Scott R. Smedley, Elizabeth Ehrhardt, and Thomas Eisner.
Defensive regurgitation by a noctuid moth larva (Litoprosopus futilis).
Psyche 100(3-4):209-221, 1993.

This article at Hindawi Publishing: https://doi.org/10.1155/1993/67950
CEC's scan of this article: http://psyche.entclub.org/pdf/100/100-209.pdf, 3016K
This landing page: http://psyche.entclub.org/100/100-209.html

The following unprocessed text is extracted automatically from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.

DEFENSIVE REGURGITATION BY
A NOCTUID MOTH LARVA (LITOPROSOPUS FUTILIS)' BY SCOTT R. SMEDLEY, ELIZABETH EHRHARDT~, AND THOMAS EISNER
Section of Neurobiology and Behavior
Cornell University
Ithaca, NY 14853
Larvae of the noctuid moth Litoprosopus futilis regurgitate when disturbed. The oral effluent proved deterrent to ants on near- contact, and topically irritating in a scratch test with a cockroach. Larvae regurgitated when attacked by lycosid spiders and derived some protection from this behavior. Caterpillars were able to regurgitate even when emerging from the eggs; however, at this stage, they proved vulnerable to attack by chrysopid larvae and ants.
Regurgitation of gut contents is a common defensive strategy of insects, including orthopterans, isopterans, coleopterans, lepi- dopteran~, dipterans, and hymenopterans (reviewed by Eisner 1970, 1980; Blum 1981 ; Brower 1984; Whitman et al. 1990; Bow- ers 1993). Repellent fluids are disgorged by both herbivorous (Eis- ner 1970, Eisner et al.
1974, Morrow et al. 1976, Peterson et al. 1987, Brower 1984, Smedley et al. 1990) and insectivorous larvae (Eisner et al.
1980). Foregut diverticula for storage of the expellable materials can be a morphological feature associated with this defense (Eisner et al. 1974, Morrow et al. 1976, Common & Bellas 1977), but such specialization is not requisite (Eisner et al. 1980, Brower 1984).
' Paper no. 121 in the series Defense Mechanisms of Arthropods; no. 120 is Eisner et al., Proc. Nat. Acad. Sci. 90:6716 (1993). Current address: University of Wisconsin Medical School, 1300 University Ave., Madison, WI 53706,
Manuscript received 19 July 1993.




================================================================================

210 Psyche [vo~. 100
We here show that larvae of the noctuid moth Litoprosopus futilis (Grote & Robinson), occurring in the southeastern United States, regurgitate when disturbed. We demonstrate by bioassays of the larval vomit and observations of encounters of the caterpil- lars with sympatric predators, that the emitted fluid has repellent and irritant properties. Further, we note that the larva is able to regurgitate even as it emerges from the egg, before onset of feed- ing on the host plant.
MATERIALS AND METHODS
Field Observations
Field observations were made at the Archbold Biological Sta- tion, Lake Placid, Highlands County, Florida, USA, where we found larvae of L. futilis feeding on saw palmetto (Serenoa repens). During 21-23 March 1989, we observed the behavior of larvae on two S. repens plants (Southern Ridge Sand Hill habitat on Red Hill). Larvae for experimental use were collected from other S. repens plants. Field-collected larvae and reared adults are deposited in the Cornell University Insect Collection, voucher lot # 1219.
Larval Vomit and Ant Repellency Bioassay Larval regurgitant was presented in microcapillary tubes at close range to ants feeding at a sucrose solution bait, and its repel- lent effect was scored relative to that of controls (tap water). These tests were done at natural foraging trails of the ant Para- trechina longicornis (Formicinae). A plastic plate with four coni- cal feeding wells (positioned at the corners of an imaginary square, 1.3 cmlside) replete with sucrose solution (10% aqueous) served to lure the ants. After the ants had gathered to feed at the wells, two wells were randomly selected, and the number of ants at each of the two was recorded.
Regurgitant was collected by pinching individual, recently-fed, mid- to late-instar caterpillars gently between a finger and glass microscope slide. The effluent was taken up in a microcapillary tube. The fluid was then squeezed from the tube with a rubber bulb until it collected as a suspended drop at the opposite end of the tube. Presentation to ants involved positioning the capillary tube



================================================================================

19931 Smedley, Ehrhardt & Eisner 21 1
vertically above (1-3 mm) the assigned well. The tap water control (similarly presented as a suspended droplet) was brought into posi- tion above its designated well at the same time as the regurgitant sample. After 5 sec of such paired presentation, the ants at the two wells were again counted.
It was noted that the regurgitant gradually changed in color from green to brown following discharge. A regurgitant sample was thus tested both fresh (within 10 sec after emission) and aged (after it remained in the tube for 3-4 min). Twenty-four presenta- tions were performed, 12 with fresh and 12 with aged material. A sample's repellency was scored as the difference in the num- ber of ants before and after presentation. This scoring procedure was justified since the number of ants per well prior to sample pre- sentation did not vary between the vomits and their controls: mean å± S. E. antslwell for fresh regurgitant and its control = 4.1k0.5 and 3.8k0.5, respectively, (paired tn = 0.58, p = 0.57); aged regur- gitant and its control = 3.8k0.4 and 3.9k0.5, respectively, (paired t,, = -0.12, p = 0.90). The repellency of the fresh and aged regur- gitant was compared to that of the respective controls by means of paired t-tests. To maintain an experiment-wide a = 0.05, the resul- tant probabilities were adjusted using the sequential Bonferroni procedure (Rice 1989).
Cockroach Scratch Bioassay
Topical application of chemical irritants induces scratch reflexes in the cockroach Periplaneta americana (Eisner et al. 1976). The technique has been used for assay of irritancy of defen- sive glandular products of insects. Decapitated roaches are used for this purpose, since these are non-ambulatory. To assess the irri- tancy of the regurgitant of L. futilis, twenty last instar P. ameri- cana nymphs were tested as per protocol in Eisner et al. (1976). Only fresh regurgitant was used, collected as for the preceding assay. Deionized water provided the control. Each cockroach was tested with both regurgitant and control. An interval of several minutes transpired between the two applications. The droplet vol- ume (0.5 pL) was the same for both samples, and was applied either to the left or right half of the fourth abdominal tergite. The application procedure controlled for treatment sequence (regurgi- tant or control first) and position (right or left side). The criterion for response was scratching with the hindleg directed at the site of



================================================================================

212 Psyche [vo~. 100
application within 30 sec of droplet delivery. Individuals not responding within 30 sec were scored as non-respondent. The response of roaches to fresh vomit and to the control was compared using 2 X 2 contingency test [log-likelihood ratio (G- statistic) with the Williams correction (Sokal & Rohlf 1% I)]. Encounters of Larvae with Lycosid Spiders To examine the interactions of L. futilis larvae with spiders, lycosids (Lycosa spp., including L. ceratiola) were collected at the Archbold Biological Station and placed in individual sand-lined containers (10 X 10 cm base; 6 cm height). The lycosids were maintained without food for one or two days before trials. A larva was placed with a spider and interactions were noted. Encounters of Emerging Larvae with Chrysopid Larvae and Ants Encounters between partially eclosed L. fufilis caterpillars and predatory chrysopid larvae (Ceraeochrysa cubana) were examined. Field-collected egg masses were used. A portion of frond bearing an egg cluster with emerging larvae was placed in a 15 X 50 mm petri dish and observed under a stereomicroscope. A chrysopid larva was then introduced to this arena. Chrysopid predation attempts and L. futilis defensive behavior were noted. Two such trails were performed.
A frond bearing an egg mass with emerging larvae was taped to a microscope slide and placed in the path of ants [P. longicornis and Solenopsis invicta (Myrmicinae)] that had laid trails to bait (10% sucrose solution or chopped mealworm pieces). Maternal hairs covering the eggs (Fig. 1) were removed to facilitate observa- tion. The reaction of the emerging larvae to ants was monitored with a stereomicroscope.
General Observations
At night the two aggregations of L. futilis larvae (7 and 11 caterpillarslaggregation) were visibly feeding on the unopened flower buds of S. repens. With the onset of daylight, larvae sought dark refuges (frond petiole bases, sheaths of flower petioles, or patches of frass formed by a pyralid moth larva on petiole stalk). Often larvae would enter and leave several refuges before final



================================================================================

19931 Smedley, Ehrhurdt & Eisner 2 13
Figure 1. (Top) Natural tgg clustcr of Li~rjpro.~opr j%i~/ts. The hwrq covering the eggs are prewmably detached modifid scales from the mother's body (Bar = 1 mm). (Bottom) Eggs in thc process of hatching; note mmdihles agape on larva in center foreground.




================================================================================

214 Psyche [vo~. 100
disappearance. During daylight hours, no larvae were seen on the developing inflorescences or other plant parts. At dusk caterpillars emerged from their shelters and began to feed. Cannibalism was observed among larvae taken from the field and kept collectively in the laboratory. Larval Vomit and Ant Repellency Bioassay The vomit of L. futilis repelled P: longicornis. Fewer ants remained after presentation of fresh (paired tll = 2.97, p = 0.026), but not aged vomit (paired tll = 0.30, p = 0.767) (Fig. 2). Cockroach Scratch Reflex Bioassay
Fresh vomit proved an irritant to P. americana. Vomit elicited a scratch reflex much more frequently than did the distilled water Vomit
Tap Water Control
Fresh
Fluid Tested
Figure 2. Repellency of fluid stimulant (L. futilis vomit or control) to ants (P. longicornis), as a function of fluid age. Mean repellency (# ants before - # ants after exposure to stimulant) + 1 S.E.; n = 12 paired presentations of vomit and con- trol for each of the two sets of data. Additional details in text.



================================================================================

19931 Smedley, Ehrhardt & Eisner 215
control (Gadj = 24.52, d.f. = 1, p < 0.001; Table 1). For roaches responding to vomit, scratch initiation time was 5.7 k1.6 sec. (mean ~1 S.E., n = 17). Three nymphs responded neither to vomit nor control, but none responded to the control alone. Encounters of Larvae with Lycosid Spiders In many trials (18135) spiders did not actively pursue the cater- pillars. When attacks did take place, they occurred almost immedi- ately after larval introduction. In 10 of the 17 instances where spiders seized caterpillars, the larva vomited upon the spider directly or on the sand in its vicinity (in the latter cases, we might have failed to notice fluid that contacted the spider). In most instances following larval regurgitation (7110), the spider cleaned its mouthparts by wiping them in the sand, a typical response to irritants (Eisner et al. 1972). Caterpillars that did not regurgitate when attacked (7117) did, however, raise their front end as they characteristically do prior to regurgitation. The outcome of an attack was dependent, at least in part, on the size relationship of the contenders. Large spiders (2.0-2.3 cm body length) successfully killed mid- to late-instar larvae (3 killed13 attacked), whereas medium-sized spiders (1 -0- 1.5 cm) were more successful in attacks on early (6 killed18 attacked) than on late instar (0 killed16 attacked) larvae. Five of the late instar larvae that escaped following attack appeared healthy 48 hr after the encounter.
Encounters of Emerging Larvae with Chrysopid Larvae and Ants L, futilis larvae that were emerging from eggs (Fig. 1) displayed mandibular spreading and regurgitation when stimulated with for- ceps (Fig. 3) or when attacked by larval chrysopids (Fig. 4). This Table 1. 2 X 2 contingency table examining topical irritancy of fresh L. futilis vomit as assessed in the cockroach (P. americana) scratch bioassay. Applied Number of Cockroaches
Fluid Respondent Non-Respondent Totals
Vomit 17 3 20
Control 2 18 20
Totals 19 2 1 40




================================================================================

216 Psyche
Figure 3. ('t'op) Emergmi lama ~cspondlng 70 pi-oddmg wth forceps by biting with its mandibles. C30tmm) Same, showing larva regnrgitating directly upon the con~actlng forceps.




================================================================================

Smedley, Ehrhardt & Eisner
defensive behavior, however, did not prevent the chrysopids from killing and eating such larvae. The chrysopids also consumed unhatched eggs.
Two of three larvae that had recently emerged from eggs regur- gitated when probed by the chrysopid's mandibles. Nevertheless, all three of these encounters proved fatal to L. futilis. When attacked by ants, larvae that were hatching from eggs responded by spreading their mandibles. Ants were observed car- rying off eggs containing larvae. The reaction of recently emerged caterpillars to the attacks included head swaying, mandibular spreading, and vomiting. These responses seemed to have no deter- rent effect on the ants.
Regurgitation in larval L. futilis appears to be a defensive behavior. The oral effluent of mid- to late-instar larvae is repellent to ants and irritant to cockroaches, properties that could well be indicative of general anti-arthropodan deterrency. The finding that the effluent induced cheliceral cleaning in lycosids, and that larvae had a good chance of surviving after their rejection by the spiders, provides direct evidence for the defensive potential of the vomiting behavior.
In the scratch test with P. americana, the larval regurgitant proved as effective as two well-known chemical defensive agents, one a common component of arthropodan secretions (2-methyl- 1,4-benzoquinone), the other a defensive metabolite of plants (pulegone). The time-delay to scratching with these two com- pounds had previously been determined (Eisner et al. 1990) and is comparable to what we found it to be with the larval regurgitant. The fact that the effluent proved repellent on near-contact in the tests with the ant P. longicornis indicates that one or more compo- nents of the fluid can effect their deterrency as vapors. No defini- tive explanation can be given for the decline in repellency following aerial exposure. Such decline could be due to evapora- tive loss of active principles, or to any number of reactive transfor- mations (oxidations?) triggered in the fluid upon emission. A question of obvious interest concerns the origin of the active principles in the oral effluent. Whether the chemicals are produced by the larva itself, are derived from the diet, or are of dual



================================================================================

218 Psyche [vo~. 100
endogenous and exogenous origin, remains unknown. We cannot even rule out the possibility that activity is attributable to a single compound rather than a mixture.
Vomiting on the part of eclosing larvae proved ineffective vis a vis the predators that we selected for testing (the chrysopid C. cubana and the ants P. longicornis and S. invicta). We are reluc- tant to conclude from this that eclosing larvae are generally vulner- able, and that they would have fared similarly in tests with other predators. Mandibular snapping and oral emission could well prove effective in predation contexts that remain to be examined. Conversely, it is possible that the effluent does not become active until the caterpillar has commenced feeding on its host plant. We found L. futilis to have the habit, shared with many other lepidopteran larvae, of ingesting their egg shells after hatching (Fig. 4). Whether by doing so they acquire not only nutrients, but also chemicals that contribute to the deterrency of the regurgitant, remains unknown. The question may be pertinent to other species as well. Does shell ingestion provide newly eclosed insects with the option of reusing defensive chemicals bestowed upon the eggs by the mother?
Dissection of L. futilis larvae revealed no special morphological refinements of the foregut, such as diverticula. We noted no diver- ticula, such as are known for storage of regurgitatable oils by euca- lyptus-feeding larvae of the lepidopteran genus Myrascia (Oecophoridae) (Common & Bellas 1977).
A number of larvae that we collected in the field and that subse- quently pupated in the laboratory succumbed to a tachinid fly para- sitoid (adults are deposited in the Cornell Insect Collection, voucher lot no. 1219). L. futilis that we raised indoors from field- collected eggs remained unparasitized. In another noctuid moth, Agrotis ipsilon, larval regurgitant functions as a kairomone that elicits larviposition by the tachinid Bonnetia comta (Clement et al. 1986). One wonders whether in L. futilis the oral effluent has a similar signal function.
Finally, our observations of larval cannibalism corroborate the earlier field and laboratory studies of Semlitsch and West (1988), who noted larval cannibalism during an outbreak of L. futilis in a maritime forest in South Carolina. Unlike our observation of strictly nocturnal activity, they found larvae active during both day and night.




================================================================================

19931 Smedley, Ehrhardt & Eisner 219
Figure 4. (Top) Chrysopid larva (C. cubana) feeding on L. fitilis egg. (Bottom) Newly emerged larvae immediately after consuming egg shells.



================================================================================

Psyche
[Vol. 100
We express our gratitude to the staff of the Archbold Station for their hospitality and support, to Maria Eisner for help with photog- raphy, and to Dr. Katherine Tauber for identifying the chrysopid. This research was conducted as part of a field course, "Explo- ration, Discovery, and Follow-up," given by Cornell University at the Archbold Biological Station, under auspices of the Graduate School. Support was also provided by a NIMH pre-doctoral traineeship (to S. R. S) and by NIH grant AI-02908. BLUM, M. S.
1981. Chemical Defenses of Arthropods, Academic Press, New York. 562 pp. BOWERS, M. D.
1993. Aposematic caterpillars: Life-styles of the warningly colored and unpalatable, pp. 331-371. In N. E. Stamp and T. M. Casey [eds.], Caterpillars: Ecological and Evolutionary Constraints on Foraging, Chapman and Hall, New York. 587 pp.
BROWER, L. P.
1984. Chemical defense in butterflies, pp. 109-133. In R. I. Vane-Wright and P. R. Ackery [eds.], The Biology of Butterflies, Symposium of the Royal Entomological Society of London, Number 11, Academic Press, London. 429 pp.
CLEMENT, S. L., W. L. RUBINK, AND D. A. MCCARTNEY 1986. Larviposition response of Bonnetia comta [Dipt.: Tachinidae] to a kairomone of Agrotis ipsilon [Lep.: Noctuidae]. Entomophaga 31: 277-284.
COMMON, I. F. B. AND T. E. BELLAS
1977. Regurgitation of host-plant oil from a foregut diverticulum in the lar- vae of Myrascia megalocentra and M. bracteatella (Lepidoptera: Oecophoridae). J. Aust. ent. SOC. 16: 144-147. EISNER, T.
1970. Chemical defense against predation in arthropods, pp. 157-217. In E. Sondheimer and J. B. Simeone [eds.], Chemical Ecology, Academic Press, New York. 336 pp.
1980. Chemistry, defense, and survival: case studies and selected topics, pp. 847-878. In M. Locke and D. S. Smith [eds.], Insect Biology in the Future, "VBW 80," Academic Press, New York. 977 pp. EISNER, T., J. S. JOHNESSEE, J. CARREL, L. B. HENDRY, AND J. MEINWALD 1974. Defensive use by an insect of a plant resin. Science 184: 996-999. EISNER, T., A. F. KLUGE, J. C. CARREL, AND J. MEINWALD 1972. Defense mechanisms of arthropods. XXXIV. Formic acid and acyclic ketones in the spray of a caterpillar. Ann. Entomol. Soc. Amer. 65: 765-766.
EISNER, T., I. KRISTON, AND D. J. ANESHANSLEY 1976. Defensive behavior of a termite (Nasutitermes exitious). Behav. Ecol. Sociobiol. 1: 83-1 25.




================================================================================

19931 Smedley, Ehrhardt & Eisner 22 1
EISNER, T., K. D. MCCORMICK, M. SAKAINO, M. EISNER, S. R. SMEDLEY, D. J. ANESHANSLEY, M. DEYRUP, R. L. MYERS, AND J. MEINWALD 1990. Chemical defense of a rare mint plant. Chemoecology 1: 30-37. EISNER, T., S. NOWICKI, M. GOETZ, AND J. MEINWALD 1980. Red cochineal dye (carminic acid): its role in nature. Science 208: 1039-1042.
MORROW, P. A., T. E. BELLAS, AND T. EISNER 1976. Eucalyptus oils in the defensive oral discharge of Australian sawfly lar- vae (Hymenoptera: Pergidae). Oecologia 24: 193-206. PETERSON, S. C., N. D. JOHNSON, AND J. L. LEGUYADER 1987. Defensive regurgitation of allelochemicals derived from host cyano- genesis by eastern tent caterpillars. Ecology 68: 1268-1272. RICE, W. R.
1989. Analyzing tables of statistical tests. Evolution 43: 223-225. SEMLITSCH, R. D. AND C. A. WEST
1988. Size-dependent cannibalism in noctuid caterpillars. Oecologia 77: 286-288.
SOKAL, R. R. AND F. J. ROHLF
1981. Biometry. 2nd ed. W. H. Freeman and Co., San Francisco. 859 pp. SMEDLEY, S. R., K. D. MCCORMICK, AND T. EISNER 1990. Interaction of Pyrausta panopealis (Pyralidae) with a newly-reported host, the endangered mint Dicerandra frutescens (Labiatae). J. Lepid. SOC. 44: 156-162.
WHITMAN, D. W., M. S. BLUM, AND D. W. ALSOP 1990. Allomones: Chemicals for defense, pp. 289-351. In D. L. Evans and J. 0. Schmidt [eds.], Insect Defenses: Adaptive Mechanisms and Strategies of Prey and Predators. State University of New York Press, Albany, NY. 482 pp.




================================================================================




================================================================================