No. 5(34] THE FIXATION OF CHARACTER 723 



undifferentiated, is able, upon necessity, to give rise to all 

 structures that the whole tissue would normally have 

 produced. A sufficiently large group of cells from any 

 portion of the blastula of an echinoderm, for example, will 

 produce a normal larva; but the moment the process of 

 gastrulation begins, this power of producing the whole 

 animal is definitely lost by those very, cells which pos- 

 sessed it but a few hours previously; for, now that differ- 

 entiation has begun to take place, a piece which shall give 

 rise to a normal larva must include a little of both the 

 primitive ectoderm and entoderm and can not be taken at 

 random from anywhere in the embryo. Any portion of 

 the primitive gut, which later develops, is able to produce 

 the c<i'lomic pouch, should the normal region of origin of 

 that structure be removed, but this "equipotency" lasts 

 only so long as there is no differentiation, for if the pouch 

 once begins to develop and then is removed it can never 

 be produced again even by the cells which a short time 

 previously had the power to form it. This process of 

 ontogenetic segregation results in the continual loss of 

 potentialities, in the progressive narrowing down of 

 the possibilities at the command of every living cell. 

 The situation in phytogeny is very similar, for the possi- 

 bilities before a simple, plastic and comparatively undif- 

 ferentiated organism — the lines of evolution along which 

 its descendants may go — are much greater than those 

 before one which is highly developed and sharply special- 

 ized. Increased differentiation is followed so regularly 

 by decreased plasticity, both in phylogeny and ontogeny, 

 as to suggest the possibility of a common cause. 



There is also a similarity between structural fixation 

 and certain psychological phenomena. The performance 

 of an action is always uncertain and variable at first, but 

 constantly tends to become stereotyped and habitual. The 

 simpler types of animal activity are directed by instincts 

 which are comparatively changeable and plastic, but where 

 behavior has become highly specialized and complex, in- 



