Foraging, mating, and thermoregulatory behavior of Cyrtopogon willistoni Curran (Diptera: Asilidae).
Psyche 102:121-130, 1995.
Searchable PDF, 796K
Stable URL: http://psyche.entclub.org/102/102-121.html
At Hindawi: http://dx.doi.org/10.1155/1995/53430
The following unprocessed text is extracted from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.
FORAGING, MATING, AND THERMOREGULATORY
BEHAVIOR OF CYRTOPOGON WZLLISTONI CURRAN (DIPTERA: ASILIDAE)
Department of Entomology
Montana State University
Bozeman, MT 59717
The robber fly Cyrtopogon willistoni Curran was studied in SW Montana, where it was an opportunistic predator of relatively small insects from 25 families in 7 orders. The most common prey were Diptera (44%) and Homoptera (21 %), with Cicadellidae, Bibion- idae, and Formicidae comprising 44% of the prey. The elaborate courtship behavior of males included audible airborne visual dis- plays that made use of silvery-white combs of hairs on the males' foretarsi. While perching, the flies exhibited both lateral and dorsal basking postures, and were apparently capable of strong flight only when direct sunlight was available. I compare the foraging and courtship behaviors of C. willistoni to those of other Cyrtopogon, and their thermal responses to those of other robber flies in the same habitat.
The robber flies of genus Cyrtopogon occur throughout North America (Wilcox and Martin, 1936), where they appear to be gen- eralist predators (Bullington and Lavigne, 1992; Lavigne, 1970). Males of this genus in particular, and of the subfamily Dasypogo- ninae in general, exhibit elaborate courtship displays not evident in other subfamilies of Asilidae (Alcock, 1977). Here, I describe the foraging, courtship, and thermoregulatory behaviors of Cyrtopogon willistoni Curran, a species previously subject to a brief report by Lavigne and Holland (1969). Cyrtopogon willistoni is a relatively Manuscript received 25 February 1995.
122 Psyche [vol. 102
stout asilid, 1 1 - 15 mm in length, with a shiny black body surface. The four distal segments of the males' foretarsi have a dense comb of silvery-white hairs that appear to play a role in courtship. Although most Cyrtopogon live within or near coniferous forests, the C. willistoni studied in Montana inhabited mesic slopes of gul- lies on grassland.
MATERIALS AND METHODS
I made most of my observations 15 km south of Three Forks, Gallatin Co., Montana (45'45'N, 1 1 1 O35'W; 1300 m altitude) dur- ing May and June of 1989 through 1994. However, in June 1990, I also obtained several prey records at the Red Bluffs Research Ranch, 2.5 km ENE of Morris, Madison Co., Montana. Predator head widths were determined to the nearest 0.1 mm using a micrometer accurate to 0.05 mm. Prey sizes are reported as the product of body length (not including antennae or ovipositors) and maximum width (not including appendages). In most cases, a fly could be approached and observed within several meters without apparently disturbing it.
When a perched fly was located, I used methods described in O'Neill, Kemp, and Johnson (1990) to record: 1) soil surface tem- perature (To) of the nearest unshaded patch of bare soil > 5 cm in diameter, 2) air temperature at the exact location of the perched fly's thorax (TTu) (TTH is an important variable because the air temperature will affect the rate of convective heat loss, and body temperature is a function of the relative magnitudes of radiative heat gain and convective heat loss), 3) perch height, 4) the orienta- tion of the longitudinal axis relative to the ground, as either approximately parallel to the soil surface (horizontal) or at an angle of 30-90' to the surface (oblique/vertical), and 5) the orien- tation (relative to the position of the sun) of the fly's longitudinal axis in the horizontal plane. For the latter, an angle of 0' indicates that a fly was facing the position on the horizon above which the sun was located, while at 180' a fly oriented directly away from the sun. A fly with an approximately 180' orientation relative to the sun and with an oblique or vertical orientation relative to the soil surface was considered to be in a "dorsal basking" orientation. A fly perched at an angle of 90' or 270' on a horizontal surface (i.e., oriented perpendicular to the direction of the sun) was considered
to be in a "lateral basking" orientation. When data sets were bimodal, as in the case of angles clustered around either 90' or 270å¡ the data were divided into two subsets and analyzed sepa- rately. Data on angles of orientation were analyzed as described in Batschelet (1981) and O'Neill et al. (1990). To examine the orien- tation of longitudinal axis in the horizontal plane, I used Rayleigh's test and the V-test, for which significance indicates that the null hypothesis of uniform or random distribution of perch direction is rejected; for the V-test, significance further indicates that the observed values are clustered around some hypothesized angle (Batschelet, 198 1).
Temperatures were measured with copperlconstantan thermo- couples and a Cole-Parmer ~i~i-~ense~ thermometer. Air and soil surface temperatures were measured with the tip of the probe tem- porarily shaded from direct solar radiation. I used To as a general measure of environmental temperature, because it is relatively sta- ble over short periods of time and is correlated with the intensity of solar radiation on clear days (O'Neill et al., 1990). RESULTS AND DISCUSSION
Habitat and Seasonal Activity
At the site south of Three Forks, C. willistoni was primarily restricted to mesic north-facing slopes of a 1.3 km long gully that ran from SW to NE down to the Madison River. The robber flies were active primarily in expanses of herbaceous vegetation in which lupine (Lupinus sp.) predominated, but could also be found on snowberry shrubs (Symphoricarpos sp.). I did not find Cyrto- pogon willistoni on the south-facing slopes, on level portions of the drier areas on grasses and Yucca (Yucca glauca Nutt.), nor on the north rim and narrower segments of the gully where there were small trees (primarily Prunus sp.). Cyrtopogon willistoni shared the mesic habitat with a smaller robber fly, Dicolonus sparsipilo- sum Back, which appeared to prefer the lower levels of the vegeta- tion. Three larger species, Machimus occidentalis Hine, Efferia staminea (Williston), and Stenopogon inquinatus Loew were active in both mesic and xeric areas during the latter half of the seasonal activity of C. willistoni, which typically lasted from late May to late June. At the Red Bluff site, C. willistoni was observed on
124 Psyche [VOL 102
herbaceous vegetation (including Geranium sp.) in a narrow mesic zone bordering a small seasonal stream.
Adults Cyrtopogon willistoni are ambush predators that leave their perches 4-70 cm high (mean = 24.3, SE = 1.4, N = 103) on vegetation to pursue airborne prey. The flies sometimes took cir- cuitous routes while pursuing prey for distances of up to 50 cm away from the perch and as high as 20 cm above it. The single prey not taken in mid-air was a cicadellid leafhopper that flew into a perched male who immediately subdued it. After most prey cap- tures, the flies immediately returned to a perch, sometimes the same one from which they took off. Occasionally, however, a fly first hovered in place while manipulating the prey with its legs. Two of these potential prey were released (or escaped). The 120 prey of C. willistoni included insects from 7 orders and 25 families: EPHEMEROPTERA Ephemerellidae (5), Baetidae (I), Heptageniidae (1); HOMOPTERA Aphididae (1 ), Cicadellidae (23), Delphacidae (1); HEMIPTERA Miridae (13); COLEOPTERA Curculionidae (1); LEPIDOPTERA Tortricidae (2); DIPTERA Anthomyiidae (4), Asilidae (2, both male D. sparsipilosum), Bibionidae (16), Bombyliidae (3), Chironomidae (I), Dolichopodi- dae (2), Empididae (5), Muscidae (4), Sarcophagidae (4), Sepsidae (7), Syrphidae (4), Tachinidae (2); and HYMENOPTERA Bra- conidae (I), Ichneumonidae (2), Proctotrupidae (I), Formicidae (14).
Prey records varied among years, although this may have been due to small sample sizes. For example, all of the Ephemeroptera were taken in 1989 and 88% of the bibionids were from 1994. Like five other species of Asilidae at the same site (O'Neill, 1994), C. willistoni took the alate ants, 93% of which were males, during mating swarms of Formica subpolita Mayr. Prey were always smaller than the predator and varied widely in size, from minute Sepsidae and Aphididae to the relatively large alate ants. There was a significant correlation between prey size and predator size among female C. willistoni (rs = 0.41, N = 63, P < 0.001), but not among males (rs = 0.17, N = 49, P = 0.20). The foraging mode of the flies I observed differed from what Lavigne and Holland (1969) reported for a population of C. willis- toni along an intermittent stream in Wyoming. There, they pnmar- ily foraged from perches on soil or rocks, and commonly subdued
prey only after both predator and prey fell to the ground following initial contact in mid-air. In a smaller sample, they reported a nar- rower range of prey types (17 insects from 7 families, all Diptera and Hymenoptera). Calliphoridae was the only family on their list that did not occur in the prey of the Montana population. A wide diet breadth, similar to that found for C. willistoni in my study, was reported for C. montanus James (6 orders of prey; Bullington and Lavigne 1992), C. auratus Cole (7 orders, 29 families), and C. glarealis Melander (6 orders, 19 families) (Lavigne, 1970). The three most common families of C. willistoni prey in Montana, Cicadellidae, Bibionidae, and Formicidae, are quite different in size and behavior suggesting that the flies display the flexibility in prey use that is typical of many asilids (O'Neill, 1992a, b). Courtship and Mating
Males commonly flew slowly through the topmost vegetation, frequently pausing and turning towards plants or an insect perched on a leaf. Males initiated courtship, but not prey pursuit, during these flights. Copulation was always preceded by prolonged courtship of a perched female, during which a male alternated between hovering near the female and perching next to her on the same plant. While hovering 5-10 cm above and to the side of a female, the male repeatedly darted -0.5 cm towards her while rapidly waving his legs forwards and vibrating them. During one continuous hover that lasted 5 min and 13 s, a male flashed his front legs over 30 times. When flashing, the pitch of wing buzzing, which is audible to humans from at least 5 m, increased and there was a distinct flash from the light reflecting off a dense comb of silvery-white hairs on the four distal segments of his foretarsi. The time between flashes varied from 6-15 s (mean 9.4 s, SE = 0.4, N = 31 in observations of 8 males). Often (9% of 22 courtships), two males simultaneously courted a single female. Females usually remained still during the males' displays, but some reacted by fly- ing away (sometimes followed by the male) or by flying directly at the male in an apparent attempt to drive him off. One female darted at a male eight times before he terminated courtship. Upon landing next to a female, the male remained motionless or tried to mount her from behind. Females usually rejected these mating attempts, either passively (N = 8) or by striking at the male
126 Psyche [vo~. 102
with her abdomen. When successful at coupling, the male always immediately dismounted and faced away from the female until copulation ended. He maintained the end-to-end position even when the female took flight and towed him behind her. I observed eight matings, including three from the initiation to the end of cop- ulation that had a mean duration of 13.2 min (range = 10.6-1 6.1). Males sometimes courted feeding females (N = lo), and females sometimes captured prey while being courted (N = 3); one male initiated copulation with a female that continued feeding. Males also directed "courtship" to other perched insects, including four sarcophagid flies, a syrphid fly, a moth, and seven conspecific males, all approximately the same size and color of the females. On three occasions, males vigorously "courted" shrivelled, desic- cated flower heads that resembled perched females, because they were black and had dried bracts that protruded laterally. The courtship displays of males of different species of Cyrto- pogon exhibit commonalities, but also a great deal of diversity in form and elaboration. Of the six species studied, only C. willistoni males (in both Montana and Wyoming) include an aerial display in their courtship, although such displays occur in other genera of Dasypogoninae (Alcock, 1977). In the other five species of Cyrto- pogon, the display occurs on a substrate with the male approaching the female from the front in C. auratus and C. glarealis (Lavigne, 1970), from behind in C. vanduzeei Wil. and Mart. (Powell, 1969) and C. montanus (Bullington and Lavigne, 1992), and from the side in C. marginalis Loew (Lavallee, 1970). All but C. vanduzeei and C. montanus are reported to exhibit conspicuous movements of the foretarsi, which are silvery-white or yellow in many species (Wilcox and Martin, 1936). Males of some species touch the female during courtship displays, with C. marginalis and C. van- duzeei exhibiting a prolonged repetitive tactile display. As in C. willistoni, males of C. glarealis and C. montanus sometimes directed "courtship" at non-females. All Cyrtopogon apparently mate in an end-to-end position.
Flies perched where air temperatures (TTH) ranged from 12-30å¡C with the Tru being correlated with Ts (rs = 0.61, N = 103, P < 0.001). When perching at times when the sun was not
obscured and To was 5 25OC, the flies apparently oriented them- selves to maximize the amount of direct solar radiation intercepted. The particular orientation adopted was contingent on the orienta- tion of the surface on which they were perched. On horizontal plant surfaces, the flies oriented so that their longitudinal axes were approximately perpendicular to the direction of the sun (i.e., either 90 or 270å¡ Table 1). When landing on oblique/vertical sur- faces, they oriented at approximately 180' to the direction of the sun, thus maximally exposing their dorsal surfaces; such exposure is enhanced by the fact that the perched flies held their wings in V- shaped array, so that the surface of the abdomen was unshaded. Observed angles of orientation did not differ from angles hypothe- sized to maximize exposure to direct solar radiation. In combined data for all flies perched when To was <: 25OC, the observed orien- tation deviated by <: 15' from expected in 69% of the observations, with only 4% of the flies deviating by > 30å (N = 86). When To was > 25O, the flies did not display a directional bias on either type of surface.
The flies were highly dependent on direct solar radiation in order to undertake foraging and courtship. On 4 June, 1994, when clouds intermittently obscured the sun over a period of 4 h, I counted the number of C. willistoni seen flying, including those flushed as I walked through the area. During 15 sunny intervals totalling 200 min, I saw 105 flies (0.53/min), including 11 with prey and 16 males that were courting females. In contrast, just 3 Table 1. Orientation of the longitudinal axis with respect to the sun (horizontal plane) as a function of soil surface temperature (To). Angle of surface Rayleigh's Observed
on which flies Test Hypothesized mean (å±SD V-test T~* perched Probability angleB angle N probabilityc <25OC horizontal P < 0.001 90' 91.3% 16.6 20 P<O.OOl
P < 0.001
270' 267.3 å 17.8 13 P < 0.001
oblique/vertical P < 0.001 1 80å 179.7 å 19.4 53 P < 0.001 >25OC horizontal P > 0.20 - 8 -
oblique/vertical P > 0.50 - 7 -
*in nearest unshaded area; 'angle hypothesized only if Rayleigh's test significant, indicating that there was a significant bias in the angle of orientation; Significance indicates that angle clustered around the hypothesized angle.
128 Psyche [vo~. 102
flies (0.08lmin) were seen during 14 cloudy intervals totalling 40 min; all three were observed in the first minute after the sun was obscured and none had prey. I may have seen fewer flies during cloudy periods simply because they were inactive. Those flies seen during cloudy intervals appeared incapable of strong flight. When prodded with grass stems, they did not fly off (the usual behavior during sunny intervals), but simply rotated on their perches to face the stem (N = 1) or just dropped off the plant (N = 2). These responses are typical of robber flies at cooler times of day (La- vigne and Holland 1969). Several of these flies flew away after I warmed them in my hand. Some flies apparently also hid during cloudy periods. One feeding female that I was watching just as a cloud obscured the sun immediately moved to the underside of the leaf.
Because C. willistoni perched on vegetation well above the soil surface, they never experienced TTu as high as those apparently preferred by other species of Asilidae in the drier portions of the same habitat. During cooler times of day, other species of robber flies at this site raise their body temperatures by hunting from perches on insolated soil surfaces, where the boundary layer pro- vides a relatively warm, non-windy environment allowing the flies to bask while minimizing convective heat loss (O'Neill and Kemp, 1990, 1992; O'Neill et al., 1990). Late in morning, Machimus occi- dentalis (O'Neill and Kemp, 1992), Stenopogon inquinatus (O'Neill and Kemp, 1990), and Efferia starninea (O'Neill et al., 1990) adopt perch locations where the Tru averages 31 to 34OC. Because Cyrtopogon willistoni always experienced TTH below that range, either they occupy a habitat with suboptimal microclimate or they are adapted to these lower temperatures. The latter explana- tion seems more likely because C. willistoni became active each day before the other species and apparently ceased orienting to the sun when To was greater than 25OC.
O'Neill et al. (1990) demonstrated that orientation by robber flies of the genus Efferia had a significant effect on body tempera- ture and that the temperature attained was highly correlated with the intensity of solar radiation. Like other species (O'Neill and Kemp, 1992; O'Neill et al., 1990), C. willistoni were probably able to raise body temperatures above ambient by orienting to maximize their exposure to insolation. Unlike the other species, however,
which exhibited only lateral basking, C. willistoni also displayed dorsal basking. Perhaps dorsal and lateral basking are equally effective in C. willistoni because the ratio of the area of the lateral profile to dorsal profile of the thorax of C. willistoni is 1 : 1 (mean ratio of lateral to dorsal profile = 1.04, SE = 0.03, N = 6, y2 = 3.37, P = 0.65, 5 d.f.). In contrast, the thoraces of Machimus, Steno- pogon, and Efferia have lateral profiles that are 20-30% greater in area than their dorsal profiles. Thus, for these species, lateral bask- ing is probably more effective than dorsal basking. In addition, lat- eral basking allows the flies of these other species to stay within the boundary layer near the soil surface. While the orientations of perched C. willistoni may also be constrained by a need to maxi- mize visibility of passing insects, the orientations adopted are con- sistent with the hypothesis that they are basking to raise body temperatures. Basking was shown to have a significant effect on the body temperatures of other species of robber flies (O'Neill and Kemp, 1992; O'Neill et al., 1990) and may be enhanced by black coloration in C. willistoni, which perhaps evolved because of its effect on thermoregulation. Furthermore, after perched flies rotated their bodies when potential prey passed, they rotated back to their original basking orientation. Their reaction to clouds obscuring the sun also supports the hypothesis that the flies depend on direct insolation to raise body temperature.
I thank Ruth O'Neill and Catherine Seibert for assistance with the research and for commenting on the manuscript. Patricia Denke also made helpful comments. C. Riley Nelson identified the robber flies. This work was supported by the USDA-ARS, USDA-APHIS- PPQ, and the Montana Agricultural Experiment Station. Contribu- tion #J-3050 from the Montana Agricultural Experiment Station. Alcock, J. 1977. The courtship behavior of Heteropogon stonei (Diptera: Asilidae). J. Kans. Entomol. SOC. 50:238-243.
Batschelet, E. 198 1. Circular Statistics in Biology. Academic Press: New York. Bullington, S.W., and R.J. Lavigne. 1992. Ethology of Cyrtopogon montanus wilcoxi James (Diptera: Asilidae) in Wyoming. Proc. Entomol. Soc. Wash. 94:123-131.
130 Psyche [VOI. 102
Lavigne, R.J. 1970. Courtship and predatory behavior of Cyrtopogon auratus and C. glarealis (Diptera: Asilidae). J. Kans. Entomol. SOC. 43:163-171. Lavigne, R.J., and F.R. Holland. 1969. Comparative behavior of eleven species of Wyoming robber flies (Diptera: Asilidae). Univ. Wyoming Agric. Exp. Sta. Sci. Mono. 18:l-61.
O'Neill, K.M. 1992a. Patch-specific foraging by the robber fly Megaphorus willis- toni. Environ. Entomol. 21:1333-1340
1992b. Temporal and spatial dynamics of predation in a robber fly popu- lation (Diptera: Asilidae, Efferia staminea). Can. J. Zool. 70:1546-1552. 1994. The male mating strategy of the ant Formica subpolita Mayr (Hymenoptera: Formicidae): Swarming, mating, and predation risk. Psyche 101:93-108.
O'Neill, K.M., and W.P. Kemp. 1990. Behavioral responses of the robber fly Steno- pogon inquinatus (Diptera: Asilidae) to variation in the thermal environment. Environ. Entomol. 19:459-464.
1992. Behavioral thermoregulation in two species of robber flies (Diptera: Asilidae, Machimus) occupying different grassland microhabitats. J. Therm. Biol. 17:323-33 1.
O'Neill, K.M.; W.P. Kemp, and K.A. Johnson. 1990. Behavioural thermoregulation in three species of robber flies (Diptera, Asilidae: Efferia). Anim. Behav. 39~181-191.
Powell, J.A. 1969. Prey hunting and courtship behavior of Cyrtopogon vanduzeei Wil. and Mart. Pan-Pac. Entomol. 45:67.
Wilcox, J.H., and C.H. Martin. 1936. A review of the genus Cyrtopogon Loew in North America (Diptera-Asilidae). Entomologica Americana 16: 1-95.
Volume 102 table of contents