The extent to which interference with one set of images will change the other may be well seen by a simple experiment. Most people, if they attempt a silent recitation while opening the mouth to its fullest stretch or holding the tongue firmly between the teeth, will notice curious transformations in the auditory images. How the experiment should be interpreted is uncertain, but it is of use in making the presence of both kinds of verbal imagery evident to those who may have overlooked them hitherto. Images of articulation should not, however, be confused with those minimal actual movements which for some people (for all, as behaviourists maintain) accompany the silent rehearsing of words. These two forms of tied imagery might also be called verbal images, and supply the elements of what is called the ‘formal structure’ of poetry. They differ from those to which we now proceed in being images of words, not of things words stand for, and in their very close connection with the visual sensations of printed words. _Free Imagery_.—Free images, or rather one form of these, visual images, pictures in the mind’s eye, occupy a prominent place in the literature of criticism, to the neglect somewhat of other forms of imagery, since, as was remarked in a preceding chapter, for every possible kind of sensation there is a corresponding possible image. The assumption, natural before investigation, that all attentive and sensitive readers will experience the same images, vitiates most of the historical discussions from that of Longinus to that of Lessing. Even in the present day, when there is no excuse for such ignorance, the mistake still thrives, and an altogether too crude, too hasty, and too superficial form of criticism is allowed to pass unchallenged. It cannot be too clearly recognised that individuals differ not only in the type of imagery which they employ, but still more in the particular images which they produce. In their whole reactions to a poem, or to a single line of it, their free images are the point at which two readings are most likely to differ, and the fact that they differ may very well be quite immaterial. Fifty different readers will experience not one common picture but fifty different pictures. If the value of the poem derived from the value _qua_ picture of the visual image excited then criticism might well despair. Those who would stress this part of the poetic reaction can have but crude views on pictures.