750 



THE AMERICAN NATURALIST 



[Vol. XLII 



science, built up by systematizing the verifiable facts alone, with 

 that which searches for the psychic processes underneath what 

 is observed. The difficulties of making a positive science from 

 the unverifiable psychic implications of the actions of animals 

 is well illustrated by the conditional and potential forms in 

 which the author is forced throughout to clothe her statements. 

 Thus, in discussing the psychic aspect of orientation to light 

 (p. 184), the predicates of six successive sentences are: we "can- 

 not imagine"; we "may conjecture*'-, "is the human experience 

 most closely resembling"; "appears to be"; "may have"; it 

 ' ' is possible that. ' ' The experimentalist becomes convinced more 

 than ever of the need of building up his own positive science 

 of behavior, composed of verifiable propositions, and omitting 

 psychic factors — though there is no reason why he should look 

 with an unfriendly eye on the attempt, as a separate thing, to 

 supply conjecturally the missing psychic elements. 



The difficulties in preparing a satisfactory account of the 

 animal mind are further increased by the high degree in which 

 the experimental science of behavior shares the provisional and 

 uncompleted character of all science. Animal behavior even 

 as a purely objective science is merely in its beginning. No 

 greater mistake could be made (and this our author evidently 

 recognizes) than to suppose that our present experimental 

 knowledge is sufficient for defining sharply the psychic powers 

 of animals. It is quite possible that the picture of the mind of 

 one of the higher animals that miuht. be drawn by an observing 

 and judicious dog lover would be much more adequate than the 

 rude sketch which experimental science is now able to give us. 

 The material furnished by the old Anecdotal School, and by the 

 Lovers of Nature, doubtless contains much most-important truth, 

 to which the experimental method has not yet succeeded in at- 

 taining: only, as Miss Washburn says, it is not possible to tell 

 what is true, what false. This material furnishes valuable finger 



»l»l< to distin-insl tl in , „ ,}, f ;1 | s m m mil behavior, it 



addition of one verifiable fact to another, which has proved the 

 method of advance for other sciences. At any given time then 



