Temperature regulation in burying beetles (Nicrophorus spp.:Coleoptera: Silphidae): effects of body size, morphology and environmental temperature

The ability of insects to regulate body temperature and warm the body endogenously has been extensively studied in bees, flies and moths (Heinrich, 1993, 1996). Endothermy was first noted in the Coleoptera in 1945 (Krogh and Zeuthen, 1941) but received little attention for 30 years(Bartholomew and Heinrich,1978; Heinrich and Bartholomew, 1979; Chappell,1984; Morgan,1987; Oertli,1989; Oertli and Oertli,1990; Chown and Scholtz,1993; Schultz,1998). Additionally, despite the enormous number of beetle species(nearly half of all known animal species), the thermal biology and thermoregulatory ability of only a handful of taxa have been studied(Oertli and Oertli, 1990), and not all of the findings agree. For example, studies on large, tropical ground beetles (Bartholomew and Casey, 1977a,b)and on large, African dung beetles(Bartholomew and Heinrich,1978) suggest that only beetles of ≥2 g are capable of maintaining a thoracic temperature independent of ambient temperature, whereas smaller beetles (<2 g) are unable to regulate thoracic temperature due to a high surface area to volume ratio. By contrast, studies on beetles that are neither large (≤1 g) nor tropical(Chappell, 1984; Morgan, 1987; Oertli, 1989; Oertli and Oertli, 1990; Chown and Scholtz, 1993)indicate that mass is not always a reliable indicator of thermoregulatory ability and that wing loading and wing beat frequency may also play a critical role (Oertli, 1989; Chown and Scholtz, 1993). There is a need for more data on the thermal biology of flying beetles from different groups (Chown and Scholtz,1993).

The ecology and reproductive behavior of burying beetles (Coleoptera:Silphidae: Nicrophorus) are well studied(Eggert and Müller, 1997; Scott, 1998; Smith and Merrick, 2001);however, little is known about the thermal ecology of this genus. Burying beetles must secure an ephemeral resource (in this case, the carcass of a small vertebrate) to complete their life cycle. Body size and the potential for increased thermal stability in larger species may play important roles in securing these resources from both conspecifics and heterospecifics. The ability to regulate and maintain a thoracic temperature independent of ambient conditions during and after flight may provide competitive benefits, both in terms of carcass defense and speed of burial and the ability to search for carcasses and mates over a wider temperature range, as demonstrated for Plecoma spp. (Morgan,1987). There is also evidence to suggest that there are differential tolerances to environmental temperatures among Nicrophorus species (Wilson et al., 1984; Trumbo,1990; Sikes, 1996; Scott, 1998), which may be related to factors such as body size, pigmentation or evolutionary history and may play a role in species distributions and coexistence. Sympatric species that utilize different thermal and temporal windows for activity have been shown to coexist in the same habitat types without much direct contact(Wilson et al., 1984), and larger species in northern Europe and northeastern North America have been shown to be nocturnal (Scott,1998).

Regulation of body temperature in Nicrophorus could be accomplished by behaviors such as basking, posturing and seeking shade (Casey, 1981, 1992; Heinrich, 1996), by regulating(conserving or losing) heat generated endogenously by flight muscles in the thorax (Kammer, 1981;Heinrich, 1993, 1996), altering metabolic rate or wing beat frequency or a combination of both behavioral and physiological thermoregulation (Coelho,2001). Nicrophorus have been described as “good,persistent fliers that are capable of covering large distances in a short period of time” (Eggert and Müller, 1997), and members of this genus fall within the size class of smaller beetles known to regulate their body temperatures during flight (0.003–1.8 g; Chappell,1984; Morgan,1987; Oertli,1989; Chown and Scholtz,1993). Smaller beetles like Nicrophorus may rely on high wing loading and wing beat frequencies, insulation and morphology to counteract the amount of body heat lost to convection during flight(Chown and Scholtz, 1993), as opposed to passive heat retention as a result of a large body.

We tested whether three Nicrophorus species (N. hybridus, N. guttula and N. investigator) are able to (1) warm up before flight via endogenous heat production and (2) regulate body temperature during and following flight. We further investigated whether wing loading or body mass affected thermoregulatory ability, as well as whether the thermal environment influences the daily activity patterns of burying beetles. We assess thermoregulatory ability using two measures: (1) the relationship between thoracic temperature during flight and ambient temperature and (2) the relationship between thoracic temperature during flight and the effective(operative) temperature of a dead beetle in the flight position (index of thermoregulatory performance; Bishop and Armbruster, 1999). Additionally, we describe daily activity patterns for the three species over two 24-h observation periods and investigate how these patterns could be explained in part by the thermal environment that Nicrophorus experiences and by differential thermoregulatory abilities among species.

Laboratory studies were conducted on the campus of Idaho State University,Pocatello, ID, USA, and the majority of field studies were carried out along the South Fork of Mink Creek drainage, Bannock County, ID, USA, during the spring and summer of 2001, near the South Fork of Mink Creek in the Caribou National Forest. Additional field measurements of field activity and flight temperatures for N. investigator were made at the Rocky Mountain Biological Laboratory along the East River drainage, Gunnison County, CO, USA. In Idaho, two sites were established: South Bench and South Fork. South Bench is located at 2133 m on the lower slopes of Scout Mountain and consists of a mixed Douglas fir/shrub stand including chokecherry (Prunus virginiana), snowberry (Symphoricarpos oreophilus), serviceberry(Amelanchier alnifolia) and snowbrush (Ceanothus velutinus). South Fork is located at 1828 m in a clearing at the edge of a sage-steppe riparian area, dominated by aspen (Populus tremuloides), willows(Salix spp.) and mixed forbs. In Colorado, two established sites were used for beetle collection and observation: Kettle Ponds and Bellview (see Smith and Merrick, 2001 for site descriptions).

Beetles were captured in traps consisting of metal cans (17 cm deep, 15.5 cm in diameter) pierced to allow drainage, half-filled with soil and covered with wire screening formed into a funnel. Each trap was suspended approximately 40 cm above the ground and baited with fresh chicken and water added to the soil (Smith and Merrick,2001).

To determine the thermoregulatory ability of Nicrophorus before and after flight, we conducted a series of flight trials in a 2.44 m×3 m Weatherport® or in a 1.2 m×1.2m screen tent throughout the season and over a range of ambient temperatures. Field-caught beetles were used within 24 h of capture or held in the laboratory and used within two weeks of capture. We tested a total of 137 beetles, including individuals of both sexes, of N. hybridus Hatch and Angell (N=53), N. guttula Motschulsky (N=28) and N. investigatorZetterstedt (Idaho N=7, Colorado N=49). Although N. defodiens were collected, no flight information was obtained from this species.

For each flight trial we randomly selected a beetle from a holding container and immediately measured both its abdominal and thoracic temperatures using a 29-gauge hypodermic temperature probe (Model HYPO-33-1-T-G-60-SMP-M; Omega Engineering Inc., Stamford, CT, USA) and a digital microprocessor thermometer (Omega Model HH23). Abdominal temperatures(T_ab) were taken by inserting the probe between two abdominal sclerites, and thoracic temperatures (T_thx) were taken by inserting the probe into the ventral metathorax. After measuring the temperatures, a beetle was then placed on the middle of a stick (8–10 cm diameter, 0.5 m length) secured at a 45° angle, where it would walk to the end of the stick and make preparations to fly. After take-off and flights of 2–4 s duration, the beetle's thoracic and abdominal temperatures were measured again. We wore gloves while handling the beetles and measurements were made within 2–3 s after landing. Because Nicrophorus have the ability to raise T_thx to a temperature adequate for flight (≈25–30°C) even on cooler days by basking (M. J. Merrick and R. J. Smith, personal observation), every captive flight took place in the absence of direct sunlight, in addition to being sheltered from wind.

The time between the preflight T_thx measurement and actual flight was frequently long, often taking more than 5–10 min. We were therefore unable to determine when the warming occurred or if the temperature immediately prior to flight was different from the post-flight temperatures, although continuous flight measurements indicate that during flight in the shade, thoracic temperatures remain near the initial flight temperature. We assume that a beetle's T_thx immediately before flight and T_thx measured post-flight were similar since each captive flight was short (<4 s.). During the time prior to flight, we did not observe any shivering, wing vibrations or outward signs of muscle contraction. Beetles spent this time cleaning foretarsae and antennae clubs, in addition to extending and retracting wings from below the elytra.

We recorded the ambient temperature using both a bare thermocouple and the effective environmental temperature (T_e) of a null temperature model in the flight position (T_e flight); a dead, dry beetle equipped with thermocouples, wings and elytra raised, and suspended approximately 0.75 m above the ground(Hertz et al., 1993) after each short flight. We also measured hourly the operative temperatures of null temperature models placed in positions that represented other behaviors a beetle could adopt in response to its thermal environment: (1) on bare ground,1–2 cm below the ground surface, and (2) beneath foliage (at the base of short grasses or under fresh leaves resting upon a soil substrate). Operative temperatures, also known as T_e, take into account the effects of radiative and convective heat gain and loss in addition to ambient temperatures on the body temperature of a nonthermoregulating organism(Bakken, 1992; Heinrich, 1993) and more closely approximate the thermal environment that beetles actually experience(Bakken, 1992; Bishop and Armbruster, 1999). Differences between T_e and actual body temperature(T_b) can indicate some form of thermoregulation by the animal (i.e. T_b>T_e indicates endogenous heat production or basking).

To address the point raised by Stone and Willmer(1989a) that some insects actually warm up upon cessation of flight, which could cause overestimation of post-flight T_thx measurements, we measured T_thx continuously for 10–12 min after 24 successful flight trials for N. hybridus (N=4), N. guttula(N=3) and N. investigator (N=17) individuals to determine whether beetles are able to regulate an elevated thoracic temperature after flight. We recorded T_thx in four tethered N. hybridus individuals during continuous (12–39 min)flights and subsequent cooling to determine whether T_thxis regulated during longer flights.

A regression of post-flight thoracic temperature against ambient temperature was used as an index of thermoregulatory precision(Oertli, 1989). Additionally,we calculated a thermoregulatory performance index for each species. Bishop and Armbruster (1999) define this index as the slope of T_thx against T_e, which indicates how a real beetle regulates its thoracic temperature in flight compared with a non-regulating, metabolically inactive one. Differences between T_thx and T_e would then be the result of physiological or behavioral control of body temperature (Casey,1992). A slope equal to or close to one(T_thx=T_a) is considered evidence for thermal conformity, whereas slopes close to or approaching zero are evidence of thermoregulation.

Cooling rates (deg. min^–1) were calculated for beetles representing three species and ranging in mass from 0.11 g to 0.62 g. Beetles were observed cooling after short flights (N=24) and after artificial heating (N=13), where live beetles were heated to a T_thx of 40°C and then allowed to cool. Differences in cooling rates between size classes indicate the amount of heat retention that is a result of body size alone. Based on physiological properties alone,larger beetles should cool more slowly because of a decreased surface area to volume ratio and higher thermal inertia. Three body size classes were assigned for all species based on the median body mass ± 25th and 75th percentiles (small ≤25th percentile; large ≥75th percentile; medium 25th–75th percentile). Cooling rates of live, artificially heated beetles (N=13) were determined by placing the 29-gauge hypodermic thermocouple probe into the lateral metathorax of a beetle that had been cross-pinned to a Styrofoam block. The beetle was then placed inside a 28 cm×18 cm×18 cm Styrofoam box and heated to 40°C with an incandescent lamp, then allowed to cool to within 1–2° of ambient air temperature while its thoracic temperature was recorded every 30 s. Cooling rates of live beetles after flight (N=24) were also measured for 10–12 min following flight (see previous section), and the differences in cooling rates between size classes for post-flight and artificially heated beetles were compared in a one-way analysis of variance(ANOVA). For comparisons, cooling rates for artificially heated beetles were also calculated between the approximate average ambient temperature recorded for outdoor flights and ambient laboratory temperature.

To understand the role that wings, elytra and thoracic pile play in insulation and maintenance of body temperature, we divided 28 beetles(representing three species; N. hybridus, N. investigator and N. guttula) into three groups: (1) no treatment, (2) thoracic pile removed with a scalpel and (3) wings and elytra removed(Chown and Scholtz, 1993). We determined cooling rates for live beetles in the three treatments as above and compared cooling rates using a one-way ANOVA.

We employed methods similar to Chown and Scholtz(1993) and Coelho(2001) to determine whether or not Nicrophorus have the ability to transfer heat produced in the thorax to the abdomen to regulate T_thx. A live beetle was cross-pinned to a Styrofoam block, and one thermocouple probe was inserted into the lateral thorax and another into the abdomen. The head and thorax were heated with an incandescent lamp while the abdomen was shielded with Styrofoam wrapped with aluminum foil. Measurements of T_thx and T_ab were taken simultaneously every 30 s until a thoracic temperature of 40°C was attained. The beetle was then euthanized with an injection of ethyl acetate and allowed to cool. Once the T_thx had returned to within 1°C of T_a, the procedure was repeated for the same individual,now dead. A total of 12 N. hybridus individuals were used in this investigation. We used paired t-tests to compare the rates of warm-up between alive and dead T_thx and T_ab,between alive T_thx and T_ab and between dead T_thx and T_ab. If physiologically mediated heat transfer to the abdomen were occurring, one would expect the rate of abdominal heating to be higher in the living beetle than the dead beetle.

We measured wing loading to quantify differences in the amount of power output required for flight between the species. Species with a higher average wing loading may produce more excess heat in flight. To calculate wing loading, one wing was removed at its base from each beetle that flew successfully (N=137). Each beetle was weighed to the nearest milligram following flight and temperature measurements. Wing area was determined in the lab by scanning (Epson 636U scanner) wings taped to graph paper (five squares per centimeter) into NIH image® software, where we measured the area of each wing in mm². Wing loading was calculated in mg mm^–2 for each beetle.

To see how daily activity patterns were affected by temperature, two 24-h field observations were carried out on 28 June and 28 July at the South Bench site in Idaho. During each 24-h observation period, traps were checked every 2–3 h during the day and twice during the night (after midnight), and the number of beetles and species of each was recorded. During daylight hours,the number of beetles seen flying in the vicinity of the traps was also recorded. These observations allowed us to estimate the relative beetle activity at different times of the day.

We measured the T_thx and T_ab of beetles caught flying into a trap, recorded their weight and removed one wing to calculate wing loading (see previous section). Each time that traps were checked or a beetle was seen flying overhead, we recorded the T_a and T_e of a beetle in a flight position. Measurements from the other T_e models (on bare ground, 1–2 cm below ground, under foliage) were taken every 2–3 h during a 24-h observation period.

We also developed a model to predict T_thx for beetles throughout the day. We measured T_e throughout an entire day in late July 2001 and then applied these values to the regression equation of T_thx against T_e for each species to predict post-flight thoracic temperatures at each T_evalue. We then used the range of post-flight T_thx actually measured for the three species in flight trials and in the field and compared these ranges to the predicted T_thx throughout the day to estimate windows of possible flight times.

All statistical tests were carried out using StatView® 5.0 for Macintosh. ANOVAs were followed by Fisher's PLSD post-hoc multiple comparison tests. Paired t-tests were used to determine significant differences between pairs of measurements taken for individuals and to test if there were significant differences in T_thx pre- and post-flight, between T_thx and T_ab and in cooling/warming rates for living vs dead individuals. Simple,linear regression was used to test whether the relationship between T_b and T_te or T_awas significantly different from zero. Mean values, unless otherwise specified, are reported ±1 s.e.m.

Beetles warmed themselves on average 4.7±0.3°C before initiating flight, which represents a significant increase in T_thxprior to flight (paired t=18.852, mean difference=4.663, P<0.0001, N=136). The mean post-flight thoracic temperature (across species and sites) was 30.0±0.2°C(N=136) during flight trials. Post-flight T_thxwas significantly higher than T_a (paired t=19.358, mean difference=5.3°C, P<0.0001, d.f.=135). After flight, T_thx was also significantly higher than T_ab (paired t=11.852, mean difference=1.692, P<0.0001, d.f.=128). Among species, there was a significant difference in post-flight temperature excess(T_thx–T_a), with N. investigator having the highest mean excess (7.3±0.5°C),followed by N. guttula (4.5±0.3°C) and N. hybridus (3.7±0.4°C) (one-way ANOVA F=24.746, d.f.=2, P<0.0001). Most beetles went through a series of behaviors before actually attempting flight; these consisted of abdominal pumping, unfolding and extending both wings, elevation of elytra and grooming of antennae clubs and foretarsae. Some individuals would repeat these behaviors many times before flight. All flight trials took place in the shade and so the increase in T_thx between the initial pre-flight measurement and the measurement directly following short flight is most likely due to endogenous heat production prior to flight.

Post-flight temperature excess in the thorax decreased significantly with increasing ambient temperature (slope=–0.523, r²=0.303, N=135, P<0.0001). Above approximately 30°C, the temperature excess between T_thx and T_a becomes negative,indicating that beetles lost heat before or during flight, and no energy for warming up is required (Fig. 1). Post-flight T_thx did not differ significantly among species (for all Idaho and Colorado sites; one-way ANOVA, F=0.803, d.f.=2, P=0.4502), and post-hoccomparisons show that there were no significant differences between species.

Recordings of continuous T_thx measurements for four sustained (>5 min), tethered flights(Fig. 2) show that N. hybridus has the ability to sustain and regulate thoracic temperatures to some degree during long flights and that the length of the flight may be dependent upon ambient temperature, as the longest flight occurred at the lowest T_a. Mean T_thx for the four sustained flights was 31.5±2.6°C compared with a mean T_a of 26.13°C.

Nicrophorus hybridus is significantly larger than N. guttula (t=5.364, P<0.0001, d.f.=79) and N. investigator: N. hybridus mean mass=500±20 mg, N. guttula mean mass=340±20 mg and N. investigator mean mass=293±10 mg. The slopes of T_thxvs T_a and T_thxvs T_e of an operative model (thermoregulatory performance index) indicate that N. hybridus is better able to regulate body temperature prior to and during short flights compared with N. guttula and N. investigator. The slope of the regression of T_thx against T_a for N. hybridus (0.315; r²=0.227) is much closer to zero than the slopes calculated for N. guttula (0.771; r²=0.784) and N. investigator (0.610; r²=0.221) and is also significantly different from the slopes for these smaller species (N. hybridus vs N. guttula t=33.853, P<0.0001, d.f.=77; N. hybridus vs N. investigator t=14.7795, P<0.0001, d.f.=107). The slopes for T_thxvs T_e(thermoregulatory performance index) were similar to those comparing T_thx and T_a, and, again, the regression slope for N. hybridus (0.370; r²=0.411) is closer to zero than the slopes calculated for N. guttula (0.636; r²=0.679) and N. investigator (0.575; r²=0.327) and is also significantly different from the other two (N. hybridus vs N. guttula t=19.777, P<0.0001, d.f.=77; N. hybridus vs N. investigator t=12.059, P<0.0001, d.f.=107)(Fig. 3).

Beetles cooling after short flights (mean post-flight T_thx=30.4°C) cooled at the same rate as beetles that were artificially heated when artificial cooling rates were calculated from 30.4°C (t=0.275, d.f.=35, P=0.7847). No beetles warmed up following flight unless flight was initiated while the post-flight temperatures were being measured, in which case the trial was omitted from the analysis of cooling rates. Post-flight and artificial cooling rates (from 30.4°C) were then combined, and cooling rates among size classes were compared. As expected, based on surface area to volume ratios, large beetles cooled more slowly than small individuals and significantly slower than medium-sized individuals (one-way ANOVA, F=4.081, d.f.=2 P=0.0261; Fisher's PLSD, large vs medium mean difference=–0.745 deg. min^–1, P=0.0074). Beetles in the large size class (N=10) had a mean cooling rate of–1.062±0.121 deg. min^–1, beetles in the medium size class (N=21) had a mean cooling rate of–1.807±0.166 deg. min^–1, and beetles in the small size class (N=5) had a mean cooling rate of–1.610±0.341 deg. min^–1. Cooling rates for the three species following short flights are summarized in Table 1. Artificial cooling was only measured for N. investigator and N. guttula.

To determine the importance of the wings, elytra and thoracic pile as insulation for burying beetles, we compared the cooling rate of individuals assigned to one of three treatments: intact (I), wings and elytra removed(WER), and thoracic pile removed (TPR). There was a significant effect of treatment (F=4.332, P=0.0248, d.f.=2, power=0.696); beetles with elytra removed cooled significantly faster (mean=–0.145 deg. min^–1) than intact beetles (mean=–0.098 deg. min^–1) (Fisher's PLSD, P=0.0110). Beetles with thoracic pile removed also cooled faster than intact beetles, but only marginally (Fisher's PLSD, P=0.0655). Because of the small sample size in each treatment (I, N=14; WER, N=7; TPR, N=7), we increased the power of the test (α=0.10, power=0.816),which resulted in a significant difference between the cooling rates of intact beetles compared with those with insulation removed.

The rate of abdominal warming between living and freshly killed individuals was not different (paired t=0.989, P=0.3461, d.f.=10),indicating that living beetles do not actively shunt heat from the thorax to the abdomen. Live beetle T_thx increased on average 0.451 deg. min^–1 faster than in dead beetles, but the difference was not significant (paired t=1.898, P=0.0869, d.f.=10).

Wing loading was significantly different among species, with N. hybridus (mean=4.61±0.08 mg mm^–2) and N. guttula (mean=4.55± 0.13 mg mm^–2) exhibiting significantly higher levels of wing loading than those calculated for N. investigator (mean=3.37±0.09 mg mm^–2) (one-way ANOVA, F=58.617, d.f.=2, P<0.0001; Fisher's PLSD, N. guttula vs N. investigator mean difference= 1.182 mg mm^–2, P<0.0001; N. hybridus vs N. investigator mean difference=1.244 mg mm^–2, P<0.0001). There was also a positive relationship between the post-flight temperature excess (post-flight T_thx–T_a) and wing loading in N. hybridus (slope=1.189, r²=0.066, P=0.06) and N. investigator (slope=0.868, r²=0.026, P=0.292) but not for N. guttula (slope=–0.04, r²<0.0001, P=0.924). Wing loading measurements for each species are summarized in Table 1.

We found four species of burying beetle co-occurring at the two Idaho study sites: N. hybridis, N. guttula, N. investigator and N. defodiens. During a 24-h period, beetle activity began in mid-morning and then peaked in the late afternoon/early evening, when ambient temperatures were ∼20–30°C. Around mid-day, ambient temperatures ranged from 25°C to 30°C. Although this temperature range is moderate, the temperature of operative models on the ground surface and suspended approximately 1 m above the ground in a flight posture often approached lethal temperatures, especially during periods of full sun (mean lethal temperature from previous cooling experiments= 46.15±0.6°C, N=8; M. J. Merrick and R. J. Smith, unpublished data; Fig. 4).

The windows of activity time available for flight throughout the day differed among the three species. Based on the amount of time throughout the day that the post-flight T_thx ranges for the three species(N. hybridus, 24.4–32.7°C; N. guttula,22.4–34.5°C; N. investigator, 21–38°C) fell within the predicted thoracic flight temperatures (based on the species-specific regression of T_thxvs T_e), it appears that N. investigator would have the largest potential window of activity time, followed by N. hybridusand then N. guttula (Fig. 5). This assumes that the post-flight T_thxmeasured for each species following short flights approximates actual T_thx in the field and that the range of ambient air temperatures within which we collected flight data reflects the range of ambient air temperatures in which these species are active under field conditions.

Following short flights, the T_thx was significantly higher than the T_thx measured prior to flight, and significantly higher than ambient air temperatures. This indicates that each species considered here has the ability to elevate T_thxabove T_a either prior to short flights or during them. Because the captive flights were short (≤5 s), it is more likely that T_thx is elevated prior to taking off. The mean post-flight T_thx excess was significantly different among the three species, and the degree to which post-flight T_thx is elevated above T_a does not appear to be related to body mass, and only very weakly related to wing loading, in N. hybridusand N. investigator. However, the ability to regulatethoracic temperature independent of T_a or T_e appears to be directly related to the mean body mass of a species. The difference between T_thx and T_a or, more accurately, T_e is attained by temperature regulation on the part of the organism, and in this study N. hybridus, the species with the largest mean mass, was better able to regulate T_thx in short flights independent of T_e or T_a(Fig. 3). Both the slope of T_thx on T_a and the index of thermoregulatory precision for N. hybridus are similar to published values for insects considered to be good thermoregulators(Table 2). The slope of T_thx on T_e (index of thermoregulatory performance) is considered more useful here as it provides a better, more realistic measure of the degree to which an organism regulates its T_b (Hertz et al.,1993; Bishop and Armbruster,1999).

Data from four continuous, sustained flights(Fig. 2) show that N. hybridus can maintain elevated T_thx over long periods of flight. Longer flights may be possible at cooler T_as,suggested by the fact that the longest flight occurred at 28°C, the coolest T_a measured for any of the continuous flights. It is possible that smaller species also regulate T_thx during sustained flight, based on results from Oertli(1989) where small beetles(7–93 mg, Nicrophorus size range=110–880 mg) regulated T_thx during short flights (≤5 s) viatemperature-dependent changes in wing beat frequency. Further studies of continuous flights, including measurements of wingbeat frequencies for the smaller species (N. guttula, N. investigator) are warranted. Although continuous flight data obtained for N. hybridus indicate that this species is able to maintain a relatively constant T_thxduring longer flights, one caveat is that these results do not fully represent actual flight conditions. Continuous flight data were taken from individuals flying while supported by a thermocouple probe, and so the T_thx measurements may not take into account effects of generating lift and thrust that would occur in free flight. During free flight, lift and thrust, in addition to heat generated from muscle contractions, may increase T_thx during longer flights above measurements of T_thx made in this study, especially at low wind speeds where less heat is lost via forced convection(Church, 1960; Casey, 1992). Tethered flights also do not allow an animal to carry out normal flight behavior and do not consider the effects of solar radiation during flight(Casey, 1992). Field data indicate that flight activity ceases when solar radiation is most intense(Fig. 4), and T_e models confirm that activity during these times could potentially be lethal, so wind speed and solar radiation may limit the degree to which a beetle can regulate T_thx during actual flights in the field.

Because of an inherently lower surface area to volume ratio, larger insects should be able to regulate and maintain higher temperature excesses because they cool more slowly and have a higher thermal inertia(Bartholomew, 1981; Stone and Willmer, 1989b). This is true for Nicrophorus and is evident in the marked difference in mass and subsequent thermoregulatory ability between N. hybridus, N. investigator and N. guttula. Similar results have been shown for other beetles much larger than Nicrophorus (Bartholomew and Casey, 1977a,b; Bartholomew and Heinrich, 1978)and for other insects such as desert robber flies(Morgan and Shelly, 1988),where larger species were better thermoregulators.

Data from beetles cooling from flight or from artificial heating provide evidence that larger beetles may be able to maintain elevated T_thx longer after flight, which may confer a competitive advantage once a carcass is located. Because of slower cooling rates, larger beetles stay warmer longer, which may be one reason why larger beetles tend to be more successful in competitive interactions(Otronen, 1988; Trumbo, 1990). This would need to be tested with a series of carcasses in the field, where body temperatures of individuals arriving at the carcass and later competing for the carcass were measured, as was done previously for dung beetles arriving at dung piles(Bartholomew and Heinrich,1978). Controlled laboratory experiments in which an individual beetle is warmed to a temperature that approximates an after-flight T_thx and is then placed in an arena with a carcass and other beetles (potential competitors) may be a way to approach this question. The use of infrared thermocouples is an appealing, non-invasive approach for obtaining field temperature measurements, allowing for more natural behavior.

Wings, elytra and thoracic pile all played a role in slowing heat loss, and beetles with both wings and elytra removed cooled significantly faster. Removal of thoracic pile also resulted in faster cooling rates. Other studies have reported that thoracic pile had no effect on the cooling rates of beetles(Nicolson and Louw, 1980; Morgan, 1987; Chown and Scholtz, 1993), but this does not seem to be the case in Nicrophorus as the cooling rates of beetles with and without pubescence differed. A difference in cooling rates with and without pubescence is also the case for desert locusts (Church,1960b), where the insulating ability of the pubescence was dependent upon its density, and potentially for Colias butterflies, where it was shown that fur thickness increased with elevation(Kingsolver, 1983). Nicrophorus species vary in the density of thoracic pile and the actual area covered by the pile (M. J. Merrick and R. J. Smith, personal observation). Studies to determine if this variation is related to a species'thermoregulatory ability and distribution would be of great interest.

During flight, Nicrophorus flies with the elytra elevated and wings extended. With these structures held away from the body, a beetle in flight probably loses a great deal of endogenous heat by forced convection(Church, 1960; Casey, 1992). Even without air movement, beetles cooled significantly faster with wings and elytra removed,which indicates that this is a substantial avenue for heat loss during flight.

The ability to regulate body temperatures in flight requires that heat be generated and maintained at cooler ambient air temperatures and dumped at higher air temperatures. One mechanism for dumping excess heat is to shunt it to the abdomen, where it dissipates faster because of the large surface area of the abdomen and increased airflow and convection during fast flight(Heinrich, 1993, 1996). This mechanism of heat transfer is common in large moths (Heinrich, 1993, 1996) but it has not been observed in beetles (Chown and Scholtz,1993), Nicrophorus (present study), honeybees (which have counter-current heat exchangers in the petiole; Casey, 1992) or in cicada killer wasps (Coelho, 2001). Although there does not appear to be any physiological mechanism for control of heat transfer to and from the abdomen, the fact that the abdomen of Nicrophorus is significantly cooler than the thorax following short flights means heat can be lost to the abdomen by simple diffusion of hemolymph from the thorax to the abdomen. This is supported by the observation that abdomens of living and dead beetles, shielded from a heat source, heated up at the same rate. Because the abdomen is exposed during flight and contains large spiracles along its margins, the opportunity for heat to be lost by forced convection is great. This presents a problem for retaining heat during flights at cooler temperatures if no mechanism (such as counter-current heat exchangers) to prevent heat dissipation to the abdomen exists(Casey, 1992; Heinrich, 1996). Inability to control heat loss to the abdomen in cooler temperatures may limit the ambient temperatures at which flight is possible for Nicrophorus.

Higher wing loading increases the amount of heat produced as the flight muscles do more work to beat faster and maintain lift. Wing loading increased with body size in Nicrophorus, and wing loading and body mass were also negatively correlated with the slope of T_thx against T_a and the index of thermoregulatory performance,indicating that better thermoregulators had higher wing loading and were heavier in general. These results are consistent with those found for other beetles (Oertli, 1989),noctuid moths (Casey and Joos,1983) and bees (Stone and Willmer, 1989b; Bishop and Armbruster, 1999).

Another mechanism for thermoregulation that may be important in Nicrophorus includes heat dissipation from the head. Studies on thermal stability in honeybees (Roberts and Harrison, 1999), desert carpenter bees(Chappell, 1982), dragonflies(May, 1995) and cicada killer wasps (Coelho, 2001) indicate that cooling at high temperatures is facilitated by shunting warm hemolymph to the head, where heat is then dissipated over either a large surface area as in carpenter bees or via regurgitated fluid (see Heinrich, 1996, chapter 6 for a review). This mechanism for cooling is a possibility for Nicrophorus, as burying beetles regularly secrete fluid from the mouth and anus. Because of these secretions, however, water balance becomes important for burying beetles. Without access to water, beetles in captivity quickly die (M. J. Merrick and R. J. Smith, personal observation), and Nicrophorus are not common in hot, dry habitats.

Nicrophorus species found in southeastern Idaho are highly influenced by the thermal environment they experience. Operative temperature models can be used to gauge the thermal environment that an organism is experiencing and how habitat features such as substrate, orientation and solar radiation affect body temperatures. High operative temperatures corresponded to times of inactivity for beetles in the field(Fig. 4) and, given the wide fluctuations in operative temperatures, it is clear that microhabitat choice could influence the body temperature of an individual. Solar radiation is likely to play a large role in limiting the activity of Nicrophorusnot only in flight but also in terrestrial activity. Nicrophorusflies at a lower T_thx range (20–38°C) than other diurnal insects that can tolerate high heat loads, such as the hawkmoth Macroglossum stellatarum(Herrera, 1992; T_thx range 39–46°C) or cicada killers(Sphecius speciosus; Coelho,2001; T_thx range 37–42°C). Members of Nicrophorus probably cannot tolerate high heat loads imposed by flying at midday on warm, calm and clear summer days. On windy days, the window of flight opportunity may widen, as high wind speeds and fast forward flight increase convective heat loss(Casey, 1992). Beetles left on the ground in direct sunlight quickly die (M. J. Merrick and R. J. Smith,personal observation) and a pair of beetles tending to a carcass left on bare soil (where T_e rapidly approaches lethal temperatures around 45–46°C) will quickly work together to move the carcass under vegetation (M. J. Merrick and R. J. Smith, personal observation).

Flight activity appears to be restricted to ambient temperatures between approximately 14°C and 36°C, which would, at higher elevations,restrict flight to primarily a diurnal activity. If thermal tolerances for flight and terrestrial activity are determined for a species, one might be able to predict areas where nocturnal flight is possible, based on the mean nighttime temperature for a given habitat and the thermoregulatory abilities of the species being studied. This information could also allow for predicting where a species is distributed geographically (in latitude and elevation) and what habitat types it might utilize. For example, Nicrophorus nigritais a burying beetle that occurs along the Pacific coast of North America but lacks dorsal maculations on the elytra (i.e. it is completely black). Sikes(1996) found that this beetle is not active during the middle of the day and that it preferred to locate and bury carcasses in “moist, cool, redwood-forested canyons”. Perhaps because of its dark pigmentation, this species cannot tolerate high incident sunlight, or higher operative temperatures, leading one to predict that it would live in a shady habitat or have crepuscular or nocturnal activity patterns. Conversely, this species' black pigmentation may be selectively advantageous for heating up faster in cooler environments.

Temperatures that are restrictive to flight are not necessarily restrictive to other activity, and temperatures restrictive to flight in one species may not restrict flight in another. In the present study, N. investigatorflew over a wider range of ambient air temperatures than the other two species, and elevated T_thx on average 7.3°C above ambient. Since most of the N. investigator individuals documented in this study came from Colorado (higher elevations, cooler daily temperatures),this population may be able to tolerate a wider range of environmental temperatures. Beetles were also observed beneath carcasses and walking near them or in baited traps at night and at dawn (temperatures between 12°C and 15°C), with a mean T_thx excess of approximately 4.5°C.

Predictions of potential flight times for the three species(Fig. 5) show that the largest species (N. hybridus) and the species with the widest thermal tolerance (N. investigator) may be able to be active longer throughout the day compared with N. guttula. These predictions for activity times are based on the range of ambient temperatures recorded following each flight during flight trials but do not consider how well each species is able to regulate its body temperature during flights throughout the day. N. investigator may have the widest `window' of possible activity times but may not be able to fly for very long before it becomes too cool or too hot to maintain flight. By contrast, because N. hybriduscan regulate its body temperature better, it may be able to fly for longer periods throughout the day. For a beetle that must fly in search of carcasses for food and reproduction, the capability for sustained flights might increase an individual's chances of finding the rare carcass resource and thus their subsequent fitness.

This study incorporates body temperature data, thermal profiles of various microhabitats and actual observations of animal activity in the field to show that (1) burying beetle activity is influenced by environmental temperatures,(2) burying beetles have the ability to elevate thoracic temperatures prior to flight and (3) thermoregulation during flight is influenced by body mass,morphological features such as wing loading and insulation. We provide a preliminary framework for predicting and testing hypotheses about burying beetle activity times and distributions based on thermal tolerances and thermoregulatory ability. Body size has been shown to influence competitive outcomes (Otronen, 1988; Trumbo, 1990), speed of carcass burial (Smith et. al. 2001) and reproductive success(Trumbo, 1990) in burying beetles and we show here that it also influences thermoregulatory ability,which may help to further explain these observed relationships between body size and fitness. We suggest that additional studies examining the relationships between body size, morphology and thermoregulatory ability and determining thermal tolerances and how these relate to distributions and activity times among different species of burying beetles will advance our current understanding of species distributions, niche partitioning among sympatric species and the relationship between body size and reproductive success.

We thank C. F. Williams, N. Huntly and an anonymous reviewer for providing critical reviews and suggestions that greatly improved this article. Partial support for this project was provided by the Idaho State University Graduate Research and Scholarship Committee, grant no. F00-06 to M.J.M. The work reported here was part of the work submitted to Idaho State University in partial fulfillment of the requirements for an M.Sc. degree.