PHYSIOLOGICAL FACTORS OF THE ATTENTION-PEOCESS. 333 two of the linear groupings can be made to alternate with one another to the almost complete exclusion of the third form. And the grouping can be controlled not only in the absence of eye-movements, but in spite of eye-movements which favour the predominance of one form of grouping only. A fine thread is laid across the figure parallel to one of the three sets of rectilinear rows and the eyes are kept moving to and fro along this thread ; this, of course, greatly favours the predominance of the grouping in rows parallel to the direction of movement of the eyes, but it is possible, after some practice, to call up either of the other linear groupings at will while the eyes continue to make the to-and-fro move- ments in the direction of the third set of rows. This effect also may be very satisfactorily observed on projecting the after-image of fig. 2 upon a smooth grey wall on which a horizontal line has been drawn. Then, while the eyes move to and fro along this line, the oblique rows can be called up at will ; and indeed it is not possible by a voluntary effort aided by such eye-movements to exclude their appearance ; after some seconds, during which the horizontal rows alone appear, one or other of the oblique sets of rows appears involuntarily. These simple observations seem to prove that neither the actual muscular-adjustments of the sense organs, nor the central innervations that produce such adjustments are essential factors in the voluntary or non-voluntary control of the direction and mode of attention to sense impressions ; that they are not the means whereby this control is effected in voluntary attention, nor the cause of the direction of non- voluntary attention, but are rather effects of the direction of attention to this or that object, effects which tend to main- tain and to facilitate that direction of attention. The frequent exaggeration of the importance of the motor- element in attention is probably due to two facts : (1) the fact that all cerebro-ideational process tends to issue in appropriate motor excitement; (2) the fact that voluntary innervation of groups of muscles is a familiar process of unquestionable reality. Those who assume that the motor adjustment is the primary fact in voluntary attention and that voluntary control of attention can only be effected in- directly by voluntary control of muscular innervation admit the power of voluntarily calling up an idea in the case of ideas of movement, but deny it in the case of ideas of every other class. But there is no justification for the assumption of any such difference, and if, as all admit, we can voluntarily and directly re-enforce an idea of movement we should be
