CEREBRAL HEMISPHERES AND MEMORY 257 



the circulation is normal. It harmonises with these 

 facts that mental activity does not influence the 

 phenomena of oxidation, as Speck has proved by very 

 careful experiments. But from this we must not con- 

 clude that the activity of the brain takes place with- 

 out chemical changes, only that the chemical changes 

 which are determined by mental activity are too 

 slight to be recognised. The statement that dilatation 



J O 



of the blood-vessels of the brain produces a sensation 

 of happiness is not based upon any fact that has 

 been proved scientifically. 



7. The amoeboid changes in the ganglion-cells have 

 been utilised to account for the phenomena of asso- 

 ciation. As far as normal processes of association 

 are concerned, these amoeboid changes cannot play 

 any role, as they are much too slow. We notice 

 migrations of the cones and the pigment in the 

 retina, yet the idea that these protoplasmic motions 

 play any role for space- or colour-perception has to 

 be abandoned for the same reason. 



Other authors hold that conditions of incomplete 

 association, as in the case of dreams, or interruption 

 of association, as in the case of deep sleep or 

 narcotics, are due to a partial or complete discon- 

 nection of the ganglion-cells by a shortening of the 

 processes. It does not seem to me that the obser- 

 vations which we thus far possess prove anything of 

 that character (9, 12). 



