174 



THE AMERICAN NATURALIST [Vol. LII 



them, possibly the "slowest," limits action and causes 

 death. As indicated later, according to the phase rule 

 the greater the number of "phases" interacting the 

 lesser the number of possibilities of change. This is 

 not a condition limited to organisms, but is a general law. 

 It is perhaps in some sense as this that we can concede 

 "differentiation" as a cause of death. 



The animal as an agent, or individual, behaves accord- 

 ing to its own system, to the extent that it is an independ- 

 ent unit, and these activities are cyclical. All systems 

 tend to perpetuate themselves. Bancroft's law for all 

 systems is that: "The broadest definition of it is that a 

 system tends to change so as to minimize an external dis- 

 turbance." In other words this is a perpetuating tend- 

 ency, a method of assimilation, of which reproduction 

 may be considered but a special phase ; it is not solely a 

 peculiarity of organisms, as is often stated, but of all 

 systems. Sedgwick ( '10, p. 177) has said: "It is a prop- 

 erty of living matter to react in a remarkable way to ex- 

 ternal forces without undergoing destruction. . . . This 

 property of reacting to the environment without under- 

 going destruction is, as has been stated, a fundamental 

 property of organisms." In these features the animal 

 acts only as other systems tend ; as a catalyzer, it hastens 

 changes and maintains itself. The activity of the animal, 

 its centrifugal stress, causes it to collide with its environ- 

 ment, while, on the other hand, there is the environmental 

 bombardment, both of which, within certain limits, tend 

 to interfere or destroy the animal. On the other hand, 

 the tendency of the system is to "minimize disturbance," 

 to change within, to minimize, to "retreat" from inter- 

 ference (absolutely or relatively), and in this manner to 

 a large degree, the system is perpetuated. To be sure, 

 many individuals perish, but the system of the species 

 continues. The rate of change of the system can be modi- 

 fied only as fast as its slowest member can change, and 

 on this account many individual systems are destroyed. 



In addition to influences which "interfere" with sys- 



