874 
ME, G. J. ROMANES AND PROFESSOR J. C. EWART 
tlieir function of localising and closing round a seat of stimulation are thus shown to 
be completely destroyed by injury of the external plexus, other nervous connexions, 
upon which another function of the spines depends, are not in the smallest degree 
impaired by such injury. This other function is that which brings about the general 
co-ordinated action of all the spines for the purposes of locomotion. That this function 
is not impaired by injury of the external plexus is proved by severely stimulating an 
area within a closed line of injury on the surface of the shell; all the spines over the 
whole surface of the animal then manifest their bristling movements, and by their 
co-ordinated action move the animal in a straight line of escape from the source of 
irritation. 
We have, therefore, to distinguish between what may be called the local reflex 
function of the spines, which they show in common with the pedicellarise and which is 
exclusively dependent upon the external plexus, and what we may call the universal 
reflex function of the spines, which consists in their general co-ordinated action for 
the purposes of locomotion, and which is wholly independent of the external plexus. 
Apparently, therefore, this more universal function must depend upon some other 
set of nervous connexions (which, however, we have not been able to detect histo¬ 
logically), and experiment shows that these, if present, are distributed over all the 
internal surface of the shell. For if the internal surface be painted with acid, or 
scoured out with emery paper and brick-dust, the spines and pedicellarise, after a 
short period of increased activity or bristling, become perfectly quiescent, lie flat, and 
lose both their spontaneity and irritability. After a few hours, however, the 
spontaneity and irritability of the spines return, though in a feeble degree, and 
also those of the pedicellariae in a more marked degree. These effects take place over 
the whole external surface of the shell, if the whole of the internal surface be painted 
with acid or scoured with brick-dust; but if any part of the external surface be left 
unpainted or unscoured, the corresponding part of the external surface remains un¬ 
injured. From these experiments we conclude :—1st, that the general co-ordination 
of the spines is wholly dependent on the integrity of the hypothetical internal plexus ; 
2nd, that the hypothetical internal plexus is everywhere in intimate connexion with 
the external, apparently through the calcareous substance of the shell; and 3rd, that 
complete destruction of the former, while profoundly influencing through shock the 
functions of the latter, nevertheless does not wholly destroy them. 
Echini may be divided into pieces, and the pedicels, spines, and pedicellariae upon 
these pieces will continue to exhibit their functions of local reflex irritability, however 
small the pieces may be. If an entire double row of pedicels be divided out as a 
segment, and then placed upon its ab-oral end, it may rear itself up on its oral end by 
the successive action of its pedicels, and then proceed to crawl about the floor of the 
tank. We have therefore to meet the question Is the action of the ambulacral feet 
in executing these righting movements of a merely serial kind—A, B, and C, first 
securing their hold of the tank floor, owing to the stimulus supplied by contact, and 
