BEHAVIOR IN LOWER METAZOA 251 



been spoken of as random movements (Holmes, 1905). The word 

 "random," of course, implies only that these movements are not defined 

 by the position of the stimulus; it does not signify that the move- 

 ments are undetermined. The principle of cause and effect applies to 

 these movements as well as to others. But the causes lie partly within 

 the animal; each phase of the movement aids in determining the suc- 

 ceeding phase. The earthworm may turn to the right at a given instant 

 merely because it has just before turned to the left. Reactions in which 

 a succeeding phase is determined by a previous one have sometimes been 

 called chain reflexes (Loeb, 1900; Driesch, 1903). If this term is used, 

 it needs to be kept in mind that in most cases the succeeding phase is not 

 invariably and irrevocably called up by the preceding one, as is implied 

 by this term. On the contrary, the relation between the two is extremely 

 variable. One type of action may be repeated many times before the 

 second type comes into play, and the order of the different actions is by 

 no means always the same. Thus the preceding phase is only one factor 

 in deciding what shall be the present action. The latter depends upon 

 the entire physiological state of the organism, which is determined by 

 various factors. Illustrations of this are seen in the righting reaction of 

 the starfish and many other animals ; in the series of reactions by which 

 Stentor responds to a mass of carmine grains in the water (p. 174); in 

 that by which Stoichactis gets rid of waste matter lying on thedisk (p. 202), 

 and the like. 



The diverse physiological states of lower organisms have been little 

 studied. This is partly because it is rarely possible to observe them di- 

 rectly; it is only through their effects upon action that they become 

 evident. Thus the real data of observation are the actions ; if we con- 

 sidered these alone, we could only state that a given organism reacts under 

 the same external conditions sometimes in one way, sometimes in another. 

 This would give us nothing definite on which to base a formulation and 

 analysis of behavior, so that we are compelled to assume the existence 

 of changing internal states. This assumption, besides being logically 

 necessary, is, of course, supported by much positive evidence drawn from 

 diverse fields, and there is reason to believe that in time we shall be able 

 to study these states directly. Before we can come to a full understand- 

 ing of behavior, we shall have to subject the physiological states of 

 organisms to a detailed study and analysis, as to their objective nature, 

 causes, and effects. 



The most noticeable and therefore best-known physiological states 

 of lower animals are those which depend upon changes in metabolism. 

 The reactions of the starfish and the planarian to many chemical and 

 mechanical stimuli depend, like those of the sea anemone, on the 



