Laws and Hypotheses for Behavior 275 



and changes in the bonds by which particular movements 

 follow particular physical events. To find when and how 

 animals whose natures remained nearly or quite unchanged 

 by the satisfying and annoying effects of their behavior, 

 gave birth to animals that could learn, is perhaps a still 

 harder task. But these tasks concern problems that are 

 intelligible matters of fact. They do not require a student 

 to get out of matter something defined as beyond matter, 

 or to get volition out of tropisms, or to get ideas of space 

 and time out of swimming and sleeping. 



The evolution of the sensitivities and of the action- 

 systems of animals has already been subjected to matter-of- 

 fact study by naturalists. The evolution of the connection- 

 system will soon be. Each reflex, instinct or capacity, 

 each bond between a given situation presented to a given 

 physiological state and a given response, has its an- 

 cestral tree. Scratching at an irritated spot on the skin 

 is older than arms. Following an object that is moving 

 slowly does not have to be explained separately, as a 

 'chance' variation in dogs, sheep and babies. The me- 

 chanical trades of man are related to the miscellaneous 

 manipulations of the apes. Little as we know of the con- 

 nection-systems possessed by animals, we know enough 

 to be sure that a bond between situation and response 

 has ancestors and children as truly as does any bodily 

 organ. Professor Whitman a decade ago showed the pos- 

 sibility of phylogenetic investigation of instinctive con- 

 nections in a study which should be a stimulus and model 

 for many others. In place of any further general account 

 of the study of the phylogeny of the connection-system, 

 I shall quote from his account of the concrete phylogeny 

 of the instinct of incubation. 



