashi, J., Bruin, J., and Posthumus, M. A. (1990). Plant strategies of manipulating predator-prey interactions through allelochemicals: Prospects for application in pest control. J. Chem. Ecol. 16:3091-3118. Duncan, C. (1939). A contribution to the biology of North American vespine wasps. Stanford Univ. Publ. Biol. Sci. 1-272. Flint, H. M., Salter, S. S., and Walters, S. (1979). Carophyllene: An attractant for the green lacewing Chrysopa carnea Stephens. Environ. Entomol. 8:1123-1125. Gould, W. F., and Jeanne, R. L. (1984). Polistes wasps (Hymenoptera: Vespidae) as control agents for lepidopteran cabbage pests. Environ. Entomol. 13: 150-156. Greany, P. D., and Hagen, K. S. (1981). Prey selection. In Nordlund, D. A., Jones, R. L., and Lewis, W. J. (eds.), Semiochemicals: Their Role in Pest Control, Wiley, New York, pp. 121135. Grevstad, F. S., and Klepetka, B. W. (1992). The influence of plant architecture on the foraging effieiencies of a suite of ladybird beetles feeding on aphids. Oecologia 92: 399-404. Heinrich, B. (1979). Foraging strategies of caterpillars. Leaf damage and possible predator avoidance strategies. Oecologia 42: 325-337. Heinrich, B. (1993). How avian predators constrain caterpillar foraging. In Stamp, N. E., and Casey, T. M. (eds.), Caterpillars: Ecological and Evolutionary Constraints on Foraging, Chapman and Hall, New York, pp. 224-245. Heinrich, B., and Collins, S. L. (1983). Caterpillar leaf damage and the game of hide-and-seek with birds. Ecology 64: 592-602. Hirose, Y., and Takagi, M. (1980). Attraction of two species of Polistes wasp to prey wounded by them. Appl. Entomol. Zool. 15(1): 108-110. Jeanne, R. (1972). Social biology of the neotropical wasp Mischocynarus drewseni. Bull. Mus. Comp. ZooL Harvard Univ. 144(3): 63-150. Kareiva, P. M., and Sahakian, R. (1990). Tritrophic effects of a simple architectural mutation in pea plants. Nature 345: 433-434. Lawson, F. R., Rabb, R. L., Guthrie, F. E., and Bowery, T. G. (1961). Studies of an integrated control system for hornworms on tobacco. J. Econ. Entomol. 54(1): 93-97. Lederhouse, R. C. (1990). Avoiding the hunt: Primary defenses of lepidopteran caterpillars. In Evans, D. L., and Schmidt, J. O. (eds.), Insect Defenses, Adaptive Mechanisms and Strategies of Prey and Predators, SUNY Press, Albany, NY, pp. 175-189. Litte, M. (1979). Mischocyttarus flavitarsis in Arizona: Social and nesting biology of a polistine wasp. Z. Tierpsychol. 50: 282-312. Niemela, P., and Tuomi, J. (1987). Does the leaf morphology of some plants mimic caterpillar damage? Oikos 50: 256-257. Obrycki, J. J. (1986). The influence of foliar pubescence on entomophagous species. In Boethel, D. J., and Eikenbary, R. D. (eds.), Interactions of Plant Resistance and Parasitoids and Predators oflnsects, Ellis Horwood, Chichester, pp. 61-83. Rasmy, A. H. (1977). Predatory efficiency and biology of the predatory mite, Amblyseius gossipi (Acarina: Phytoseiidae) as affected by plant surfaces. Entomophaga 22(4): 421-423. Raveret Richter, M. (1988). Prey Hunting and Interactions Among Social Wasp (Hymenoptera: Vespidae) Foragers and Responses of Caterpillars to Hunting Wasps, Ph.D. dissertation, Cotnell University, Ithaca, NY. Raveret Richter, M. A., and Jeanne, R. L. (1985). Predatory behavior of Polybia sericea (Olivier), a tropical social wasp (Hymenoptera: Vespidae). Behav. Ecol. Sociobiol. 16" 165-170. Sokal, R. R., and Rohlf, F. J. (1981). Biometry, W. H. Freeman, New York.


781

Tietz, H. M. (1972). An Index to the Described Life Histories, Early Stages and Hosts of the Macrolepidoptera of the Continental United States and Canada, A. C. Allyn, Sarasota, FL, pp. 467-468. Turlings, T. C. J., Tumlinson, J. H., and Lewis, W. J. (1990). Exploitation of herbivore-induced plant odors by host-seeking parasitic wasps. Science 250: 1251-1253. Wood, D. L. (1982). The role of pheromones, kairomones, and allomones in the host selection and colonization behavior of bark beetles. Annu. Rev. Entomol. 27: 411-446. Yamamoto, R. T., and Fraenkel, G. S. (1960). The specificity of the tobacco homworm Protoparce sexta (Johan.) to solanaceous plants. Ann. EntomoL Soc. Am. 53: 503-507.