THE FUTURE OF THE LIVING WORLD. l8l 



tion into definite types the chance for future modifi- 

 cation becomes rapidly less. It is only the abso- 

 lutely undifferentiated which has infinite possibilities, 

 for as soon as a single step is taken in any direction, 

 the possibilities become finite. Now it is plain that 

 this continued specialization cannot go on forever. 

 Since evolution does not retrace its steps, every step 

 in advance limits the possible lines of development. 

 All the descendants of the vertebrate line must 

 conform jto the vertebrate type. The vertebrates 

 become separated into fish, reptile, and mammal, and 

 each group is still further fettered in its develop- 

 ment by the special line which its ancestors have 

 taken. The descendants of the animals which have 

 started the order of birds cannot take any new line. 

 They can develop this type to perfection, or they 

 may lose their special characters, but there they must 

 stop. And thus, with every step in advance, the 

 possibilities of expansion are constantly decreasing. 



Now a continued specialization of this sort is sure 

 to reach a limit ; it must run to extremes and event- 

 ually stop. Physical laws will of themselves set 

 limits to every line of advance, even if there be no 

 such limits determined by the organism itself. It is 

 easy to find examples which will show that such has 

 been the general history of groups in the past. Some 

 have reached the extreme of their development in 

 the distant past, and have ceased to advance, or, 

 perhaps, have disappeared. Others seem even now 

 to be at the summit of their advance, and others 

 still are yet advancing. The line of development 

 represented by the trilobites has completely ex- 



