THE PERCEPTION OF SPACE. 147 



of each other and in definite order they must ajopear as parts in 

 a vaster sensible extent ivhich can enter the mind simply and all 

 at once. I think it will be seen that the difficulty of esti- 

 mating correctly the form of one's body by pure feeling 

 arises from the fact that it is very hard to feel its totality as 

 a unit at all. The trouble is similar to that of thinking for- 

 wards and backwards simultaneously. When conscious of 

 our head we tend to grow unconscious of our feet, and there 

 enters thus an element of time-succession into our percep- 

 tion of ourselves which transforms the latter from an act of 

 intuition to one of construction. This element of con- 

 structiveness is present in a still higher degree, and carries 

 with it the same consequences, when we deal with objective 

 spaces too great to be grasped by a single look. The rela- 

 tive positions of the shops in a town, separated by many 

 tortuous streets, have to be thus constructed from data ap- 

 prehended in succession, and the result is a greater or less 

 degree of vagueness. 



That a sensation he discriminated as a part from out of a 

 larger enveloping space is then the conditio sine qua nan of its 

 being apprehended in a definite spatial order. The problem 

 of ordering our feelings in space is then, in the first instance, 

 a problem of discrimination, but not of discrimination pure 

 and simple ; for then not only coexistent sights but coex- 

 istent sounds would necessarily assume such order, which 

 they notoriously do not. Whatever is discriminated will 

 appear as a small space within a larger space, it is true, but 

 this is but the very rudiment of order. For the location of 

 it within that space to become precise, other conditions still 

 must supervene ; and the best way to study what they are 

 will be to pause for a little and analyze lohat the expression 

 * spatial order' means. 



Spatial order is an abstract term. The concrete percep- 

 tions which it covers are figures, directions, positions, mag- 

 nitudes, and distances. To single out any one of these 

 things from a total vastness is partially to introduce order 

 into the vastness. To subdivide the vastness into a multi- 

 tude of these things is to apprehend it in a completely 

 orderly way. Now what are these things severally ? To 



