Terrence D. Fitzgerald

Collective behavior in social caterpillars

The repertoires of social caterpillars are drawn from four categories of behavior centered on shelter building, thermoregulation, colony defense, and trail-based communication. This chapter provides an overview of these collective patterns of behavior and assesses the potential role of individual-level behavior in their overt expression. While few of the 300 or more species of caterpillars that form sib-aggregations have been studied in any detail, our review of the literature indicates that the most promising fronts for the investigation of emergent phenomena in social caterpillars lie in the areas of collective shelter building and trail-marking behavior. Of particular interest is the resemblance of trail-based chemical recruitment communication and collective flexibility in recruitment in caterpillars to similar phenomena in the Hymenoptera. The decision rules underlying recruitment patterns have been explored in ants, and are likely to prove generalizable to the more sophisticated of the recruitment systems found among the lasiocampid caterpillars, but a set of mechanisms dependent on such species-specific factors as resource patchiness, mode of recruitment, and physical properties of silk are likely to uniquely influence collective foraging patterns and shelter-building behavior in social caterpillars.

Trail Marking by the Larva of the Madrone Butterfly Eucheira socialis and the Role of the Trail Pheromone in Communal Foraging Behavior

The larva of the Madrone butterfly Eucheira socialis (Lepidoptera: Pieridae) secretes a trail pheromone from the ventral surface of the posterior tip of its abdomen. Caterpillars mark trails by bringing the secretory site into brief contact with the substrate during a locomotive cycle. Foragers mark most heavily when they move onto new branches and little, if at all, when they move over established trails or when they return to the communal shelter after feeding. The caterpillars make careful comparisons of alternative pathways at choice points and select newer and stronger trails over older and weaker trails. Differential marking of new and established trails during nightly forays, coupled with sensory discrimination of trails by strength and age, leads colonies to abandon old trails in favor of new trails. When applied at a rate as low as 2.5 × 10−10g/mm, caterpillars followed synthetic trails prepared from 5β-cholestane-3-one, a trail pheromone previously reported from the tent caterpillars (Malacosoma spp.). Although both Eucheira and Malacosoma mark with the tip of the abdomen and have near-identical sensitivites to 5β-cholestane-3-one, our study shows that Eucheira employs a relatively unsophisticated system of trail-based communication and does not recruit to food. The trail-based communication system of Eucheira appears to represent an early stage in the evolution of cooperative foraging that is derived from, and motivationally linked to, conflict behavior.

The role of tactile and chemical stimuli in the formation and maintenance of the processions of the social caterpillar Hylesia lineata (Lepidoptera: Saturniidae)

Colonies of the social caterpillar Hylesia lineata (Lepidoptera: Satumiidae) form long, single-file, head-to-tail processions as they move between their shelters and distant feeding sites. Although investigations of other processionary species have implicated a silk trail in the processionary process, silk plays little or no role in initiating or maintaining processions in H. lineata. Studies we report here implicate both tactile stimuli and a trail pheromone in the establishment and maintenance of processions. Processionaries elicit locomotion in the individual preceding them in line by brushing their heads against prominent sulci that project from the tips of their abdomens. Caterpillars mark their pathways with a pheromone deposited by brushing the ventral surfaces of their last abdominal segments against the substrate. The persistent pheromone is soluble in hexanes and appears to be secreted from glandular setae found on the proximal regions of the anal prolegs and the venter. In Y-choice tests, caterpillars selected newer trails over older trails and stronger trails over weaker trails. They did not distinguish between trials deposited by newly fed caterpillars and those deposited by starved caterpillars. Despite the unidirectional nature of processions, there is no indication that caterpillars can determine from the trail alone the direction in which the procession advanced. The significance of these findings to the foraging ecology of the caterpillars is discussed.

COMMUNAL FORAGING BEHAVIOR AND RECRUITMENT COMMUNICATION IN Gloveria sp.

The caterpillars of Gloveria sp. mark trails with a pheromone they deposit by dragging the ventral surface of the tip of the abdomen along branch pathways as they move between their communal nest and distance feeding sites. The threshold sensitivity of the caterpillar for an extract prepared from the secretory site was approximately 0.5 × 10−3 caterpillar equivalents/cm of trail. Bioassays show that Gloveria follows neither authentic trails of Malacosoma americanum nor artificial trails prepared from 5β-cholestane-3-one, a chemical previously reported to elicit trail following from other social caterpillars. Although our observations show that fed caterpillars mark heavily as they return to their nest, we found no evidence that individual caterpillars are able to recruit hungry nestmates to new food finds. In this species, recruitment to food occurs only after many caterpillars have reinforced a trail to a newly discovered food source. In contrast, hungry caterpillars of the confamilial species M. americanum, tested under identical conditions, responded strongly to the postprandial trails of individual caterpillars and rapidly abandoned depleted sites in favor of new food finds. We postulate that the difference in the efficiency with which these two species recruit nestmates to food evolved in response to differences in the spatial distribution of their food supplies.

Leaf shelter-building caterpillars harness forces generated by axial retraction of stretched and wetted silk

Leaf shelter-building caterpillars generate most of the force required to pull leaf surfaces together by stretching silk strands while spinning. Axially retractive forces produced by columns of stretched strands enabled caterpillars in our study to generate forces as great as 0.3 Newtons (i.e., a 30-g force). We found that caterpillar silk also contracts instantly when wetted, producing an additional, though smaller, axially retractive force. Contraction ratios (final length/ original length) of the wetted silk of 19 species ranged from 0.21 to 0.93 and were smallest among species that use their silk to make leaf shelters. Our study, the first to identify the specific sources of the energy harnessed by caterpillars to tie, roll, or fold leaves, indicates that silk properties and caterpillar behavior have coevolved to facilitate the leaf shelter-building process.

Identification of trail pheromone of larva of eastern tent caterpillarMalacosoma americanum (Lepidoptera: Lasiocampidae).

Previous studies have shown that larvae of the eastern tent caterpillar (Malacosoma americanum F.) mark trails, leading from their tent to feeding sites on host trees, with a pheromone secreted from the posterior tip of the abdominal sternum. 5β-Cholestane-3,24-dione (1) has been identified as an active component of the trail. The larvae have a threshold sensitivity to the pheromone of 10−11 g/mm of trail. Several related compounds elicit the trail-following response. Two other species of tent caterpillars also responded positively to the pheromone in preliminary laboratory tests.

Winter foraging patterns and voluntary hypothermia in the social caterpillar Eucheira socialis

1. Analysis of 28 years of weather data for the Sierra Madre Occidentals of Mexico showed that while flight, mating, and oviposition of the social caterpillar Eucheira socialis (Lepidoptera: Pieridae) occurred in the warmest and wettest months, much of the caterpillar’s feeding and growth occurred in the winter when nocturnal temperatures often fell below 0 °C.

Depletion of Host-Derived Cyanide in the Gut of the Eastern Tent Caterpillar, Malacosoma americanum

Using a colorimetric procedure, we assessed the HCN-p of black cherry leaves (Prunus serotina) ingested by the eastern tent caterpillar, Malacosoma americanum, and the cyanide content of the bolus as it passed thorough the caterpillars digestive tract and into the detritus pool. The mean HCN-p of leaves in our study area was 1902 ± 174 (SE) ppm. Young leaves found at the tips of growing branches, which the caterpillars preferred, had a significantly higher HCN-p (3032 ± 258 ppm) than older leaves found at the middle (1542 ± 243 ppm) or base of the shoot (1131 ± 159 ppm). Following a bout of overnight feeding on young leaves, the cyanide content of the foregut and midgut boluses of early sixth-instar caterpillars averaged 631 ± 161 ppm, and 14 ± 3 ppm, respectively, indicating that host-derived cyanide is rapidly depleted as the bolus transits the gut. Some cyanide, however, remains. In three studies, the mean cyanide content of fresh fecal pellets ranged from approximately 20 to 38 ppm, while the dried, compacted pellets ranged from 63 to 85 ppm. Food in the foreguts of mature caterpillars dispersing over the ground in search of pupation sites had 417 ± 99 ppm cyanide. The potential impact of this egested and caterpillar-transported cyanide on the consumer and detritivore communities is discussed.

A Hidden past: the Hypermetamorphic Development of Marmara arbutiella (Lepidoptera: Gracillariidae)

Abstract Development in Marmara arbutiella Busck is hypermetamorphic, with 3 behaviorally and morphologically distinct larval forms. There are 6–8 sap-feeding and 2 nonfeeding, structurally differentiated instars. The early instars are legless, dorsoventrally compressed, prognathous sap feeders. The 1st of the nonfeeding instars, which we call the transition instar, never issues from the cuticle of the previous (feeding) instar; instead there is apolysis without ecdysis. It is characterized by reduction—the cuticle is transparent, the feeding structures are nonfunctional, and the legs are vestigial. Moreover, the stadium is ephemeral, lasting no ≤24 h. The 2nd nonfeeding stage is a fully legged and ambulatory instar, with rudimentary feeding structures and a functional spinneret. Upon issuing from the mine, this instar spins a cocoon that is elaborately decorated with clusters of 31–114 pearly bubbles that are extruded from its anus and then individually attached to the exterior of the cocoon. We contrast the unique development of Marmara with other gracillariids and interpret the transition instar as an ontogenetic bridge, a stage of risk and metabolic cost without function.

Trail marking and processionary behavior of the larvae of the weevil Phelypera distigma (Coleoptera: Curculionidae)

We present here the results of an investigation of the behavioral bases of the first documented instance of trail marking and processionary behavior in a beetle. The larvae of Phelypera distigma (Coleoptera: Curculionidae) forage communally, moving over the host plant in head-to-tail processions. Our study shows that the larvae secrete a pheromone from the ventral surface of the posterior abdomen that both elicits and guides the collective locomotion of the cohort. The pheromone is soluble in acetone and other nonpolar solvents and is relatively short-lived, eliciting trail following for less than 4 h after its deposition. When in processionary formations, larvae stimulate locomotion in others by rapidly bobbing their heads against sets of setae that occur on the lateral flanks of the posterior tips of the abdomens of precedent individuals. Larvae are also strongly attracted to tactile or chemotactile stimuli found at the tip of the abdomen of other larvae and their response to lures made of eviscerated abdomens show that such stimuli take precedence over the trail pheromone in eliciting and orienting locomotion. The cycloalexic formations adopted by resting larvae maximize the amount of body contact possible in a two-dimensional aggregate and allow tactile signals to rapidly radiate through the groups, alerting all members of a cohort to the onset of bouts of activity.