I assume, then, that, granting the theory of evolution, the early stages of the process of construction--discrimination, localization, and outward projection--are the same in kind throughout the whole range of animal life, wherever we are justified in surmising that psychical processes occur, and the power of registration and revival in memory has been established. As will be gathered, however, from what I have already said, I hold that the nature of the constructs produced is and must be for us human-folk, since we are human-folk, to a large extent a matter of speculation. Remembering this, then, endeavouring never to lose sight of it for a moment, let us consider what we may fairly surmise concerning the constructs and the process of construction in animals. * * * * * There can be no question that the animals nearest us in the scale of life--the higher mammalia--form constructs analogous to, if not closely resembling, ours. I do not think the resemblance can be in any sense close, seeing to how large an extent our constructs are literally our _handi_-work. For though in many animals the tongue and lips are delicate organs of touch--not to mention the trunk of the elephant--and though in the monkeys and many rodents the hands are used for grasping, still we have no reason to suppose that in any other mammal the geometrical sense of touch plays so determining a part in the formation of constructs as in man. On the other hand, in the dog and the deer, for example, not only must the marvellously acute sense of smell have a far higher suggestive value, but smells and odours must, one would suppose, be built into the constructs in a far larger proportion. But although their constructs may not closely resemble ours, the constructs of animals may, I believe, be fairly regarded as closely analogous to our own. And as with us, so with them, a comparatively simple and meagre suggestion may give rise, through association in experience, to the construction of a complex object. And again, as with us, so with them, the suggested construct may be very vague and indefinite. A dog, for example, is lying asleep upon the mat, and hears an unfamiliar step in the porch without. There can be no question that this suggests the construct man. But from the very nature of the case, this must be vague and indefinite. So, too, when a chamois, bounding across the snow-fields, stops suddenly when he scents the distant footprints of the mountaineer, the construct that he forms cannot be in any way particularized--no more particularized than is to me the sheep that I hear bleating in the meadow behind yonder wall.