LIGHT AND THE BEHAVIOR OF ORGANISMS 463 



this total act the significant part so far as ultimate orientation 

 is concerned is not the making of the five trial and error move- 

 ments but the final " selection " of the best one as a line to be 

 followed in the real locomotor act. Now it is difficult for the 

 reviewer to see how this " selection " can be accomplished 

 without involving just that feature of symmetrical stimulation 

 which is the essential part of the tropism idea. The worm 

 presumably moves finally in the direction of that trial which 

 through the stimulation of symmetrical points on its body has 

 been found to be most favorable for orientation. Without 

 complicating this problem by raising the question as to the 

 means by which the worm could retain impressions of the five 

 trial and error movements preparatory to following up one of 

 them, it must be evident that these movements in their effect 

 on orientation are quite subordinate to symmetrical stimulation, 

 in fact, not essentially involved in orientation at all. Thus, 

 trial and error movements in the orientation of such animals 

 as the earthworm are wholly . secondary occurrences and the 

 orientation of these animals depends in reality upon tropic methods. 

 It is to be regretted that this view of the question should have 

 received no attention in Professor Mast's book, for it is on the 

 basis of criticisms, such as this, that some investigators have 

 regarded the trial and error method when applied to many 

 higher invertebrates as of no real significance. 



The minor defects in the volume are slight. Omitted letters 

 occur on page 236, eleventh line from the bottom, and on page 

 239, eighth line from the top. The bibliography contains no 

 reference to Smith's article on the relations of negative and 

 positive animals in volume thirteen of the American Journal of 

 Physiology. It is difficult to understand the ground for the 

 sequence of authors' names such as^ occurs in the categorical 

 statement made on pages 265 and 266, where, contrary to the 

 common practice, the order of dates is not followed. This 

 looseness at times comes to be really misleading where, for instance 

 on page 52, Verworn precedes Loeb with dates 1886 and 1887, 

 while in the bibliography Loeb's earliest reference is given with 

 the date 1888, and Verworn 's with that of 1889. Notwithstand- 

 ing these minor defects, as well as the general one which was 

 pointed out in the preceding paragraph and which characterizes 



