THE INSTABILITY OF THE HOMOGENEOUS. 425 



destroy the organic compounds, but only modify them: the 

 parts most exposed to the modifying forces being most modi- 

 fied. To elucidate this, suppose we take a few cases. 



Note first what appear to be exceptions. Certain 

 minute animal forms present us either with no appreciable 

 differentiations or with differentiations so obscure as to be 

 made out with great difficulty. In the Rhizopods, the sub- 

 stance of the jelly-like body remains throughout life unor- 

 ganized, even to the extent of having no limiting mem- 

 brane ; as is proved by the fact that the thread-like processes 

 protruded by the mass coalesce on touching each other. 

 Whether or not the nearly allied Amoeba, of which the less 

 numerous and more bulky processes do not coalesce, has, as 

 lately alleged, something like a cell-wall and a nucleus, it is 

 clear that the distinction of parts is very slight ; since parti- 

 cles of food pass bodily into the inside through any part of 

 the periphery, and since when the creature is crushed to 

 pieces, each piece behaves as the whole did. Now these cases, 

 in which there is either no contrast of structure between ex- 

 terior and interior or very little, though seemingly opposed 

 to the above inference, are really very significant evidences 

 of its truth. For what is the peculiarity of this division of 

 the Protozoa f Its members undergo perpetual and irregu- 

 lar changes of form — they show no persistent relation of 

 parts. What lately formed a portion of the interior is now 

 protruded, and, as a temporary limb, is attached to some 

 object it happens to touch. What is now a part of the sur- 

 face will presently be drawn, along with the atom of nutri- 

 ment sticking to it, into the centre of the mass. Either the 

 relations of inner and outer have no permanent existence, 

 or they are very slightly marked. But by the hypothesis, 

 it is only because of their unlike positions with respect to 

 modifying forces, that the originally like units of a living 

 mass become unlike. We must therefore expect no estab- 

 lished differentiation of parts in creatures which exhibit no 

 established differences of position in their parts; and we 

 29 



