ON THE STUDY OF ANIMAL INTELLIGENCE. 179 12. There is also a special point in which our interpreta- tions of the feelings which underlie the activities of animals, are far more liable to error than our interpretations in the case of other men savages, for example. It is that what may be termed the ratio of the senses is so different. All our knowledge of sensible things is acquired through, and built up out of, sensible impressions, however much they may be impressed with our humanity and our personality. A material difference in the ratio of the senses must, we may suppose, make a material difference in the nature of the mental product. Compare, e.g., man with his delicate sense of touch, the deer with its marvellously acute sense of smell, the eagle with its intense vision, the ant with unknown antennary senses. I speak of known objective differences, as observed in, or inferred from, the study of habit ; and I submit that, if we can infer anything of the underlying mental states, these known objective differences must be accompanied by unknown psychical differences. 13. The element of pain also presents a point of special difficulty. We have abundant evidence how strong a tendency there is at the present time to credit animals with a sensibility equal to, or greater than, our own in this respect. This seems at first sight justified by the cries of a wounded animal, and the howls of a dog under the horse- whip. But there are many facts which point in an opposite direction. We read, for example, that " A post-horse came down on the road with such violence, that the skin and sinews of both the fore fetlock joints were so cut that on its getting up again the bones came through the skin, and the two feet turned up at the back of the legs, the horse walking upon the ends of its leg bones. The horse was put into a field close by, and the next morning it was found quietly feeding about the field, with the feet and skin forced some distance up the leg- bones ; and, where it had been walking about, the holes made in the ground by the leg-bones were three or four inches deep " (G. A. Rowell, Essay on the Beneficent Distribution of the Sense of Pain, quoted by E. D. Girdlestone, Vivisection, p. 18). I submit that in this matter we are not in a position to draw inferences sufficiently exact to be of value for scientific purposes. 14. We thus see how profoundly liable to error are our ejective inferences in Animal Psychology. Let us now try and assign to them their true place and value in the study of Animal Intelligence. We clearly cannot afford to neglect them altogether. And yet it is difficult to see how we can hope to frame a science of Comparative Psychology out of such materials. Could we frame a science of Astronomy if the only method of procedure were to observe the stars and
