OA' PHYSIOLOGICAL MORrHOLOGY. 57 



towards and at last over the blood vessel and form a sheath 

 around it, while the gaps between the blood vessels become 

 empty of chromatophores. In this way the chromatophores 

 are arranged in stripes, and changes in the surface tension, and 

 not a preformed arrangement of the germ, determine the 

 marking. We do not know what processes determine the 

 coloration of animals which owe their markings to interference 

 colors, but the task of deriving such a coloration in the adult 

 from a similar arrangement of molecules in the germ plasm 

 would prove too much even for a genius like Huyghens, and 

 without the possibility of such a derivation the theory is of 

 no use. 



3. The reasons why roots grow on the under side of the 

 stem of Antennularia and stems on the upper side, can only be 

 given when the special physical and chemical conditions inside 

 the stem of Antennularia have been worked out. At present 

 we can only think of possibilities. It is possible that the 

 hypothetical root substances of Sachs may have a greater 

 specific gravity than the substances which form stems, and 

 therefore take the lowest position in the cell. Since outgrowth 

 can take place only at the free surface of a stem or branch, 

 roots could grow only at the under side and stems only at the 

 upper side of an element. But there are still other possibilities 

 which we must omit here. In the case of Margelis and other 

 hydroids, it might happen that contact with solid bodies 

 produced an increase of surface in the touched elements in 

 case they contained specific root substances, while the opposite 

 took place in the case of elements containing polyp substances. 

 The consequence would be an increase in the surface of the 

 roots if they came into contact with solid bodies, while polyps 

 only would grow out in the opposite direction. I found, indeed, 

 in some forms at Naples that roots of hydroids which grew free 

 in the water began to grow much faster and to branch off more 

 abundantly when brought into contact with solid bodies. But 

 in these cases we must wait with our attempts at explanation 

 until the physical and chemical conditions for the form are 

 worked out. For the same reasons I will not go into a discus- 

 sion of the question of what determines the polarization of 



