_Annornia arcens._--This is the so-called 'driver' or 'marching' ant of West Africa, which in habits and intelligence closely resembles the military ants of the other hemisphere. I shall therefore not wait again to describe these habits in detail. Like the Ecitons, the marching ants of Africa have no fixed nest, but make temporary halts in the shade of hollow trees, overhanging rocks, &c. They march in large armies, and, like the Ecitons, always in the form of a long close column; but in this case the relative position of the carriers of spoil and larvæ is reversed, for while these occupy the middle place the soldiers and officers march on either side. These have large heads armed with powerful jaws, and never take part in carrying; their function is to maintain order, act as scouts, and attack prey. The habits of these ants resemble most closely those of the blind Ecitons in that they very frequently, and indeed generally, build covered ways; they do so apparently in order to protect themselves from the heat of the African sun. Their line of march is therefore marked by a continuous arch or tunnel, which is always being constructed by the van of the column. The structure is made of earth moulded together by saliva, and is very quickly built. But it is only built in places where the line of march is exposed to the sunlight; at night, or in the shadow of trees or long grass, it is not made. If their camp is flooded by a tropical rainstorm, the ants congregate in a close mass, with the younger ants in the centre; they thus form a floating island. It is remarkable that ants of different hemispheres should manifest so close a similarity with respect to all these wonderful habits. The Chasseur ants of Trinidad, and, according to Madame Merian, the ants of visitation of Cayenne, also display habits of the same kind. _General Intelligence of Various Species._ Many of the foregoing facts display an astonishing degree of intelligence as obtaining among ants; for I think that however much latitude we may be inclined to allow to 'blind instinct' in the way of imitating actions elsewhere due to conscious purpose, some at least of these foregoing facts can only be fairly reconciled with the view that the insects know what they are doing and why they are doing it. But as I am myself well aware of the difficulty that arises in all such cases of drawing the line between purposeless instinct and purposive intelligence, I have thought it desirable to reserve for this concluding division of the present chapter several isolated facts which have been observed among sundry species of ants, and which do not seem to admit of being reasonably comprised under the category of instinctive action, if by the latter we mean action pursued without knowledge of the relation between the means adopted and the ends attained.