The Behavior of a Stinkbug, Euschistus Conspersus Uhler (Hemiptera: Pentatomidae)

The small, bro.wn-backed, pale-bellied stinkbug Euschistus conspersus Uhler is a common insect along the Pacific coast from California to British Columbia (Essig, 1929). In some areas it is a pest feeding on pears and other commercially important fruits; the natural history of several of these, populations has been reported by Borden, Madsen and Reran (I952). In addition aspects, of the developmental cycle of this species have been studied in the laboratory (Hunter and Leigh, 1965). This paper presents new information on the basking, feeding, dispersal and courtship behavior of one. popula.tion of E. conspersus in Seattle, Washington. From the. middle of May to. the mid-October 1971 I watched the stinkbugs which inhabited a large vacant lot near the University of Washington. This area bordered Lake. Washington and was covered by marsh and field grasses, willows, alders, cattails, and a dense stand of blackberry bushes. The results which follow have been derived from field notes and photographs made throughout the observation period.


Basking Behavior
My attention was drawn to the bugs initially because so many could be found standing conspicuously on the upper surfaces of blackberry leaves i the full sun, particularly in the. morning. The:se insects do not feed on leaves.
As Fig. illustrates with a single day's census, the .stinkbugs were rarely found on leaves until t'he sun had appeared in the morning. Then a very substantial proportion of the population of bugs moved quickly to. places where they were fully exposed to the sun's rays (Fig. 2). Dark green blackberry leaves provide a flat heat-absorptive surface for basking. Bugs. tended to flatten themselves against the leaf on which they stood. They often chose leaves which were' slanted upwards so that the bug's brown back was oriented more or less. at right angles to the sun. And, particularly in the early morning and late afternoon, when the sun was low, some bugs tilted their bodies 1971] dlcock Behavior o/ Stinkbug :7 to one side to achieve the same effect (Fig. 3). This behavior presumably increases the rapidity with which the bug is warmed by directly exposing the. maximum surface area of the animal to the sun.
By midday basking was on the' wane ( Fig. :) although in those areas where the afternoon sun reached blackberry leaves some bugs could be seen basking in the late afternoon. (The figures are based o.n data collected from an area on the eastern side. of the lot which was wellshaded by I6.OO.) Feeding Behavior E. conspersus feeds on fruits and is in fact extremely catholic in its tastes (Borden,Madsen and Retan,19..52). Many bugs were watched as they fed on grasses and on the fruit and stems, of the blackberry Rubus laciniatum. Although I have no quantitative, evidence to confirm this point, it .seemed clear that the. bugs preferred, both as nymphs and adults., unripe pinkish-green blackberries to fully ripe fruit. Bugs. also fed on an ornamental, Pyrecantha coccinea, and on the pods and stems of a sweet pea, Lathyrus latifolius. But by far the most remarka.ble choice of a foodplant was the bracken fern, Pteridium aquilinum. This plant occurred in .scattered patches, along the edge of the. lot amidst the blackberry bushes. Both adults and nymphs were commonly found on the fern; adults were seen many times with their rostrums inserted into, the stem of the plant.
Groups of adults, congregated ,on a few ferns in the early summer and second generation bugs occurred in large number.s on almost alI living ferns, in the late summer. Those few ferns which had hosted bugs in the early s.ummer turned yellow and died while, the remainder on which few or no. stinkbugs had been seen remained quite, healthy. The stems of dying ferns were covered with small reddish marks presumed to be points of insertion of the bugs' rostrums. There was a complete die-o.ff of all ferns in mid-September long before the. first frosts. It seems likely that the stinkbugs were' capable of killing their fern hosts. On the. other hand, blackberry stems which supported large numbers of bugs, for many days did not die back or appear injured in any way.
Feeding activity wa.s most pronounced during the afternoon with many bugs moving from leaves to blackberry stems during this time ( Fig. ) E. cons#ersus readily discharged its glands when handled roughly, or pecked by a bird, and several times I detected the odor coming from bugs trapped in a spider's web. This, action may protect some bugs from some predators, particularly birds, which are not strongly motivated to feed. However, the insect is not solely dependent upon its secretion as it engages in a variety of defensive maneuvers remaining immobile much of the time, scuttling under leaves or dropping from them to the ground when approached or to.uched, kicking at ants, buzz-flying and wing-whirring. The defensive function of the latter two, behaviors is purely speculative. However, bugs flying some distance' definitely produced a fairly loud and, to this observer at least, a very bee-like buzz which might deter some aerial predators from attacking. Wing-whirring co.nsisted of lifting and rapidly vibrating the wings, producing a loud buzz. Five of a group of about fifty bugs which I picked up, handled, and returned to a leaf remained on the. leaf and wing-whirred. Similar behavior has been reported for some species of stinkbugs occurring on coco.a plants (Callan,I944) and might serve to startle o.r to warn a predator not to attack.

Reproductive Behavior
Only one generation of stinkbugs mated in Seattle although in California some populations have two reproductively active, generations in a single summer (Borden, Madsen and Retan, 1952). The adults which overwintered in leaf litter beneath the blackberries had emerged in large numbers by the time observations were begun (mid-May). Many adults and mated pairs were seen in the lot through late June but by the end of the first week in July very few adults could be found. By late July the seco.nd generation bugs were appearing in abundance with many feeding on the blackberries which were just beginning to ripen at that time.
Mated pairs were rarely seen in the morning, never at midday, and often rom 5.oo to dusk (21.oo). I believe that the initiatio.n of courtship and copulation occurred almost exclusively in the late afternoon and early evening. The mated pairs seen in the morning probably had coupled the previous evening. Hunter and Leigh Psyche [December (965) report that in the lab bugs remain in copulo from 3-35 hours. A total of six successful courtships was observed, largely in late May, with copulation occurring at I5.35, I8.39, 8.46, 9.27, x9.5o, and 20.05. The description of courtship behavior which follows is based primarily on observations of one aggregation of between o-2o bugs regularly present on one bracken fern in late May and early June. Courting males were active despite the cool evenings; the females present were pressed flat against the stem o.r leaves of the fern. Males (I) palpated the upper surfaces of females with their antennae, (2) appeared to attempt to raise, the female's abdomen by placing their heads under the side of the female and lifting upwards, and (3) turned and pressed their aedaegus, which might be completely extruded and partially inverted, against the side, belly, or tip o.f the abdomen of the female (o.r occasionally against the stem of the fern or even against another male courting the same female). These behavior patterns occurred generally in the order presented but since females were often unresponsive might be repeated over and .over again.
Sometimes in response to a courting male a female would lift her abdomen slowly upward away from the surface on which it had been resting. The male would continue to court, particularly with activities (2) and (3), and often (4) palpated the undersides of the female's abdomen while standing directly behind her. This might induce the would-be mate to raise her abdomen still higher until her body formed an approximately 30 angle with the stem to. which she clung. A.t this point the male turned away and with completely extruded and inverted aedaegus backed toward the female all the while acing directly away from her. The male's abdomen was also raised and upon touching the female he. pressed his aedaegus about the tip of the female's abdomen until it entered the female genital opening. After a series of small movements copulation was. firmly achieved. The bugs might then move a short distance before settling down to. remain in copulo for many hours. Males were extremely persistent courters often attending to a single female for 5-3o minutes before copulation was accomplished or before the male left in search of a more receptive female. One male vas watched for almost one hour as it courted without success.
However, on return to the plant 45 minutes, later a mated pair rested on the branch where the persistent male had been active.
Unreceptive females often simply remained flattened against a stem and did not move. Rejection o.f a male. could be more active how-1971]  Given the considerable amount of sexual activity which occurred in these first generation groups I initially felt that their function was exclusively sexual in nature. However, I later discovered that aggregations were also formed by nymphs and second-generation nonbreeding adults. Groups of 5-6 late instar nymphs could be found basking on the same. leaf together, some even with their heads under the side of a companion, an action reminiscent of the abdomen lifting behavior of courting males. As fall neared large numbers of bugs (up to. 6) occurred on a single fern. Smaller contact groups of 5-o individuals were. common (Fig. 4) and one large tightly clumped duster of between 5 and 35 bugs, was found on exactly the same portion of a blackberry stem .on various dates from 28 September to 7 October. At the same time not all individuals, showed a tendency to. clump together in semi-permanent groups. On the contrary many E. conspersus were highly active and could be seen walking substantial distances along blackberry stems from one. plant to another. On warm days in the afternoon bugs were often seen flying five to. ten meters. Of 3o' bugs marked on 2 May, I could find only 2 the next day. The apparent mobility and dispersal of many stinkbugs, contrast sharply with the seeming stability of some groups.. Although there have been ew reports of basking by pentatomids (however see Thomas, 1954) it nevertheless may be practiced quite commonly by them. Chlorochoa sayi, a less abundant stinkbug than E. conspersm in the study plot, also. basked on blackberry leaves.. This behavior is videspread among insects in general (particularly in deserts and northern latitudes where t'here are great temperature fluctuations). Some species, or example the desert locust (Walof/, 1963), may also orient the body to achieve maximum exposure to the sun. Basking is appropriate or the Northwest where summer nights are usually quite chilly and summer days oten partly cl.oudy and cool.

Feeding Behavior
Although many pentatomids eed on a wide range o plant species, E. conspersus is the first stinkbug known to exploit a ern. Very ew insects attack erns (Brues, 192o;Soo Hoo and Fraenkel, 1964). The only other Heteroptera associated with erns are, some members o the mirid subfamily Bryocorinae (Southwood and Leston, 1959;Woodward, Evans and Easton, 197o).
The stinkbug attacked the stems o adult bracken ferns avidly despite the fact that the plant when ull grown is avoided by almost all herbivores and is reputed to. be toxic to. cattle (Viuenscher, 1939).
Moreover, bracken pinnae o at least some populations contain analogs o ecdysone (Kaplanis et al., 1967) although whether stems contain the substance is not proven. The bug's ability to eed on bracken is all the more remarkable because, unlike other ern herbivores, E. conspersus is not a specialist limited to erns.

Reproductive Behavior
The courtship behavior o very ew stinkbugs has been reported in any detail (but see Kullenberg, 1947;Teyrovsky, 1949;Southw.ood and Hine, 195o;Leston, 1955;Kaumann, 1966). Judging rom these cases courtship among stinkbugs must be highly diverse. For example, the male o Calidea dregii irst aces the. emale and then climbs i:orward onto her head (Kaumann, 1966); the males of Dolycoris laccarum first creep under the abdomen o the emale (Teyrovsky, I949). However, in every previously studied case the male eventually climbs onto the back o the emale and aces in the same direction 1971] dlcock--Behavior o/ Stinkbug 225 as its partner. It is in this. position that insertion o the aedaegus occurs. Usually the male. then turns, dismounts and faces away rom the female. Copulation proceeds in this end-to-end pose. This is a very common sequence of events among the Heteroptera in general (GrassY, I95 Weber,93o). E. conspersus is then the only stinkbug known to omit the male-above position and to effect a copulatory union directly end-to-end.
How might .such behavior have evolved? Weber noted (I93O, p. 3o7) that there were only two basic patterns of Hemipteran copulation--() the male and female facing in the same direction with the male by the side or above the female and (2) the male and female united end-to-end facing away from each other. At the time no. species was known which copulated in position (2) without (irst assuming the side-by-side or male-above position. From this evidence Weber argued that the end-to-end position was an elaboration of the presumably more primitive position (I). Therefore it seems reasonable that the unusual courtship pattern of E. conspersus which omits the male-above component altogether is probably a relatively recent evolutionary invention derived rom the second basic pattern. Representatives of all three copulatory patterns occur in the Orthoptera (and doubtless other insect orders) Alexander's (964) explanation for the evolution of direct end-to-end initiation of copulation in the Orthoptera is essentially the same as that outlined above for the Hemiptera.
Omission of the back-climbing phase of courtship makes some sense or pentatomids given their bulky flattened shape which could make coupling awkward in a male above/female bel.ow position (N. T. Davis, pets. comm.). Indeed in type (2) copulations the. function o.f the male's dismounting and turning away from the female may be to achieve a more stable and easily maintained copulatory position.
Initiation o copulation directly in the. end-to-end position might have had its origin in premature turning behavior. After a number of unsuccessful attempts to copulate while, on the back o.f a female, some males might tend to. dismount and turn despite the act that coupling had not yet taken place. Teyrovsky (949) observed a male D. baccarum attempt a direct end-to-end union with another male it had been unsuccessfully courting. Presumably in the evolution of the sexual behavior o E. conspersus selection qrst favored males with a low threshold for dismounting and turning. Gradually selection has eliminated mounting the female entirely while favoring those males which act to induce the emale to assume that position which makes direct end-to-end mating most easy.
Psyche [December The Second, clumping could be an anti-predator adaptation. A group of toxic, warningly-colored insects may more effectively advertize potential unpalatability than scattered individuals could. In addition should one member of a group be taken, the predator would be likely to avoid the. others. However, the formation of groups by nymphs and adults of a cryptically colored species such as E. conspersus can hardly serve this function. _As previously noted, the bug proved to be edible, although apparently distasteful, to redwinged blackbirds. The formation .of groups must make these, bugs more vulnerable, not less, to. hungry redwings as well as to their wasp predators which can probably learn to return repeatedly to a productive searching area (Evans,966). The fact that the bugs are not terribly nimble also places a premium .on avoiding detection by their enemies. Thus there must be some other advantage for group formation by second generation adults and nymphs which outweighs increased risk of attack by predators. What this advantage is remains uncertain. However, given the death of the ferns it seems probable that the bugs were injecting toxic substances into their foodplants as Hemiptera are known to. do (Nuorteva,I958 Adams and McAllen,I968). The salivary glands of some mirids contain pectinases (Lau:'ema and Nuorteva, I96I) as does as least one lygaeid (Adams and McAllen, I958). The pentatomid D. baccarum possesses salivary proteases and amylases (Nuorteva,I954). Thus it is possible that by feeding in groups individual bugs may extract plant juices more easily than otherwise. In any event, the relative advantages of group formation as opposed to. dispersal must be rather closely balanced given the great variability in the tendency to aggregate exhibited by E. conspersus. UMMARY This paper describes the behavior of members of one population of Euschistus conspersus Uhler, a small brown stinkbug. These bugs characteristically bask for some time after sunrise, often on black-berry leaves. La,ter in the day they feed .on a wide variety of plants including bracken fern, the rst pentatomid reported to, attack a fern. Adults which have overwintered mate in the late afternoon and evening in May and June. Males engage, in several tactile courtship activities which may induce the female to raise her abdomen upward. Males then turn away rom females and back toward them with aedaegus extruded. Copulation is initiated in this end-to-end position. In all other pentatomids and most Heteroptera copulation begins with the male above the female, although later the male often dismounts and faces away from its mate. Aggregations of adults and nymphs are quite common and may have. two. functions-(I) to bring reproductively active individuals together and (2) to acilitate extraction of plant juices.