THE ASSOCIATION OF IDEAS 51 



In the first place: in a series of activities of the muscles, the 

 successive reactions would not have to wait for their initiation, 

 on the completion of the preceding reactions, but could be initi- 

 ated from the shorter cerebellar transit. A great saving, espe- 

 cially in a rapid series, would thus be accomplished, although the 

 afferent current from the completed reactions would always serve 

 as a check on the series, in case of its going wrong or suddenly 

 needing to be modified. 



In the second place : in a series of associated ideas, the muscular 

 activities would not be needed at all, after the series had been 

 efficiently formed; the series of discharges between the cerebellum 

 and the cerebrum being entirely sufficient, since they integrate, 

 and are integrated with, the total neural activity as thoroughly 

 as are the primitive reactions involving the muscles. For con- 

 sciousness therefore, the cerebro-cerebellar system suffices (after 

 it has once been established by reactions), except in so far as 

 further modifications are necessary in the series, or as a slight 

 renewal of the learning process may be necessary to offset for- 

 getting: little further modification of the series being possible after 

 it has been handed over to the shorter system. 14 We should 

 expect to find, however, vestiges of the original muscular activi- 



14 Excepting of course deterioriations. As is well known, all series tend to 

 fall to pieces in time, although certain series resist the tendency to a surprising 

 extent. The extent to which modifications in old series, or the formation of new 

 series, may occur without muscular intervention if such modification or forma- 

 tion be possible at all is an important field for experimental work. The phe- 

 nomena of temporary memory, as in the transcribing of words or figures offers a 

 good starting point. In listing nine-place numbers on an adding machine, where 

 the numbers are read from a sheet at the left of the machine, errors are excluded 

 if the numbers are distinctly articulated as read : without distinct articulation it 

 is apparently impossible for the ordinary operator to avoid errors unless he refers 

 back to the sheet once or more during the listing of a number. The question here 

 is whether actual articulation is not essential in all such cases, even for two place 

 numbers; or whether a certain (and manifestly inferior) association is possible 

 without articulation. In such cases, the possibility of finger innervations func- 

 tioning instead of articulation ; and in the case of expert operators, functioning 

 better; must not be overlooked. Serious fallacies result from the consideration 

 of vocal movements as the sole motor components in thought-processes : the 

 whole striated musculature has important thought-functions, although the vocal 

 muscles have, in intelligent adults, the leading r61e. 



