146 EVOLUTION AND GENETICS 



these specifications. They are small fusiform creatures with a 

 dorsal tubular nerve cord, well developed segmental musculature, 

 a notochord, and numerous pharyngeal clefts. They are able 

 either to lead a sedentary life or to swim about. 



Amphioxus has some characters which are not primitive, but it 

 corresponds so nearly in its general structure with the hypothetical 

 ancestor of the vertebrates that we are justified in considering 

 it an approximation of their true progenitors. In the course of 

 time the lancelets of the present have apparently taken on only a 

 few new characters while the vertebrates have been attaining their 

 remarkable diversity (Fig. 82). 



The Amphioxus Theory of Vertebrate Origin. The possibilities 

 of development of lancelets or similar ancestral forms is suggested 

 by the habits of the existing species. They live along sandy shores 

 and pass some time buried in the sand, but in response to tidal 

 action they are able to swim rapidly against the currents. Their 

 food concentrating mechanism is of an unusual type, well adapted 

 to sedentary life. The animal is one of those peculiar forms men- 

 tioned in the last chapter, able to respond with equal facility to 

 widely different conditions of environment. As pointed out be- 

 fore, the constant action of one environmental condition would 

 be expected to call forth only one response. Hence a shift into 

 the still waters of the ocean might develop its sedentary characters 

 and produce something akin to the tunicates, while the opposite in- 

 fluence, constant motion of the waters in swift flowing streams, 

 might bring out the characters which fit Amphioxus for active 

 resistance to such motion. Here the geological record supplies the 

 interesting information that the earliest fossil fishes show associa- 

 tion with fresh waters, and not with marine deposits. 



The Annelid Theory of Vertebrate Origin. The annelid theory 

 is based on the resemblance of embryonic characters of vertebrates 

 to certain annelid structures, and on the resemblance of the adult 

 vertebrates in general plan to an inverted annelid, with a different 

 mouth and anus (Fig. 83). Embryology shows that the mouth 

 and anus of vertebrates actually develop secondarily, and are not 

 at the original ends of the primitive gut, as in the annelid. More- 

 over, several highly developed invertebrates of the present live 

 normally in an inverted position. The theory is thus less peculiar 

 than it seems at first thought. 



Reversal of the annelid results in the nerve cord lying above the 



