Through the sense organs our mind receives information about the external world. The traditional classification of the sensations divided them into five groups. But the distinction of five senses has been found to be insufficient. At least twice as many must be distinguished.
When psychologists tried to explain all human knowledge in terms of experience, they met with some difficulty in the description of our experience of solid bodies. Tactual sensation was found to be insufficient for this explanation, since it informs us only of the side-by-side position of things, that is, of only two dimensions. It was soon recognized that the movements of our limbs were important factors in this experience, and the question was asked: How do we perceive the spatial relations of our limbs and the resistances offered to changes in these spatial relations, that is, to movements? The first answer to this question was, that the muscles, being obviously a kind of sense organ which gives us the familiar sensations of fatigue and muscular pain, are also capable of sending in definite groups of afferent nervous processes according to their conditions of contraction and tension. This answer was quite true, as far as it went; and about 1870 the sensory neurons of muscles were actually discovered. The tendons connecting the muscles with the bones were also found to contain sensory neurons.
But this cannot be all, for we are able to judge the position of our limbs even when the muscles are completely relaxed and a limb is moved by another person. It is further a fact that a weight and the distance through which it is moved can be estimated with fair accuracy, whether the arm is sharply bent or straightened out, although the contraction and tension of the muscles is very different in these two cases. It is now known with some certainty how these estimations are made possible. The surfaces of the joints are furnished with nerves. Make a slow movement of the hand or a finger and attend to the sensation resulting from it. There is little doubt that the sensation is localized in the joint. This view is supported by the fact that electrical stimulation of a joint considerably decreases the accuracy of the estimation of weight and movement.