The one evident exception to these tastes is seen in the genus Prodotiscus, which differs in many respects—the bill and plumage especially—from all the other genera of the family. It does not eat bee larvae, and one of the species frequents, it is said, trees of the genus Sterculia when they are in flower.

How then do the normal honey guides procure their favorite food? The only bees in Africa from which the wax could come are the common honeybee, Apis mellifera (represented by a somewhat smaller African race), and the much smaller species of Trigona, which are stingless. The honeybees, we know, nearly always store their sweets in a secure place,—a cavity in a tree, among rocks, or in the ground, where the birds unaided have little chance of stealing them. The nests of Trigona are usually in hollow trees, and are if anything harder to get at. Had the honey guides the strong chisel-shaped beak of their allies, the woodpeckers, they might hew their way through the wood; but, as it is, they are without any tools for use in such a direct attack. Birds of the genus Melignothes, for example, have an exceptionally blunt beak.

Thus all the typical honey guides eat honeycomb and yet are apparently unable to secure it without help. What else can we conclude save that they adopt some method similar to that reported from South Africa, of enlisting the aid of the honey badger? In the forests of western equatorial Africa this mammal is extremely rare or wanting, and therefore we may only guess that squirrels, small carnivores, lemurs, or monkeys, are the creatures with which the honey guides carry on this commensal existence. This seems a bold assumption, but it is the best explanation I can offer of what we know to be their food habits. An alternative would be to suppose that the honey guides simply happen upon the hives after they have been robbed by some other animal. I doubt if they would get as much plunder in this manner as we know they secure.

It seems credible that the partnership began in such a way, but that sooner or later the birds took to accompanying bee-hunting mammals until finally the bird became the leader. The theoretic bearing of observations on the honey badger and its bird guide is now evident. Once the method had been well worked out with certain lower mammals, man would have been admitted into the association as a matter of course.

The breeding habits of the Indicatoridai are bizarre, and similar to those of the parasitic cuckoos of the Old World or of our North American cowbird. Each white egg is deposited in the nest of some other bird, preferably that of a barbet, which is hewn out like a woodpecker's hole. Where possible, the adult honey guide breaks the legitimate eggs, it is said, and when the young is found, it is always the sole occupant of the nest. In the two species which lead men to hives, Indicator indicator and Indicator variegatus, Dr. Alwin Haagner [Journ. S. Afr. Orn. Union, Vol. III, 1907, p. 3, Pl. 1] has found that the nestling has both upper and lower mandibles armed with a sharp, curved hook, as though for seizing any competitors and ejecting them from the nest it has usurped. It is said that these hooks are shed at about the time of leaving the nest. In two other members of the family, Melignothes conirostris and Melichneutes robustus, I have examined skins of nestlings partly fledged, but found only the usual egg-tooth on the upper mandible.