It appears that the African bee Apis mellifera might assist in the increasingly important task of protecting African smallholdings from elephant damage. Although thick-skinned, African elephants Loxodonta africana have thin trunk membranes, ears and eyes where a bee sting can have serious effect. To test the idea that bees can deter elephants we used local log hives to \'mine\' a favourite elephant foraging area of fever tree Acacia xanthophloea re-growth on Mpala Ranch in the Laikipia Plateau, Kenya.
Towards the end of the dry season we hung 30 unoccupied (but seasoned) hives and six occupied hives 1.5-2 m high in trees of the 3-5 m size range. Nearby hiveless trees acted as controls. Over the 40 day experiment, we found that the 36 control trees (no hives) on our site experienced serious elephant damage (only three were left undamaged), the 30 experimental trees with unoccupied hives experienced some damage (12 left alone) while all six trees with occupied hives were spared. This suggests strongly that the elephants sensed and avoided the hives and their immediate surroundings, taking special care with occupied hives.
The sense of smell is paramount in elephant social and foraging decisions and it is possible that smell may have been the crucial cue. Elephants also have excellent hearing and sound might therefore have been another factor in hive avoidance. To test this idea we played a tape of the sound of bees humming to a tame bull who a few years previously had been stung badly. He backed away from a tree playing the \'bee-sound\' but ignored a tree playing a Bach violin concerto. Whatever the mechanism, it seems that bees provide some protection against elephants. Hence we propose that hives might be used to protect smallholdings by placing them strategically where they could be upset easily by browsing elephants. Using bees as a selective deterrent would more than pay for itself through sales of honey, a product of the natural environment. Controlling elephants with bees is preferable to lethal measures and future large-scale experiments could determine the feasibility of this concept.
Fritz Vollrath , University of Oxford, UK and Iain Douglas-Hamilton, Save the Elephant, Kenya