Do Caterpillars Secrete “Oral Secretions”?

The oral secretions or regurgitant of caterpillars contain potent elicitors of plant induced responses. These elicitors are recognized by host plants to differentiate between simple mechanical injury and the presence of herbivores. In some cases, this level of recognition is highly specific. Despite the in-depth chemical characterization of these elicitors, little is known about the amounts delivered in regurgitant during feeding. In this study, we use a fluorescent dye to label regurgitant in order to visualize caterpillar regurgitation during feeding. The procedure is highly sensitive and allows us to visualize nanoliter amounts of regurgitant. We examined the propensity of larval Helicoverpa zea, Heliothis virescens, Spodoptera exigua, Spodoptera frugiperda, and Manduca sexta to regurgitate on various host plants. These species were selected because they have been among the most intensely studied in terms of elicitors. Our results indicate that most larvae did not regurgitate following a brief feeding bout (∼10 min) during which they ate ca. 0.40 cm² of leaf. When larvae did regurgitate, it was typically less than 10 nl. This is several orders of magnitude less than is typically used in most studies on oral secretions. The frequency of regurgitation appears to vary depending upon the host plant. Larval H. zea are less likely to regurgitate when feeding on tomato leaves compared to corn mid-whorl tissue. Our results have importance in understanding the role of oral secretions in plant recognition of herbivory. Because caterpillars did not routinely regurgitate during feeding, it is likely that they avoid the elicitation of some plant defensive responses during most feeding bouts.     

Identification and Synthesis of Volicitin and Related Components from Beet Armyworm Oral Secretions

Oral secretion of beet armyworm caterpillars (BAW), when applied to damaged tissues of corn seedlings, induces the seedlings to emit volatile compounds that attract the natural enemies of the caterpillars. The key elicitor present in BAW oral secretions is N-[17-hydroxylinolenoyl]-L-glutamine (volicitin). Analysis of the oral secretion showed that it also contained N-[17-hydroxyolinoleoyl]-L-glutamine, free 17-hydroxylinolenic, and 17-hydroxylinoleic acid, the glutamine conjugates of linolenic and linoleic acid as well as free linolenic and linoleic acid. Here we present the identification and synthesis of the hydroxy acids and of glutamine conjugates.     

Induction of parasitoid attracting synomone in brussels sprouts plants by feeding ofPieris brassicae larvae: Role of mechanical damage and herbivore elicitor

Induction of plant defense in response to herbivory includes the emission of synomones that attract the natural enemies of herbivores. We investigated whether mechanical damage to Brussels sprouts leaves (Brassica oleracea var.gemmifera) is sufficient to obtain attraction of the parasitoidCotesia glomerata or whether feeding byPieris brassicae caterpillars elicits the release of synomones not produced by mechanically damaged leaves. The response of the parasitoidCotesia glomerata to different types of simulated herbivory was observed. Flight-chamber dual-choice tests showed that mechanically damaged cabbage leaves were less attractive than herbivore-damaged leaves and mechanically damaged leaves treated with larval regurgitant. Chemical analysis of the headspace of undamaged, artificially damaged, caterpillar-infested, and caterpillar regurgitant-treated leaves showed that the plant responds to damage with an increased release of volatiles. Greenleaf volatiles and several terpenoids are the major components of cabbage leaf headspace. Terpenoids are emitted in analogous amounts in all treatments, including undamaged leaves. On the other hand, if the plant is infested by caterpillars or if caterpillar regurgitant is applied to damaged leaves, the emission of green-leaf volatiles is highly enhanced. Our data are in contrast with the induction of more specific synomones in other plant species, such as Lima bean and corn.     

In vivo Volatile Emissions from Peanut Plants Induced by Simultaneous Fungal Infection and Insect Damage

Peanut plants, Arachis hypogaea, infected with white mold, Sclerotium rolfsii, emit a blend of organic compounds that differs both quantitatively and qualitatively from the blend emitted from plants damaged by beet armyworm (BAW; Spodoptera exigua) larvae or from uninfected, undamaged plants. Attack by BAW induced release of lipoxygenase products (hexenols, hexenals, and hexenyl esters), terpenoids, and indole. The plant-derived compound methyl salicylate and the fungal-derived compound 3-octanone were found only in headspace samples from white mold infected plants. White mold-infected plants exposed to BAW damage released all the volatiles emitted by healthy plants fed on by BAW in addition to those emitted in response to white mold infection alone. When BAW larvae were given a choice of feeding on leaves from healthy or white mold-infected plants, they consumed larger quantities of the leaves from infected plants. Exposure to commercially available (Z)-3 hexenyl acetate, linalool, and methyl salicylate, compounds emitted by white mold-infected plants, significantly reduced the growth of the white mold in solid-media cultures. Thus, emission of these compounds by infected plants may constitute a direct defense against this pathogen.     

Volicitin, An Elicitor of Maize Volatiles in Oral Secretion of Spodoptera Exigua: Isolation and Bioactivity

Plants respond to insect-inflicted injury by systemically releasing relatively large amounts of several volatile compounds, mostly terpenoids and indole. As a result, the plants become highly attractive to natural enemies of the herbivorous insects. In maize, this systemic response can be induced by the uptake via the stem of an elicitor present in the oral secretions of caterpillars. Such an elicitor was isolated from the regurgitant of Spodoptera exigua larvae, identified as N-(17-hydroxylinolenoyl)-L-glutamine, and named volicitin. Here we present details on the procedure that was used to isolate volicitin and the biosasays that demonstrate its potency as an elicitor of maize volatiles that attract parasitoids. With a series of liquid chromatography purification steps, volicitin was separated from all other inactive substances in the regurgitant of larvae of the noctuid moth S. exigua. Maize seedlings that were incubated in very low concentrations of pure natural volicitin released relatively large amounts of terpenoids and became highly attractive to the parasitoid Microplitis croceipes. The identification of this and other insect-derived elicitors should allow us to determine their precise source and function, and better understand the evolutionary history of the phenomenon of herbivore-induced volatile emissions in plants.     

Herbivore-induced volatile emissions from cotton (Gossypium hirsutum L.) seedlings

The effect of herbivory on the composition of the volatile blends released by cotton seedlings was investigated by collecting volatiles from undamaged, freshly damaged (0–2 hr after initiation of feeding), and old damaged (16–19 hr after initiation of feeding) plants on which corn earworm caterpillars (Helicoverpa zea Boddie) were actively feeding. A blend of 22 compounds was consistently observed to be emitted by the old damaged plants with nine occurring either only in, or in significantly greater amounts in old damaged, as compared with freshly damaged plants. These were (Z)-3-hexenyl acetate, hexyl acetate, (E)--ocimene, (3E)-4,8-dimethyl-1,3,7-nonatriene, (Z)-3-hexenyl butyrate, (E)-2-hexenyl butyrate, (Z)-3-hexenyl 2-methylbutyrate, (E)-2-hexenyl 2-methylbutyrate, and indole. The nature of this response is compared with other studies where herbivore-induced volatile responses are also known. The presence of large amounts of terpenes and aldehydes seen at the onset of feeding and the appearance of other compounds hours later suggest that cotton defense mechanisms may consist of a constitutive repertoire that is augmented by an induced mechanism mobilized in response to attack. A number of the induced compounds are common to many plants where, in addition to an immediate defensive function, they are known to be involved in the attraction of natural enemies.     

Volatile Emissions Triggered by Multiple Herbivore Damage: Beet Armyworm and Whitefly Feeding on Cotton Plants

Plants are commonly attacked by more than one species of herbivore, potentially causing the induction of multiple, and possibly competing, plant defense systems. In the present paper, we determined the interaction between feeding by the phloem feeder silverleaf whitefly (SWF), Bemisia tabaci Gennadius (B-biotype = B. argentifolii Bellows and Perring), and the leaf-chewing beet armyworm (BAW), Spodoptera exigua Hübner, with regard to the induction of volatile compounds from cotton plants. Compared to undamaged control plants, infestation with SWF did not induce volatile emissions or affect the number and density of pigment glands that store volatile and nonvolatile terpenoid compounds, whereas infestation by BAW strongly induced plant volatile emission. When challenged by the two insect herbivores simultaneously, volatile emission was significantly less than for plants infested with only BAW. Our results suggest that tritrophic level interactions between cotton, BAW, and natural enemies of BAW, that are known to be mediated by plant volatile emissions, may be perturbed by simultaneous infestation by SWF. Possible mechanisms by which the presence of whiteflies may attenuate volatile emissions from caterpillar-damaged cotton plants are discussed.     

Volatile Stimuli Related to Feeding Activity of Nonprey Caterpillars, Spodoptera exigua, Affect Olfactory Response of the Predatory Mite Phytoseiulus persimilis

The effect of volatiles related to feeding activity of nonprey caterpillars, Spodoptera exigua, on the olfactory response of the predatory mites Phytoseiulus persimilis was examined in a Y-tube olfactometer. At a low caterpillar density (20 caterpillars on 10 Lima bean leaves), the predators were significantly more attracted to volatiles from infested leaves on which the caterpillars and their products were present or from infested leaves from which the caterpillars and their products had been removed when compared to volatiles from uninfested leaves. The predators, however, significantly avoided odors from 20 caterpillars and their products (mainly feces) removed from bean leaves. In contrast, at a higher caterpillar density (100 caterpillars on 10 Lima bean leaves), the predators avoided volatiles from caterpillar-infested bean leaves. Volatiles from infested leaves from which the caterpillars and their products had been removed were not preferred over volatiles from uninfested leaves. Volatiles from feces collected from 100 caterpillars were strongly avoided by the predators, while the behavior of the predatory mites was not affected by volatiles from 100 caterpillars removed from a plant. The data show that carnivorous arthropods may avoid nonprofitable herbivores. This avoidance seems to result from an interference of volatiles from herbivore products with the attraction to herbivore-induced plant volatiles.     

Potato Expressing Beetle-Specific Bacillus thuringiensis Cry3Aa Toxin Reduces Performance of a Moth

Expression of the Bacillus thuringiensis beetle-specific toxin Cry3Aa, which renders a genetically modified potato cultivar resistant to the Colorado potato beetle Leptinotarsa decemlineata, exerts a deleterious effect on the polyphagous moth Spodoptera littoralis. The caterpillars of S. littoralis feed less and produce smaller pupae on the genetically modified cultivar (NewLeaf Superior) than on the parental nontransgenic cultivar (Superior). The conversion efficiencies of total dry matter, combustion heat, carbon, and nitrogen from leaves to insect biomass are similar on both cultivars. In spite of similar food utilization and a relatively small difference in the body mass at pupation, female adults that developed from caterpillars fed on NewLeaf Superior lay a mean of 309 eggs compared to a mean of 713 eggs deposited by females that developed from caterpillars fed on Superior. Because of this difference and a simultaneous reduction in fertility (egg hatchability) from 78 to 48%, a pair of adults that fed as larvae on NewLeaf Superior produces only 148 larvae, whereas a pair of adults that fed as larvae on Superior produces 556 larvae. We suggest that small amounts of Cry3Aa that accumulate in insect tissue and persist until the adult stage are responsible for the decline in reproduction.     

Induction of volatile emissions in maize by different larval instars of Spodoptera littoralis

Maize plants under attack by caterpillars emit a specific blend of volatiles that is highly attractive to parasitic wasps. The release of these signals is induced by elicitors in the caterpillar regurgitant. Studies suggest that plants respond differently to different herbivore species and even to different herbivore stages, thus providing parasitoids and predators with specific signals. We tested if this is the case for different larval instars of the noctuid moth Spodoptera littoralis when they feed on maize plants. Cut maize plants were incubated in diluted regurgitant from second, third, or fifth instar caterpillars. There were no differences in total amount released after these treatments, but there were small differences in the release of the minor compounds phenethyl acetate and -humulene. Regurgitant of all three instars contained the elicitor volicitin. To test the effect of actual feeding by the larvae, potted plants were infested with caterpillars of one of the three instars, and volatiles were collected the following day. The intensity of the emissions was correlated with the number of larvae feeding on a plant, and with the amount of damage inflicted, but was independent of the instar that caused the damage. We also used artificial damage to mimic the manner of feeding of each instar to test the importance of physical aspects of damages for the odor emission. The emission was highly variable, but no differences were found among the different types of damage. In olfactometer tests, Microplitis rufiventris, a parasitoid that can only successfully parasitize second and early third instar S. littoralis, did not differentiate among the odors of maize plants attacked by different instar larvae. The odor analyses as well as the parasitoid's responses indicate that maize odors induced by S. littoralis provide parasitoids with poor information on the larval developmental stage. We discuss the results in the context of variability and lack of specificity in odorous plant signals.     

Influence ofAmaranthus hybridus L. allelochemics on oviposition behavior ofSpodoptera exigua andS. eridania (Lepidoptera: Noctuidae)

Common pigweed,Amaranthus hybridus L., is a favorite host of the beet army worm (BAW),Spodoptera exigua L. Chemicals extracted from homogenized pigweed with distilled water, ethanol, or dichloromethane and sprayed back on pigweed deterred oviposition by the BAW. Similarly, water extracts of frass from conspecific larvae or southern armyworm (SAW) larvae,S. eridania (Cramer), fed pigweed leaves and sprayed back on pigweed plants also deterred BAW oviposition, thus confirming that deterrence was due to plant allelochemics rather than specific compounds associated with the metabolic or excretory products of the larvae. Confirmation of the presence of oviposition-deterring chemicals in pigweed was used to explain a previously observed seasonal displacement of BAW by SAW on pigweed in the field.Mention of a commercial or proprietary product does not constitute an endorsement by the USDA.     

Induction of Plant Volatiles by Herbivores with Different Feeding Habits and the Effects of Induced Defenses on Host-Plant Selection by Thrips

Induced plant responses to attack by chewing insects have been intensively studied, but little is known about plant responses to nonchewing insects or to attack by multiple herbivores with different feeding habits. We examined volatile emissions by tobacco, Nicotiana tabacum, in response to feeding by the piercing–sucking insect western flower thrips (WFT), Frankliniella occidentalis, the chewing herbivore Heliothis virescens, and both herbivores simultaneously. In addition, we examined the effects of herbivore-induced plant defenses on host-plant selection by WFT. Plants responded to thrips feeding by consistently releasing five compounds. Simultaneous feeding by WFT and H. virescens elicited the same 11 compounds emitted in response to caterpillar feeding alone; however, two compounds, α-humulene and caryophyllene oxide, were produced in greater amounts in response to simultaneous herbivory. In choice tests, thrips consistently preferred uninduced plants over all other treatments and preferred plants damaged by caterpillars and those treated with caterpillar saliva over those treated with caterpillar regurgitant. The results are consistent with a previous finding that caterpillar regurgitant induces the release of significantly more volatile nicotine than plants damaged by caterpillars or plants treated with caterpillar saliva. A repellent effect of nicotine on WFT was confirmed by encircling unwounded plants with septa releasing volatile nicotine. Our results provide the first direct evidence that thrips feeding induces volatile responses and indicates that simultaneous herbivory by insects with different feeding habits can alter volatile emissions. In addition, the findings demonstrate that induced plant responses influence host-plant selection by WFT and suggest that the induction of volatile nicotine may play a role in this process.     

Protective action of midgut catalase in lepidopteran larvae against oxidative plant defenses

Catalase activity was detected in the midgut tissues and regurgitate of several lepidopteran pests of the tomato plant. Greatest activity in the midgut was detected in larvalHelicoverpa zea, followed bySpodoptera exigua, Manduca sexta, andHeliothis virescens. We present evidence that catalase, in addition to removing toxic hydrogen peroxide, may inhibit the oxidation of plant phenolics mediated by plant peroxidases. Small amounts of larval regurgitate significantly inhibited foliar peroxidase activity via removal of hydrogen peroxide. Treatment of foliage with purified catalase nearly eliminated peroxidase activity and was superior as a larval food source compared to untreated foliage. Tomato foliar peroxidases oxidize an array of endogenous compounds including caffeic acid, chlorogenic acid, rutin, coumaric acid, cinnamic acid, and guaiacol. The oxidized forms of these compounds are potent alkylators of dietary and/or cellular nucleophiles (e.g., thiol and amino functions of proteins, peptides, and amines). When tomato foliar protein was pretreated with peroxidase and chlorogenic acid and incorporated in artificial diet, larval growth was reduced compared to larvae fed untreated protein. Thus, the diminution of peroxidase activity and removal of hydrogen peroxide by catalase may represent an important adaptation to leaf-feeding. The secretion of catalase in salivary fluid during insect feeding is also suggested to be a potential mechanism for reducing hydrogen peroxide formation as an elicitor of inducible plant defenses.     

Systemically Induced Plant Volatiles Emitted at the Time of “Danger”

Feeding by Pieris brassicae caterpillars on the lower leaves of Brussels sprouts (Brassica oleracea var. gemmifera) plants triggers the release of volatiles from upper leaves. The volatiles are attractive for a natural antagonist of the herbivore, the parasitoid Cotesia glomerata. Parasitoids are attracted only if additional damage is inflicted on the systemically induced upper leaves and only after at least three days of herbivore feeding on the lower leaves. Upon termination of caterpillar feeding, the systemic signal is emitted for a maximum of one more day. Systemic induction did not occur at low levels of herbivore infestation. Systemically induced leaves emitted green leaf volatiles, cyclic monoterpenoids, and sesquiterpenes. GC-MS profiles of systemically induced and herbivore-infested leaves did not differ for most compounds, although herbivore infested plants did emit higher amounts of green leaf volatiles. Emission of systemically induced volatiles in Brussels sprouts might function as an induced defense that is activated only when needed, i.e., at the time of caterpillar attack. This way, plants may adopt a flexible management of inducible defensive resources to minimize costs of defense and to maximize fitness in response to unpredictable herbivore attack.     

Consumption and utilization of experimentally altered corn by southern armyworm: Iron,nitrogen, and cyclic hydroxamates

The effects of differential leaf water, leaf nitrogen and cyclic hydroxamate (DIMBOA) concentrations in corn seedlings were analyzed for a polyphagous insect, the southern armyworm (Spodoptera eridania Cram.). Six different combinations of nutrients and allelochemicals [DIMBOA = 2,4-dihydroxy-7-methoxy(2H)-benzoxazin-3(4H)-one] were generated using two corn genotypes (WF9 and CI3IA) and three fertility regimes (complete nutrient, Fe-deficient, and N-deficient solutions) in the University Biotron. Poorest larval growth was observed in the low-nitrogen treatments (1.2% and 1.7% leaf N) and was the result of both low consumption rates and high metabolic costs (low efficiency of conversion of digested food, ECD). Fastest growth rates were observed forthe larvae fed leaves from the high-nitrogen treatments (4.6% and 4.4% leaf N). It is noteworthy that these treatments also contained the highest concentration of cyclic hydroxamates, which are generally believed to be the primary defensive chemicals mediating resistance against the European corn borer,Ostrinia nubilalis (Hubner). If these hydroxamates do have any deleterious or costly effects (perhaps accounting for a large portion of metabolic expenditures), the high digestibility of the leaf tissue and the increased consumption rates more than compensate, resulting in rapid growth (growth rate = consumption rate × approximate digestibility × efficiency of conversion of the digested food). These studies illustrate that variation in key nutrients and allelochemicals within a single plant species (Zea mays L.) may have significantly different effects upon various potential leaf-chewing caterpillars, such as these armyworms versus corn borers (which cannot handle the cyclic hydroxamates, even if provided with young nutritious leaf tissues).     

Plant-natural enemy association in the tritrophic system,Cotesia rubecula-Pieris rapae-brassiceae (cruciferae): I. Sources of infochemicals

The role of airborne infochemicals in host selection by the parasitoidCotesia rubecula (Marshal) (Hymenoptera: Braconidae) was examined in a wind tunnel. To elucidate the role of volatile chemicals in attractingC. rubecula to cabbage infested by the host [Pieris rapae L. (Lepidoptera: Pieridae)], the potential sources of volatiles related toP. rapae infestation on cabbage were tested individually. The responses of females to nonhost plant species, bean and geranium, as well as to frass of a nonhost lepidopteran were also examined.C. rubecula was attracted to cabbage previously infested byP. rapae and to frass and regurgitate ofP. rapae. No attraction was observed to larvae ofP. rapae alone. Females were also attracted to mechanically damaged cabbage, cabbage previously infested byPlutella xylostella L. (Lepidoptera: Plutellidae) (a nonhost lepidopteran herbivore), and cabbage previously infested by snails (a nonhost, noninsect herbivore). Intact cabbage, bean, and geranium plants elicited no attraction. A low frequency of attraction was observed to mechanically damaged bean and geranium. Attraction was also observed to frass ofP. xylostella. Volatiles from cabbage related to damage, and volatiles from frass and regurgitate of the host seem to play an important role in guidingC. rubecula to plants infested by its host.     

Herbivore-Induced Plant Volatiles Mediate In-Flight Host Discrimination by Parasitoids

Herbivore feeding induces plants to emit volatiles that are detectable and reliable cues for foraging parasitoids, which allows them to perform oriented host searching. We investigated whether these plant volatiles play a role in avoiding parasitoid competition by discriminating parasitized from unparasitized hosts in flight. In a wind tunnel set-up, we used mechanically damaged plants treated with regurgitant containing elicitors to simulate and standardize herbivore feeding. The solitary parasitoid Cotesia rubecula discriminated among volatile blends from Brussels sprouts plants treated with regurgitant of unparasitized Pieris rapae or P. brassicae caterpillars over blends emitted by plants treated with regurgitant of parasitized caterpillars. The gregarious Cotesia glomerata discriminated between volatiles induced by regurgitant from parasitized and unparasitized caterpillars of its major host species, P. brassicae. Gas chromatography-mass spectrometry analysis of headspace odors revealed that cabbage plants treated with regurgitant of parasitized P. brassicae caterpillars emitted lower amounts of volatiles than plants treated with unparasitized caterpillars. We demonstrate (1) that parasitoids can detect, in flight, whether their hosts contain competitors, and (2) that plants reduce the production of specific herbivore-induced volatiles after a successful recruitment of their bodyguards. As the induced volatiles bear biosynthetic and ecological costs to plants, downregulation of their production has adaptive value. These findings add a new level of intricacy to plant–parasitoid interactions.     

Physiological, Nutritional, and Biochemical Bases of Corn Resistance to Foliage-Feeding Fall Armyworm

Three corn (Zea mays) germplasm lines [i.e., Ab24E (susceptible control), Mp708 (resistant control), and a locally selected partial inbred line FAW7050 (resistant)] were examined for Spodoptera frugiperda (J.E. Smith; Lepidoptera: Noctuidae) resistance. Nutritional [i.e., total protein content, amino acids, glucose, total nonstructural carbohydrates (TNC), protein to TNC (P/C) ratios] and biochemical (i.e., peroxidase and lipoxygenase 3) properties in the seedlings of these corn lines were examined to categorize resistance mechanisms to S. frugiperda. Physiological changes in photosynthetic rates also were examined in an attempt to explain nutritional and biochemical dynamics among corn germplasm lines and between insect-infested and noninfested corn plants within a germplasm line. Results indicated that S. frugiperda larvae survived better and developed faster in susceptible Ab24E than in resistant FAW7050 or Mp708. The three germplasm lines differed in resistance mechanisms to S. frugiperda, and the observed patterns of resistance were probably collective results of the P/C ratio and defensive proteins. That is, the susceptibility of Ab24E to S. frugiperda was due to a high P/C ratio and a low level of induced defensive compounds in response to insect herbivory, while the resistance of FAW7050 resulted from elevated defensive proteins following insect herbivory, low P/C ratio, and elevated defensive proteins in Mp708 contributed to its resistance to S. frugiperda. The elevated protein amounts in resistant Mp708 and FAW7050 following S. frugiperda injury were likely due to greater conversion of photosynthates to defensive proteins following the greater photosynthetic rates in these entries. Greater photosynthetic capacity in Mp708 and FAW7050 also led to higher amino acid and glucose contents in these two lines. Neither amino acid nor lipoxygenase 3 played a critical role in corn resistance to S. frugiperda. However, high inducibility of peroxidase may be an indicator of S. frugiperda susceptibility as observed elsewhere.     

Increased Terpenoid Accumulation in Cotton (<Emphasis Type="Italic">Gossypium hirsutum</Emphasis>) Foliage is a General Wound Response

The subepidermal pigment glands of cotton accumulate a variety of terpenoid products, including monoterpenes, sesquiterpenes, and terpenoid aldehydes that can act as feeding deterrents against a number of insect herbivore species. We compared the effect of herbivory by Spodoptera littoralis caterpillars, mechanical damage by a fabric pattern wheel, and the application of jasmonic acid on levels of the major representatives of the three structural classes of terpenoids in the leaf foliage of 4-week-old Gossypium hirsutum plants. Terpenoid levels increased successively from control to mechanical damage, herbivory, and jasmonic acid treatments, with E-β-ocimene and heliocide H₁ and H₄ showing the highest increases, up to 15-fold. Herbivory or mechanical damage to older leaves led to terpenoid increases in younger leaves. Leaf-by-leaf analysis of terpenes and gland density revealed that higher levels of terpenoids were achieved by two mechanisms: (1) increased filling of existing glands with terpenoids and (2) the production of additional glands, which were found to be dependent on damage intensity. As the relative response of individual terpenoids did not differ substantially among herbivore, mechanical damage, and jasmonic acid treatments, the induction of terpenoids in cotton foliage appears to represent a non-specific wound response mediated by jasmonic acid. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.