Oxygen consumption and body temperature of active and resting honeybees

Journal of Insect Physiology 49 (2003) 881–889 www.elsevier.com/locate/jinsphys

Oxygen consumption and body temperature of active and resting honeybees Auton Stabentheiner ∗, Jutta Vollmann, Helmut Kovac, Karl Crailsheim Institut fu¨r Zoologie, Universita¨t Graz, Universita¨tsplatz 2, A-8010 Graz, Austria Received 12 March 2003; accepted 9 June 2003

Abstract We measured the energy turnover (oxygen consumption) of honeybees (Apis mellifera carnica), which were free to move within Warburg vessels. Oxygen consumption of active bees varied widely depending on ambient temperature and level of activity, but did not differ between foragers (⬎18 d) and middle-aged hive bees (7–10 d). In highly active bees, which were in an endothermic state ready for flight, it decreased almost linearly, from a maximum of 131.4 µl O2 min⫺1 at 15 °C ambient temperature to 81.1 µl min⫺1 at 25 °C, and reached a minimum of 29.9 µl min⫺1 at 40 °C. In bees with low activity, it decreased from 89.3 µl O2 min⫺1 at 15 °C to 47.9 µl min⫺1 at 25 °C and 14.7 µl min⫺1 at 40 °C. Thermographic measurements of body temperature showed that with increasing activity, the bees invested more energy to regulate the thorax temperature at increasingly higher levels (38.8– 41.2 °C in highly active bees) and were more accurate. Resting metabolism was determined in young bees of 1–7 h age, which are not yet capable of endothermic heat production with their flight muscles. Their energy turnover increased from 0.21 µl O2 min⫺1 at 10 °C to 0.38 µl min⫺1 at 15 °C, 1.12 µl min⫺1 at 25 °C, and 3.03 µl min⫺1 at 40 °C. At 15, 25 and 40 °C, this was 343, 73 and 10 times below the values of the highly active bees, respectively. The Q10 value of the resting bees, however, was not constant but varied in a U-shaped manner with ambient temperature. It decreased from 4.24 in the temperature range 11–21 °C to 1.35 in the range 21–31 °C, and increased again to 2.49 in the range 30–40 °C. We conclude that attempts to describe the temperature dependence of the resting metabolism of honeybees by Q10 values can lead to considerable errors if the measurements are performed at only two temperatures. An acceptable approximation can be derived by calculation of an interpolated Q10 according to the exponential function VO2 = 0.151 × 1.0784Ta (interpolated Q10 = 2.12).  2003 Elsevier Ltd. All rights reserved. Keywords: Oxygen consumption; Resting metabolism; Body temperature; Activity; Thermography; Honeybee

1. Introduction The honeybee (Apis mellifera L.) is a heterothermic insect. At rest and during many duties inside the colony, she is ectothermic. She changes to the endothermic state throughout the foraging cycle (Esch, 1960; Heinrich, 1979; Dyer and Seeley, 1987; Schmaranzer and Stabentheiner, 1988; Waddington, 1990; Stabentheiner, 2001; Stabentheiner et al., 1995, 2002), and whenever additional heat is necessary to regulate the brood nest temperature (Esch, 1960; Koeniger, 1978; Ritter, 1982; ∗ Corresponding author. Tel.: +43-3163805705; fax: 3163809875. E-mail address: [email protected] Stabentheiner).

+43(A.

0022-1910/$ - see front matter  2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0022-1910(03)00148-3

Kronenberg and Heller, 1982; Harrison, 1987; Stabentheiner and Kovac, 2002) or to prevent chill coma and freezing in the swarm cluster (Heinrich, 1981) or winter cluster (Esch, 1960; Stabentheiner et al., 2003). During their life, individual honeybees can encounter a wide variety of ambient temperatures, from below freezing point to more than 40 °C (Heinrich, 1993). The energy turnover of the bees in both the ectothermic and endothermic state varies widely depending on the ambient temperature. With decreasing ambient temperature, it decreases in the resting (ectothermic) bees (Allen, 1959; Rothe and Nachtigall, 1989) but increases in the thermally active (endothermic) bees (Rothe and Nachtigall, 1989; Southwick, 1991; Crailsheim et al., 1999b; Moffatt, 2001; Blatt and Roces, 2001). However, the dependence of the energy turnover on ambient temperature of

882

A. Stabentheiner et al. / Journal of Insect Physiology 49 (2003) 881–889

both resting and thermally active (non-flying) bees has not yet been determined satisfactorily (Crailsheim et al., 1999b). In the resting bees, measurements were either made at only one ambient temperature (Goller and Esch, 1991), or it was not completely sure that all bees were truly resting, i.e. not walking and ectothermic (Allen, 1959; Abrol, 1992; Crailsheim et al., 1999b). In active bees, either the degree of activity was not recorded (Southwick, 1991; Blatt and Roces, 2001; Moffatt, 2001) or the bees were only investigated at a few ambient temperatures (Kosmin et al., 1932; Crailsheim et al., 1999b). Here, we compare the oxygen consumption and the thorax temperature of individual honeybees of different ages (1–7 h old young bees without the capability of active heat production, 5–13 d old middle-aged hive bees, and ⬎18 d old foragers) depending on their level of activity, in a wide range of ambient temperatures which individual honeybees are likely to experience throughout their life (10–40 °C).

perature sensitive because of the expansion of gases with increasing temperature. Therefore, before the start of the measurements, we investigated the effect of endothermic heat production by the bees on the air temperature in a glass vessel of a size similar to that of our 95 ml Warburg vessels. In visually highly active bees which are highly endothermic (Stabentheiner and Crailsheim, 1999), the temperature increase above the water bath temperature of 25.0 °C was only 0.17 ± 0.12 °C on average (n = 34 from three bees, maximum: 0.4 °C, measured with calibrated thermocouples). This demonstrates that the heat dissipation by the stirred water bath was effective. A model calculation with an oxygen consumption of 81.1 µl O2 min⫺1 at 25 °C (highly active bees seen in Fig. 1) showed that assuming a 0.17 °C (average) or 0.4 °C (maximum) higher air temperature inside the vessels reduced the values of oxygen consumption from only 0.06% to 0.13%, respectively. Assuming a “worst case scenario” of 131.39 µl O2 min⫺1 (mean of highly active bees at 15 °C, Fig. 1) and a 1 °C temperature increase inside the vessels, not considering this tempera-

2. Materials and methods Brood combs of Apis mellifera carnica Pollmann were kept in an incubator at 34 °C. Young bees were taken from the comb immediately after emergence, weighed and transferred to 12 ml Warburg vessels. Some of the young bees were marked and returned to a standard colony. At the age of 7–10 d (“middle-aged” hive bees), they were taken from the colony and their oxygen consumption was measured. Foragers were taken from feeding places where they foraged for sucrose solution (1–2 mol l⫺1). Control dissections showed that each bee had at least 16 mg of sugar solution in her honey stomach at the end of the experiments, so no bee was hungry while the measurements were taken. The fresh weight of the bees before the oxygen measurements ranged from 95 to 160 mg, 90% of the individuals weighing 100 mg or more. Measurements of oxygen consumption were made at temperatures of 10–40 °C according to standard procedures of Warburg manometry. The young bees were placed individually in small Warburg vessels with a volume of 12 ml. According to the expected higher values of oxygen consumption, individual “middle-aged” hive bees and the foragers were placed in large 95 ml vessels. The CO2 produced was absorbed by 20% KOH (small vessels: 250 µl, large vessels: 1 ml) in a cuvette with a perforated top containing a strip of folded filter paper to increase the absorbing surface. Water vapour was absorbed by 300 and 600 mg of silica gel in a side chamber of the small and large vessels, respectively. In the large vessels, an additional amount of 1 g was added to a perforated cuvette. All vessels were submerged in a stirred water bath for temperature control to the nearest 0.05 °C. However, constant volume respirometry is tem-

Fig. 1. (a) Oxygen consumption and (b) activity (see Table 1) of adult and young bees which were free to move within Warburg vessels, in dependence on ambient temperature. Vertical bars indicate SD. Number of values written beside means. For the conversion of µl O2 min⫺1 bee⫺1 to mW bee⫺1, a caloric equivalent of oxygen of 21.117 kJ l⫺1 O2 was used (RQ = 1; Rothe and Nachtigall, 1989).

A. Stabentheiner et al. / Journal of Insect Physiology 49 (2003) 881–889

883

Table 1 Activity levels, “mean behaviour scores”, and observed behaviours (5 s intervals) during the oxygen and temperature measurements in freely moving young bees, hive bees, and foragers. In young bees, only two activity levels were calculated (Ay and By) Activity level

“Mean behaviour score”

Observed behaviours

A

Small or no movements

0–1.5

B

Low activity

⬎1.5–3.5

C

Medium activity

⬎3.5–5.5

D

High activity

⬎5.5

(0) No body part moving (1) Sitting with movements of antennae or single legs and/or sporadic abdominal movements (2) Self-grooming (3) Sitting + walking a few steps (4) Walking with low speed (5) Walking with medium speed (6) Fast walking (7) Fast walking + flight intentions (8) Short flights + fast walking (9) Flight against the box wall (seldom observed)

ture increase might have led to an overestimation of oxygen consumption by not more than 0.35%. As oxygen consumption decreases, endothermy (and thus the temperature increase inside the vessels) decreases in parallel (Rothe and Nachtigall, 1989; Esch et al., 1994; Crailsheim et al., 1999b). So we are sure that errors due to temperature effects caused by endothermy of the bees remained below 0.5%. Since the fresh weight of honeybees may vary by more than 100% (~75 to ~165 mg) to the greatest part depending on the contents of the honey stomach and the intestine, data are presented here in µl O2 min⫺1 bee⫺1 at standard conditions (oxygen volume at 760 Torr = 101.323 kPa and 0 °C). To convert to µl O2 min⫺1 mg⫺1, it seems appropriate to divide the reported values by a “standard bee fresh weight” of 80 mg, which equals an empty honeybee with a few milligram of sugar for potential endothermy in her honey stomach (Allen, 1959). The bees were allowed to move freely within the Warburg vessels. After the vessels had been equilibrated for 10 min in the stirred water bath, the O2 consumption (decrease of pressure) was read every 3 min for 15–18 min (5–6 readings). If oxygen consumption was too low to be measured, the measurement intervals were extended to 6 min (5.2% of measurements) or 9 min (0.8%). Especially at 10 °C, the measurement interval had to be extended up to 18 min in a few cases (0.9%). Afterwards, the bees were supplied with fresh air. After the vessels had equilibrated for another 10 min, the measurements were repeated. Oxygen consumption was read with a resolution of ca. ±0.044 and ±0.26 µl O2 min⫺1 for the 3 min measurement intervals with the small and large vessels, respectively. The behaviour of the bees was recorded in intervals of 5 s. Behaviours were classified into 10 categories based on the level of activity, as defined in Table 1. If the bees performed two different activities during the 5

s, the longer lasting behaviour was recorded (Crailsheim et al., 1999b; Stabentheiner and Crailsheim, 1999). In order to allow comparison of the O2 consumption with the behavioural observations, the “mean behaviour score” during the measurement periods of O2 consumption had to be determined. Each behaviour was assigned an activity number from 0 to 9, based on the expected level of energy expenditure of that behaviour (Table 1). Then, each bee’s “mean behaviour score” for the period of oxygen measurement was determined by weighting the activity numbers of each of her observed behaviours by the frequency of occurrence of that behaviour in the measurement period. In the middle-aged hive bees and in the foragers, the mean behaviour scores were grouped into the four activity levels of Table 1, which resulted in frequency distributions with maximum observations within each of the four activity levels (Crailsheim et al., 1999b). To determine the true resting metabolism, only sporadic abdominal movements (1–2 per 2 × 5 s) were allowed in the young bees. The O2 consumption of adult bees was compared with their body temperature under conditions comparable with those during the Warburg measurements. Hive bees and foragers (age 5 to ⬎18 d) were placed in small boxes (volume ca. 120 ml) in an incubator with a fluorescent tube light inside. Inside the boxes, the bees had access to a small pot of diluted honey. The body surface temperature was measured thermographically with an AGA 782 SW real-time infrared camera through an infrared transmissive plastic front cover of the boxes (Schmaranzer and Stabentheiner, 1987; Stabentheiner and Crailsheim, 1999). Body surface temperature was calibrated using the cuticular emissivity of the honeybee (0.97) and an AGA 1010 reference radiator (Stabentheiner and Schmaranzer, 1987; Schmaranzer and Stabentheiner, 1988). The thermographic scenes were stored on a video tape for later analysis. The bees’ ambient air temperature was measured inside each box

884

A. Stabentheiner et al. / Journal of Insect Physiology 49 (2003) 881–889

within 4 cm of the bees by NiCr/Ni thermocouples (diameter of wires: 0.2 mm). In order to be sure that the young bees were not capable of endothermic heat production with their flight muscles, bees that had emerged from brood combs in an incubator were placed in small boxes (volume ca. 100 ml) in an illuminated incubator at the age of 0.5, 2, 4, 6, 9, and 12 h. The boxes were mounted on a laboratory shaker. Before the temperature measurements, the shaker was switched on for 20 s at 1000 rounds/min to stimulate the bees to heat up. Their body temperature was measured by infrared thermography.

3. Results The oxygen consumption (VO2) varied in a wide range, from 0.049 to 201.2 µl O2 min⫺1 in the foragers (age ⬎18 d), from 0.815 to 149.2 µl O2 min⫺1 in middleaged hive bees (age 7–10 d), and from 0.012 to 12.3 µl O2 min⫺1 in the young bees (age 1–7 h). It depended strongly on ambient temperature and level of activity. To enable comparisons between the different age classes and ambient temperatures, the original values of VO2 were averaged with respect to the four levels of activity of Table 1. There was no difference (11 of 16 cases) or no consistent direction of the difference (five of 16 cases) in the mean values between the foragers and the middle-aged hive bees (Table 2). Therefore, the

measurements of these two age classes were pooled for further treatment (“adult bees”, Fig. 1). In the highly active adult bees, mean oxygen consumption increased nearly linearly from 29.9 µl O2 min⫺1 at 40 °C to 81.1 µl O2 min⫺1 at 25 °C, and a maximum of 131.4 µl O2 min⫺1 at 15 °C ambient temperature (Fig. 1a). This means that on average they increased oxygen consumption by 40.7 µl O2 min⫺1 per 10 °C decrease of ambient temperature. In the bees with medium and low locomotor activity, a similar increase at lower levels was observed. This increase, however, was not accompanied by an increase of activity. The bees’ activity varied considerably in the investigated range of ambient temperature (see mean curves in Figs. 1b and 2c). Figs. 1b and 2c show that our way of classifying behaviour led to activity levels (average of mean behaviour scores) which were similar—and thus comparable—within the whole range of ambient temperature (10–40 °C). In the adult bees which showed no or only small movements, both VO2 and activity remained at a low level at all ambient temperatures (Fig. 1). At 10 °C, the bees could not keep up a high metabolism long enough to allow comparable measurements. Therefore, oxygen consumption was lower than expected by extrapolation from the higher temperatures (Fig. 1). Under conditions which resembled the situation in the Warburg experiments, the highly active adult bees regulated their thorax temperature at a high level (Fig. 2a). Mean thorax surface temperature amounted to 38.8–41.2

Table 2 Comparison of the oxygen consumption of foragers (⬎18 d) and middle-aged hive bees (7–10 d) in dependence on activity level (compare Table 1) and ambient temperature (Ta). N, number of measurements Activity level

Ta (°C)

t test (P ⬍)

Oxygen consumption (µl O2 min⫺1) Foragers (⬎18 d) Mean

Middle-aged bees (7–13 d) SD

N

Mean

SD

N

A

15 25 30 35

27.89 3.50 1.35 5.55

22.976 2.900 1.036 5.249

12 10 6 6

13.00 7.92 24.72 17.45

11.931 11.447 17.036 19.092

30 22 9 9

0.01 n.s. 0.01 n.s.

B

15 25 30 35

97.70 46.57 16.31 21.07

23.219 28.467 15.737 19.583

13 31 14 55

61.79 52.11 28.53 16.58

16.760 17.482 12.355 13.604

4 10 16 30

0.02 n.s. 0.05 n.s.

C

15 25 30 35

115.37 73.86 46.42 30.30

37.655 38.481 20.841 30.076

38 57 26 40

98.60 70.07 58.02 47.12

18.742 24.307 26.354 17.043

6 28 24 18

n.s. n.s. n.s. 0.05

D

15 25 30 35

133.12 83.43 60.66 41.15

18.938 17.012 25.905 26.803

21 23 25 41

129.26 80.67 66.85 52.27

16.076 34.196 20.315 15.376

17 29 32 17

n.s. n.s. n.s. n.s.

A. Stabentheiner et al. / Journal of Insect Physiology 49 (2003) 881–889

885

°C. With decreasing (locomotor) activity, the thorax was regulated at lower levels (Fig. 2a). The general activity of the young bees decreased from 22 to 10 °C (see mean curve in Fig. 3b). At 10 °C, movements came to an end soon after insertion in the Warburg vessels, and they succumbed to chill coma. At higher ambient temperatures, however, there was no further increase but a significant decrease towards 30–35 °C. Towards 40 °C, the mean activity increased again. In the young bees, walking was always more clumsy and slower than in the older ones. Therefore, as only the observed behaviours 0–2 of Table 1 could be directly compared with the older bees, only two levels of activity were calculated (Ay and By). True resting metabolism was determined in 1–7 h old young bees. In order to be sure that these bees were not endothermic, we measured their body surface temperature by means of infrared thermography and their activity under conditions comparable to the situation in the Warburg vessels. They were classified as endothermic when the temperature of their thorax (the source of heat production) was more than 0.1 °C

Fig. 2. (a) Thorax surface temperature (Tth), (b) thoracic temperature excess (Tth⫺Ta), and (c) activity (see Table 1) of adult bees (hive bees and foragers, age ⱖ5 d) which were free to move within small boxes, in dependence on ambient temperature (Ta). Vertical bars indicate SD. Number of values written beside means.

Fig. 3. (a) Oxygen consumption and (b) activity of truly resting (Ay) and active (By) young bees which were free to move within Warburg vessels, in dependence on ambient temperature and level of activity (see Table 1). Vertical bars indicate SD. Number of values written beside means. The mean activity (circles) in (b) was significantly lower at 10, 15, 30 and 35 °C than at 22, 25 and 40 °C (P ⬍ 0.0001; t test).

886

A. Stabentheiner et al. / Journal of Insect Physiology 49 (2003) 881–889

higher than those of the head and abdomen (Thead ⬍ Tthorax ⬎ Tabdomen; Stabentheiner et al., 2003). Despite the heavy vibrational stimulation, up to an age of 9 h, no resting or slightly moving young bee out of the 16 (observed behaviours 0–3; Table 1) was endothermic at ambient temperatures of 16, 25 and 33 °C. Therefore, we conclude that by observing the young bees’ activity during the oxygen measurements, we determined the basal metabolism in truly resting animals. The resting VO2 of the young bees increased with ambient temperature, from 0.212 µl O2 min⫺1 at 10 °C to 1.12 µl O2 min⫺1 at 25 °C, and 3.03 µl O2 min⫺1 at 40 °C (Fig. 3a). This was 343, 73 and 10 times below the values of the highly active bees at 15, 25 and 40 °C, respectively (compare Fig. 1a). However, the VO2 of the resting young bees was not strictly correlated with activity. Between 22 and 40 °C, VO2 increased by a factor of 1.98, whereas the activity remained approximately constant (Fig. 3). At lower temperatures (15 and 10 °C), the decrease in resting VO2 resembled the decrease in activity (Fig. 3). In the active young bees, VO2 was at a higher level but increased also with ambient temperature. However, there was also no comparable increase of activity (Fig. 3). Interpolation of the data on the resting metabolism by the function VO2 = 0.15096 × 1.0778Ta provided an interpolated van’t Hoff Q10 value of 2.12, i.e. VO2 was more than doubled as the ambient (and thus body) temperature increased by 10 °C. However, there were considerable deviations from this interpolated curve, with a plateau between 22 and 30 °C (Fig. 3a). Accordingly, the Q10 value varied in a wide range in a U-shaped manner (Fig. 4). It decreased from 4.24 in the temperature range 11–21 °C to 1.35 in the range 21–31 °C, and increased again to 2.49 in the range 30–40 °C.

Fig. 4. Q10 values of the resting oxygen consumption of young bees as determined from the spline interpolation of Fig. 3a. The mean Q10 value (2.42 ± 0.99, N=21) is different from the value derived from the fitted curve of Fig. 3a (2.12).

4. Discussion The oxygen consumption of individual bees varied widely. At 15 °C ambient temperature, the requirements of endothermic temperature stabilization caused the highly active (but not continuously flying) adult bees to increase the VO2 by a factor of 343 above the resting level. A similar effect of the activity level on VO2 was reported by Kosmin et al. (1932) and Crailsheim et al. (1999b). This way, the highly active bees could maintain a mean thoracic temperature excess of up to 22 °C (Fig. 2). The oxygen consumption of the bees with low, medium or high activity increased considerably with decreasing ambient temperature (Fig. 1a). In contrast to our results, Allen (1959) did not report a steady increase of the oxygen consumption with decreasing ambient temperature. This is due to the measurement procedure. Both body temperature and oxygen consumption of bees which are immobilized, as was the case in Allen (1959), are not easily predictable and usually remain at a considerably lower level than in freely moving bees (Stabentheiner and Crailsheim, 1999; Crailsheim et al., 1999b). An increase similar to that in the present study but at a lower level was reported for small groups of 10 bees by Cahill and Lustick (1976), and for individual bees by Southwick (1991). Since, however, group size may influence the energy turnover of honeybees strongly (Southwick, 1985; Fahrenholz et al., 1992), our measurements can only be directly compared with those of Southwick (1991). His mean values were generally at a lower level, with a mean VO2 of about 39 and 4 µl O2 min⫺1 (for 80 mg bees) at 15 and 35 °C ambient temperature, respectively. For comparison, our highly active bees consumed 131.4 and 44.4 µl O2 min⫺1 at these temperatures, and even our bees with low locomotor activity had a higher turnover (89.3 and 19.5 µl O2 min⫺1, respectively; Fig. 1a). We suggest that the values reported by Southwick (1991) were lower than ours because he averaged all values without considering the bees’ level of activity. Blatt and Roces (2001) measured the VCO2 of Apis mellifera carnica with open-flow respirometry under similar conditions and in the same range of ambient temperatures as in the present study. Their data (29, 94 and 138 µl CO2 min⫺1 at 40, 25 and 15 °C, respectively) resembled those of our highly active bees (29.9, 81.1 and 131.4 µl O2 min⫺1, respectively). Therefore, we conclude that their bees regulated the thorax temperature at a similar high level as our highly active bees (compare Figs. 1, 2 and 5). In honeybees foraging 50% sucrose from artificial flowers at different reward rates, Moffatt (2001) determined a somewhat steeper increase of VCO2 with decreasing ambient temperature. At a reward rate of 6.5 µl min⫺1, VCO2 at 30 and 25 °C was 42% and 20% higher (~77 and ~117 µl min⫺1) than the VO2 in our highly

A. Stabentheiner et al. / Journal of Insect Physiology 49 (2003) 881–889

Fig. 5. Variation of (a) mean thorax surface temperature (Tth) of adult bees and (b) mean oxygen consumption invested to regulate Tth, in dependence on activity. Despite the variation in ambient temperature (compare Figs. 1 and 2), the mean Tth and the accuracy of thermoregulation increased with the level of activity. Bars indicate SD.

active bees (64 and 82.5 µl min⫺1). This is probably due to the fact that the bees of Moffatt (2001) were part of the time in flight between the artificial flowers. During continuous hovering and slow forward flight, bees exhibit a higher energy turnover than highly active bees prepared for flight. At an ambient temperature of 32 °C, Wolf et al. (1989) measured 135 µl O2 min⫺1 in 80 mg bees and 192 µl O2 min⫺1 in 140 mg bees during free flight at 0.5 m s⫺1 in a wind tunnel (highly active bees of this study: ~56.2 µl O2 min⫺1). Harrison et al. (1996) and Roberts and Harrison (1999) reported a linear decrease of CO2 production from 146 µl CO2 min⫺1 at 20 °C to 94 µl CO2 min⫺1 at 40 °C (highly active bees of this study: 106.5 and 29.9 µl CO2 min⫺1, respectively). The body temperature of our highly active bees (Fig. 2) was between that of bees gathering a 0.5 M and a 1 M sucrose solution from feeders with unlimited flow (Schmaranzer and Stabentheiner, 1988; Stabentheiner and Schmaranzer, 1988), and higher than in bees foraging on flowers (Heinrich, 1979; Kovac and Schmaranzer, 1996) and water gatherers (Schmaranzer, 2000).

887

The total means of the adult bees’ activity varied considerably between different ambient temperatures (Figs. 1b and 2c). Our way of classifying activity, however, not only reduced the variability of both activity and oxygen measurements, but even more importantly, it provided remarkably constant levels of activity, and thus comparability of activity levels and oxygen measurements throughout the investigated temperature range. Therefore, we conclude that the increase of VO2 with decreasing ambient temperature (Fig. 1a) was not the direct result of a higher (locomotor) activity, though at any one ambient temperature, VO2 correlated with activity. A comparison of Fig. 1 with Fig. 2 shows that the increase of VO2 with decreasing ambient temperature originates to the greatest part from endothermic heat production with the thoracic flight muscles. It is used to keep the thorax temperature high, which allows the bees to stay ready for flight and to move fast even at low ambient temperatures. At low ambient temperature, the thoracic leg muscles surely contributed only a small fraction to total energy turnover. At high ambient temperatures, a higher relative contribution has to be assumed, because despite the constant level of activity, total heat production was less (compare Fig. 1). In the measurement situations in the Warburg vessels and in the small boxes, the motivation to escape was high in the active adult bees. The higher the desired activity (and escape motivation), the more energy was invested to regulate the thorax temperature at an increasingly higher level (Figs. 1, 2 and 5). However, not only was the level increased but also the accuracy of thermoregulation was significantly improved with activity and escape motivation. The bees compensated much better for variations in ambient temperature (Figs. 2 and 5). This resembles the situation found in dancing honeybees (Stabentheiner et al., 1995). There, highly motivated dancers (which had foraged sucrose solutions of high concentration) not only had a higher thorax temperature but also increased the accuracy of thorax temperature regulation. The general activity of the young bees was, like in the adult bees, not strictly correlated with the energy turnover. Below 20 °C, both the young bees’ activity and VO2 decreased. Between 20 and 40 °C, however, VO2 increased, whereas the bees’ activity showed a minimum between 30 and 35 °C (see mean curves in Fig. 3). This minimum in activity corresponds with the ambient temperatures the young bees prefer (Heran, 1952; Crailsheim et al., 1999a) and which they experience naturally inside a colony (Ritter, 1982; Fahrenholz et al., 1989; Stabentheiner et al., 1995). This stands in contrast to ants, where running speed increases persistently with ambient (and thus body) temperature (Shapley, 1920; Lighton et al., 1987). There have been several attempts to determine the resting metabolism of honeybees. However, as has

888

A. Stabentheiner et al. / Journal of Insect Physiology 49 (2003) 881–889

already been pointed out by Kosmin et al. (1932) and Rothe and Nachtigall (1989), to be sure to determine the true resting metabolism of insects, it is absolutely necessary to record their activity. Therefore, we recorded the bees’ activity and ensured that they were not in an endothermic state by using young bees which were not yet capable of endothermy. At 10 and 15 °C, we measured 0.38 and 0.21 µl min⫺1, which is less than that reported by Lighton and Lovegrove (1990), who only observed a weak increase of VCO2 in this temperature range (~0.62 and ~0.65 µl min⫺1, respectively). Kosmin et al. (1932) reported a resting oxygen consumption of 0.4 and 0.9 µl min⫺1 at 11 and 18 °C, which is higher than in our bees (~0.23 and ~0.67 µl min⫺1, respectively). Allen (1959) determined the VO2 of newly emerged bees in a wide range of ambient temperatures (7–47 °C). Assuming 80 mg bees, she determined a somewhat steeper increase with temperature than in the present study (~0.21 and ~2.54 µl O2 min⫺1 at 15 and 35 °C ambient temperature, respectively, vs. 0.38 and 2.09 µl O2 min⫺1 in the present study, Fig. 3). In the temperature range of 10–30 °C, our young bees’ resting metabolism (Fig. 3a) resembles that reported for adult bees by Rothe and Nachtigall (1989), who averaged all measurements which were within a range of five times the minimum value. In this temperature range, there seems to be no principal difference between young and adult bees. At 35 °C, however, Rothe and Nachtigall (1989) reported a higher value, which resembled that of our active bees (activity level By in Fig. 3a). At temperatures above 30 °C, therefore, a reinvestigation of the respiration of adult bees with simultaneous control of activity and body temperature will be needed to completely answer the question of their true resting metabolism. Interpolation of data of the resting metabolism by the function VO2 = 0.15096 × 1.0778Ta provided an interpolated van’t Hoff Q10 value of 2.12, i.e. VO2 was more than doubled as ambient (and thus body) temperature increased by 10 °C. This is similar to the increase of the mitochondrial activity of a thorax homogenate, where a Q10 of 1.9 was determined between 10 and 45 °C (Leonhard and Crailsheim, 1999). However, while at the mitochondrial level the data fitted the theoretical curve quite well (Leonhard and Crailsheim, 1999), there were considerable deviations from this interpolated curve in the resting honeybees, with a plateau between 20 and 30 °C (Fig. 3a; see also Rothe and Nachtigall, 1989). Accordingly, the Q10 value varied widely and in a Ushaped manner (Fig. 4). This U-shaped curve of the Q10 value is not restricted to honeybees. It was also reported in Pogonomyrmex rugosus (Lighton and Bartholomew, 1988) and Camponotus fulvopilosus (Lighton, 1989) ants and other insects (see Heatwole, 1996). This suggests that it has a general physiological origin (Heatwole, 1996), which, in the case of the honeybee, is unknown.

As a consequence, attempts to describe the temperature dependence of the resting metabolism of honeybees by Q10 values can lead to considerable errors if the measurements are performed at only two temperatures. Acknowledgements This work was supported by the Austrian Fonds zur Fo¨ rderung der Wissenschaftlichen Forschung (FWF) and the Austrian Bundesministerium fu¨ r Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft. We thank all the students of the “Tierphysiologisches Seminar” for their help with oxygen measurements and observation of the bees, Norbert Hrassnigg for help with bee handling, Andrea Loidl for help with data evaluation, Gabriel Stabentheiner for electronics and software support, and two anonymous reviewers for useful hints. References Abrol, D.P., 1992. Oxygen consumption of foraging bees. Journal of Animal Morphology and Physiology 39, 27–42. Allen, M.D., 1959. Respiration rates of worker honeybees of different ages and at different temperatures. Journal of Experimental Biology 36, 92–101. Blatt, J., Roces, F., 2001. Haemolymph sugar levels in foraging honeybees (Apis mellifera carnica): dependence on metabolic rate and in vivo measurement of maximal rates of trehalose synthesis. Journal of Experimental Biology 204, 2709–2716. Cahill, K., Lustick, S., 1976. Oxygen consumption and thermoregulation in Apis mellifera workers and drones. Comparative Biochemistry and Physiology 55A, 355–357. Crailsheim, K., Eggenreich, U., Ressi, R., Szolderits, M.J., 1999a. Temperature preference of honeybee drones (Hymenoptera: Apidae). Entomologia Generalis 24, 37–47. Crailsheim, K., Stabentheiner, A., Hrassnigg, N., Leonhard, B., 1999b. Oxygen consumption at different activity levels and ambient temperatures in isolated honeybees (Hymenoptera: Apidae). Entomologia Generalis 24, 1–12. Dyer, C.D., Seeley, T.D., 1987. Interspecific comparison of endothermy in honeybees (Apis): deviations from the expected sizerelated patterns. Journal of Experimental Biology 127, 1–26. ¨ ber die Ko¨ rpertemperaturen und den Wa¨ rmehauEsch, H., 1960. U shault von Apis mellifica. Zeitschrift fu¨ r Vergleichende Physiologie 43, 305–335. Esch, H., Goller, F., Burns, J.E., 1994. Honeybee waggle dances: the “energy hypothesis” and thermoregulatory behavior of foragers. Journal of Comparative Physiology B 163, 621–625. Fahrenholz, L., Lamprecht, I., Schricker, B., 1989. Thermal investigations of a honey bee colony: thermoregulation of the hive during summer and winter and heat production of members of different bee castes. Journal of Comparative Physiology B 159, 551–560. Fahrenholz, L., Lamprecht, L., Schricker, B., 1992. Calorimetric investigations of the different castes of honey bees, Apis mellifera carnica. Journal of Comparative Physiology B 162, 119–130. Goller, F., Esch, H., 1991. Oxygen consumption and flight muscle activity during heating in workers and drones of Apis mellifera. Journal of Comparative Physiology B 161, 61–67. Harrison, J.F., Fewell, J.H., Roberts, S.P., Hall, H.G., 1996. Achievement of thermal stability by varying metabolic heat production in flying honeybees. Science 274, 88–90.

A. Stabentheiner et al. / Journal of Insect Physiology 49 (2003) 881–889

Harrison, J.M., 1987. Roles of individual honeybee workers and drones in colonial thermogenesis. Journal of Experimental Biology 129, 53–61. Heatwole, H., 1996. Energetics of Desert Invertebrates. Springer, Berlin, Heidelberg. Heinrich, B., 1979. Thermoregulation of African and European honeybees during foraging, attack, and hive exits and returns. Journal of Experimental Biology 80, 217–229. Heinrich, B., 1981. The mechanisms and energetics of honeybee swarm temperature regulation. Journal of Experimental Biology 91, 25–55. Heinrich, B., 1993. The Hot-Blooded Insects. Strategies and Mechanisms of Thermoregulation. Springer, Berlin. Heran, H., 1952. Untersuchungen u¨ ber den Temperatursinn der Honigbiene (Apis mellifica) unter besonderer Beru¨ cksichtigung der Wahrnehmung strahlender Wa¨ rme. Zeitschrift fu¨ r Vergleichende Physiologie 34, 179–206. Koeniger, N., 1978. Das Wa¨ rmen der Brut bei der Honigbiene (Apis mellifera L.). Apidologie 9, 305–320. Kosmin, N.P., Alpatov, W.W., Resnitschenko, M.S., 1932. Zur Kenntnis des Gaswechsels und des Energieverbrauches der Biene in Beziehung zu deren Aktivita¨ t. Zeitschrift fu¨ r Vergleichende Physiologie 17, 408–422. Kovac, H., Schmaranzer, S., 1996. Thermoregulation of honeybees (Apis mellifera) foraging in spring and summer at different plants. Journal of Insect Physiology 42, 1071–1076. Kronenberg, F., Heller, C., 1982. Colonial thermoregulation in honey bees (Apis mellifera). Journal of Comparative Physiology 148, 65–76. Leonhard, B., Crailsheim, K., 1999. Temperature dependency of the oxygen consumption of a honeybee thorax homogenate (Hymenoptera: Apidae). Entomologia Generalis 24, 31–36. Lighton, J.R.B., 1989. Individual and whole-colony respiration in an African formicine ant. Functional Ecology 3, 523–530. Lighton, J.R.B., Bartholomew, G.A., 1988. Standard energy metabolism of a desert harvester ant, Pogonomyrmex rugosus: effects of temperature, body mass, group size, and humidity. Proceedings of the National Academy of Science 85, 4765–4769. Lighton, J.R.B., Lovegrove, B.G., 1990. A temperature-induced switch from diffusive to convective ventilation in the honeybee. Journal of Experimental Biology 154, 509–516. Lighton, J.R.B., Bartholomew, G.A., Feener, D.H. Jr., 1987. Energetics of locomotion and load carriage and a model of the energy cost of foraging of the leaf cutting ant Atta colombica Guer. Physiological Zoology 60, 524–537. Moffatt, L., 2001. Metabolic rate and thermal stability during honey bee foraging at different reward rates. Journal of Experimental Biology 204, 759–766. Ritter, W., 1982. Experimenteller Beitrag zur Thermoregulation des Bienenvolkes (Apis mellifera L.). Apidologie 13, 169–195. Roberts, S.P., Harrison, J.F., 1999. Mechanisms of thermal stability during flight in the honeybee Apis mellifera. Journal of Experimental Biology 202, 1523–1533. Rothe, U., Nachtigall, W., 1989. Flight of the honeybee, IV. Respir-

889

atory quotients and metabolic rates during sitting, walking and flying. Journal of Comparative Physiology B 158, 739–749. Schmaranzer, S., 2000. Body temperature of water collecting honeybees. Journal of Insect Physiology 46, 1187–1194. Schmaranzer, S., Stabentheiner, A., 1987. Das Temperaturverhalten der Honigbiene (Apis mellifera carnica) bei Verschiedenen Aktivita¨ ten als Bioindikator fu¨ r Subletale Wirkungen Verabreichter Pestizide. Institut fu¨ r Zoologie, Universita¨ t Graz, Graz, Austria. Schmaranzer, S., Stabentheiner, A., 1988. Variability of the thermal behavior of honeybees on a feeding place. Journal of Comparative Physiology B 158, 135–141. Shapley, H., 1920. Thermokinetics of Liometopum apiculatum Mayr. Proceedings of the National Academy of Science 6, 204–211. Southwick, E.E., 1985. Allometric relations, metabolism and heat conductance in clusters of honey bees at cool temperatures. Journal of Comparative Physiology B 156, 143–149. Southwick, E.E., 1991. The colony as a thermoregulating superorganism. In: Goodman, L.J., Fisher, R.C. (Eds.), The Behaviour and Physiology of Bees. CAB International, Wallingford, UK, pp. 28–47. Stabentheiner, A., 2001. Thermoregulation of dancing bees: thoracic temperature of pollen and nectar foragers in relation to profitability of foraging and colony need. Journal of Insect Physiology 47, 385–392. Stabentheiner, A., Crailsheim, K., 1999. The effect of activity level and ambient temperature on thermoregulation in isolated honeybees (Hymenoptera: Apidae). Entomologia Generalis 24, 13–21. Stabentheiner, A., Kovac, H., 2002. Contribution of worker bees of different age to active heat production in the brood nest of honeybee colonies. Apidologie 33, 499–500. Stabentheiner, A., Schmaranzer, S., 1987. Thermographic determination of body temperatures in honey bees and hornets: calibration and applications. Thermology 2, 563–572. Stabentheiner, A., Schmaranzer, S., 1988. Flight-related thermobiological investigations of honeybees (Apis mellifera carnica). In: Nachtigall, W. (Ed.), BIONA—Report 6. Akademie der Wissenschaften und Literatur zu Mainz, Gustav Fischer, Stuttgart, New York, pp. 89–102. Stabentheiner, A., Kovac, H., Hagmu¨ ller, K., 1995. Thermal behavior of round and wagtail dancing honeybees. Journal of Comparative Physiology B 165, 433–444. Stabentheiner, A., Kovac, H., Schmaranzer, S., 2002. Honeybee nestmate recognition: the thermal behaviour of guards and their examinees. Journal of Experimental Biology 205, 2637–2642. Stabentheiner, A., Pressl, H., Papst, Th., Hrassnigg, N., Crailsheim, K., 2003. Endothermic heat production in honeybee winter clusters. Journal of Experimental Biology 206, 353–358. Waddington, K., 1990. Foraging profits and thoracic temperature in honey bees (Apis mellifera). Journal of Comparative Physiology B 160, 325–329. Wolf, Th.J., Schmid-Hempel, P., Ellington, C.P., Stevenson, R.D., 1989. Physiological correlates of foraging efforts in honey bees: oxygen consumption and nectar load. Functional Ecology 3, 417–424.