Naturwissenschaften

, Volume 96, Issue 9, pp 1133–1136

Heat and carbon dioxide generated by honeybees jointly act to kill hornets

Authors

  • Michio Sugahara
    • Department of Bioscience and Biotechnology, Faculty of Bioenvironmental ScienceKyoto Gakuen University
    • Department of Bioscience and Biotechnology, Faculty of Bioenvironmental ScienceKyoto Gakuen University
SHORT COMMUNICATION

DOI: 10.1007/s00114-009-0575-0

Cite this article as:
Sugahara, M. & Sakamoto, F. Naturwissenschaften (2009) 96: 1133. doi:10.1007/s00114-009-0575-0

Abstract

We have found that giant hornets (Vespa mandarinia japonica) are killed in less than 10 min when they are trapped in a bee ball created by the Japanese honeybees Apis cerana japonica, but their death cannot be solely accounted for by the elevated temperature in the bee ball. In controlled experiments, hornets can survive for 10 min at the temperature up to 47°C, whereas the temperature inside the bee balls does not rise higher than 45.9°C. We have found here that the CO2 concentration inside the bee ball also reaches a maximum (3.6 ± 0.2%) in the initial 0–5 min phase after bee ball formation. The lethal temperature of the hornet (45–46°C) under conditions of CO2 concentration (3.7 ± 0.44%) produced using human expiratory air is almost the same as that in the bee ball. The lethal temperature of the honeybee is 50–51°C under the same air conditions. We concluded that CO2 produced inside the bee ball by honeybees is a major factor together with the temperature involved in defense against giant hornets.

Keywords

Japanese honeybeeApis cerana japonicaBee ballGiant hornetVespa mandarinia japonica

Introduction

The Japanese honeybee Apis cerana japonica is a wild honeybee species native to Japan. They make nests in urban as well as rural areas (Sugahara 2000) usually in dark and closed spaces such as holes in trees and undersides of roofs. However, about 10% of the nests are unexpectedly constructed in open spaces as open nest (Matsuura 2003). Artificially established open nests are useful for various behavioral studies.

The honeybee is known to fight against the giant hornet Vespa mandarinia japonica by forming a spherical assemblage called a bee ball and trapping and killing them inside the ball. The heat produced within bee balls has been measured. The mechanism of the honeybee's strategy can be ascribed to their heat tolerance greater than that of the giant hornet (Ono et al. 1995).

We have observed that the giant hornet is not killed in an incubator at 47°C corresponding to the maximal temperature in bee balls (Ono et al. 1995). However, all hornets trapped in bee balls are killed after 10 min, when the mass movement of bees of bee ball is calmed down. We therefore reinvestigated the biology of bee ball formation as a defense mechanism against the giant hornet.

Materials and methods

The Japanese honeybees of 2-year-old open nest were used as the test stage for bee ball formation. The nest was prepared as follows: a swarm of honeybees was captured in Hirakata City and kept in a special Langstroth hive box composed of a fixed top board and removable bottom and lateral boards. No hive boards were introduced. The hive box was hung under eaves. When the nest had sufficiently developed, all bottom and lateral boards were removed to create an open nest (Sugahara and Kondou 2005). Foraging workers of the giant hornet were prepared by collecting from natural nests around the Yodo river bank in Moriguchi City. All hornets and honeybees were handled after being anesthetized with CO2 and tested as described.

A hornet was anesthetized, then fixed to the tip of a thermometer probe by scotch tape; 5 min later when it was revived to fully recover from the anesthesia, the tip of a thermometer probe with a fixed hornet was touched to the open nest. A bee ball was immediately formed. The temperature inside the bee ball (around the giant hornet) was recorded using a digital thermometer (Yokogawa Model 2455) coupled with a digital video recorder (SONY DCR, TR V20). After 10 min, the bee ball attached to the probe was removed from the nest, and the remaining bees around the trapped hornet were effectively dispersed by spraying Skin Guard (Johnson Co. LTD). The survival/death of trapped hornets was based on the following criteria: alive = presence of a response to contact stimulation and movement 30 min later, dead = protrusion of the sting, no response to contact stimulation, and no movement 30 min later.

A portable gas detector (COSMOS XP-3140) was used to measure the CO2 concentration in open-nest bee balls. As the gas inlet was large (5 mm, id), two hornets were fixed to the inlet tip (Fig. 2a). Then, the inlet tip was touched to the open nest. A bee ball slightly larger than those formed in temperature measurement (ca. 6 cm in diameter) was produced. The CO2 concentration around the bee ball (sampling air flow at 250 ml/min) was recorded using a digital video recorder. The CO2 level measurements may be underestimates because the CO2 levels inside bee balls may have decreased due to the continuous sampling of gas from the balls.

To determine the 50% lethal temperature (TL50) for 10 min, an incubator (Takasaki Scientific Instruments CORP, TXY-9R-3F) was used. Each giant hornet, kept in a plastic container (100 ml volume) with a cover of punched holes in the case of normal air or in a plastic container sealed with a cover and filled with human expiratory air (300 ml volume), was placed in the constant-temperature unit. After 10 min, their condition (alive/dead) was examined. In the case of normal air, the Japanese honeybee was similarly examined in a small plastic container (20 ml volume) with small holes. In the case of human expiratory air, the bees were put directly into a plastic bag, equipped with a small battery-powered fan, filled with human expiratory air, and the plastic bag was placed in the incubator. The honeybee's condition (alive/dead) was examined in the same way as for hornets. The plastic bag made it easy to handle the specimens and control the temperature, but in the case of the giant hornet, it could pierce and escape.

Results

Observation of giant hornets trapped in bee balls

A total of 24 giant hornets were experimentally trapped for 10 min in bee balls in open nests and seven in closed nests. All hornets recovered were dead, showing protruded stings and not responding to any contact stimulation. As additional cases, four hornets were experimentally removed from open-nest bee balls within 4 min of ball formation. Among them, three were still alive, although in a critical condition. There were no stings or sting marks on the cuticles of dead hornets.

Measurement of temperature within bee balls

Temperature change at the center of the bee ball (around the trapped hornet) was traced in the open nest (Fig. 1b), being composed of two phases. The first phase is the period in which a plateau is reached following a rapid rise (0–5 min). The second is the latter half (5–10 min) of the observation period, where the temperature is generally maintained but shows a slight decreasing tendency. The maximum temperature is observed to be 45.9 ± 1.0°C (n = 15). The heat generated in the bee ball tended to be lower when the external temperature was low. The range of ambient temperatures was ca. 18–33°C.
https://static-content.springer.com/image/art%3A10.1007%2Fs00114-009-0575-0/MediaObjects/114_2009_575_Fig1_HTML.gif
Fig. 1

a The temperature within bee balls was measured in an open nest. b Typical changes in temperature at the center of the open bee ball (mean ± SD, n = 15)

CO2 concentration within bee balls

The CO2 level at the center of the bee ball was monitored at 10-s interval in two cases (Fig. 2c). The maximal concentration was observed to be 3.6 ± 0.2% (n = 4) at ca. 4 min after the start. Although the observed data shows marked variations among each trial, the CO2 level can be separated into two phases: 0–5 and 5–10 min (Fig. 2c). The CO2 level in a normal bee space of the open nest was measured as 0.7%.
https://static-content.springer.com/image/art%3A10.1007%2Fs00114-009-0575-0/MediaObjects/114_2009_575_Fig2_HTML.gif
Fig. 2

a Two hornets were introduced into the tip of the gas detector. b The CO2 level was measured within the bee ball beneath the open nest. c CO2 levels inside bee balls (two cases) at 10-s intervals. Features during the initial 5 min and the latter half changed markedly

Measurement of the 50% lethal temperature on a 10-min exposure

The lethal temperature on a 10-min exposure was examined, and the results are summarized in Table 1. The TL50 of the hornet in normal air was 47–48°C, whereas in human expiratory air (CO2 level, 3.7 ± 0.44%, n = 5), it was 45–46°C. Thus, the presence of CO2 leads to a 2°C decrease in the TL50. On the other hand, the TL50 of honeybees in normal air was 50–51°C, and that in human expiratory air was also 50–51°C.
Table 1

The mortality rate (MR) of V. mandarinia japonica and A. cerana japonica after 10 min

Temperature (°C)

43

44

45

46

47

48

49

50

51

52

53

V. mandarinia japonica

MR in normal air (%)

0

0

0

0

100

100

Tested number

0

2

2

5

6

6

6

0

0

0

0

MR in human expiratory air (%)

0

17

33

67

100

100

Tested number

2

6

9

9

3

3

0

0

0

0

0

A. cerana japonica

MR in normal air (%)

0

0

40

73

90

100

Tested number

0

0

0

0

0

10

10

10

15

10

10

MR in human expiratory air (%)

0

10

50

60

100

100

Tested number

0

0

0

0

0

10

10

10

15

10

10

When the exposure period is increased to 30 min, the TL50 of the hornet lowers to 45–46°C, while that of the honeybee becomes 48–49°C, showing a discrepancy of 3°C. Generally, it is known that the lethal temperature falls when the period of time maintained at that temperature lengthens (Schmidt-Nielsen 1997). These values are almost the same as those reported by Ono et al. 1995 (V. mandarinia japonica, 44–46°C; A. cerana japonica, 48–50°C).

The humidity of human expiratory air at the tested temperature (43–53°C) was around 30%. We controlled the moisture level in the incubator between 30% and 45% by the presence of water, so it was considered that the influence of moisture was excluded (Wigglesworth 1972).

Discussion

The giant hornets are killed in 10 min following bee ball formation by honeybees, and the maximum temperature inside bee balls is 45.9 ± 1.0°C. In controlled experiments, hornets can survive for 10 min at constant temperatures of up to 47°C. The CO2 concentration inside the bee ball rises to 3.6 ± 0.2% in the initial phase after bee ball formation, and under such air conditions, the lethal temperature of the hornet decreases to 45–46°C which is almost the same as that in the bee ball. So, we concluded that CO2 produced inside the bee ball by honeybees is a major factor together with temperature involved in defense against the giant hornet.

Both the observed temperature traces and CO2 emissions of bee balls comprised two phases: 0–5 and 5–10 min. Within the first full 10 min, all giant hornets were killed, whereas three out of four were found alive when recovered experimentally from the bee ball after 4 min. These facts indicate that the hornet may be killed during the first 0–5 min period, in which the highest level of heat production and CO2 emissions take place in honeybees, possibly consuming hydrocarbons in their body. This apparent separation into two phases may suggest that members of the bee ball are aware of the physiological condition of their entrapped victim. The latter 5–10 min period may be free running to ensure their victim's death.

Ono et al. reported that a hornet is killed by heat in bee balls in closed nests. We observed that all hornets captured in such bee balls died after 10 min (n = 7), as was observed in open-nest bee balls. It is considered that even if the maximal temperature of the bee balls in closed nests (47°C, Ono et al. 1995) was maintained for 10 min, it would be difficult to kill all hornets simply through heat.

It has been reported that bee balls are formed as a defense against Vespa simillima (Ono et al. 1987), Vespa velutina (Ken et al. 2005), Vespa magnifica (Abrol 2006) and Vespa multimaculata (Koeniger et al. 1996). By measuring at what point they died, the TL50 over the period of time required until death and O2 and CO2 concentrations in bee balls within those species, a unified analysis would become possible.

Apis mellifera cypria reportedly mobs and smothers hornets to death in bee balls (Papachristoforou et al. 2007). This is based on the result that when the opening of the spiracle in their abdomen was secured, the time required until the hornet's death was prolonged. It is understandable that Cyprian honeybees form bee balls in a different way from Asian honeybees.

The TL50 of the Japanese honeybee in human expiratory air was the same as that in normal air. It is interesting why the TL50 for 10 min in human expiratory air decreased in the case of the giant hornet, whereas the honeybee tolerated such a condition. This may be related to the fact that the CO2 level in the bee nest rose to 6% (Nicolas and Sillans 1989).

Because of the small sample sizes in the lethal temperature measurement of the giant hornet in human expiratory air, the lethal temperature may change slightly, but we think that the CO2 emission in the bee ball is a powerful weapon of honeybees against the giant hornet.

Acknowledgements

We are grateful to professor Kuwahara of Kyoto Gakuen University and Drs. Nakahari, Asai and Okazaki of Osaka Medical College, who made useful suggestions for the writing of this paper.

Copyright information

© Springer-Verlag 2009