348 
Psyche 
[Vol. 89 
the colony some ten to 12 days prior to the 21st. This would corres- 
pond to what Matsuura and Sakagami (1973) describe as the hunt- 
ing phase for V. mandarinia. By the 21st the attack had escalated to 
the slaughter phase where nearly two score of hornets were concen- 
trating upon the now weakened honey bee colony. 
The hornet attack was a campaign of slow attrition for the honey 
bees. Usually two to five hornets would position themselves at the 
colony entrance. They would engage any honey bee entering or 
exiting the hive. The large hornets had no difficulty in seizing the 
bees and would maul them with their strong mandibles. The hornets 
would normally drop the disabled bees to the ground and only 
rarely was a moribund bee observed to be eaten by a hornet. The 
guard hornets would frequently position themselves in the entrance 
passageway with only their abdomens visible to the observer. These 
hornets would engage individual guard bees just inside the entrance 
and after seizing a bee, quickly drag it out and drop it off the 
landing board to the ground. A separate cadre of hornets would 
enter the colony and position themselves on the comb containing 
capped honey. This peripheral comb was without bees which were 
concentrated on the brood comb. Honey scavenging hornets would 
spend long periods of time within the colony uncapping honey stor- 
age cells and engorging themselves on the contents. On the 21st the 
average time spent by a hornet inside the colony was 22.9 ±17.7 min 
(n = 64). Upon emerging from the hive the scavenger hornets were 
frequently antennated by the guard hornets, and an exchange of 
alimentary fluid would usually result. 
Continuous observation at the colony was not conducted on 
December 22. However, a one m 2 piece of plywood was placed 
directly in front of the hive to facilitate an estimate of adult honey 
bee mortality. Between 0930 and 1415 h 119 dead honey bees had 
been deposited on the plywood by guard hornets. At 1420 h the 
colony absconded and within ten minutes had clustered on a small 
shrub ca. 15 m north of the hive. For the remainder of the afternoon 
hornets were observed for the first time exiting the hive with larvae 
and pupae scavenged from the brood nest. At 1930 h on the 22nd we 
reintroduced the swarm cluster and queen back into the hive. 
Observations on the 23rd began at 0715 h. At 0834 h the colony 
