554 Saw-flies, Gall-flies, Ichneumons, 



ants, bees, and wasps has long been a subject of popular interest and an 

 object of much scientific observation and experimentation more or less 

 rigorously conducted. Speculation, both popular and scientific, concerning 

 the causal factors concerned has run a wide gamut, from the declaration 

 of Bethe that ants are simply complex machines responding mechanically, 

 with fixed strictly rellex reactions, to physico-chemical stimuli, to the anthro- 

 pomorphic comparisons of the natural-history popularizer, who reads into 

 the behavior of the "wonderful little ant people" human emotions, human 

 reason, intelligent discrimination, and volitional action. 



A difficulty met with at the very beginning of any discussion of the be- 

 havior of social insects is the lack of precise definitions of three presumably 

 classificatory terms distinguishing, on a basis of cause, three kinds of behavior 

 or action, viz., reflexes, instincts, and intelligence. Another more funda- 

 mental difficulty in the actual study and interpretation of animal behavior 

 is the absolute lack in ourselves of any criterion or means of interpretation 

 of action other than our exijerience of our own sensation and psychologj-. 

 Nevertheless the matter can be, and is now being, undertaken in a rational 

 and unbiased spirit, and is attaining important positive results based on 

 ob.servation and experiment conducted with rigorously scientific method 

 and expre.s.sed with scientific caution. Although little more than an ap- 

 preciable beginning has been made in this work, we can already dis- 

 tinguish some of the springs or factors, both intrinsic and extrinsic, which 

 determine the actions of these insects, and we can define scientifically some 

 of the limitations as well as some of the possibilities of their purposeful 

 behavior. 



Between the cleanly mechanical or reflex theory of Bethe, Uexkull, and 

 others, and the reflexes plus instincts and animal-memory theory of Was- 

 mann, Loeb, and Wheeler, or between this and the instincts plus intelligence 

 theory of Lubbock and Forel, there is no sharp line, although between Bethe 

 and Forel there is a wide gulf. What modern investigation has clearly and 

 positively done is to cut away the anthropomorphism of the careless popu- 

 larizer, and to compel a strong leaning toward a belief in the efficiency of 

 reflex and instinct to explain most if not all of ant behavior. What would 

 not have been heard with any patience at all a few years ago, that is, a purely 

 mechanical, i.e., reflexive reaction to physico-chemical stimuli, explanation 

 of many of the "wonderful" actions of ants, as their perception of paths, 

 their recognition of nest-mates, and swift attack on strangers, their refrain 

 from attack on other species living in s\Tnbiotic relations with them, etc., etc., 

 is now heard with careful attention. Couple with this purely reflexive theory 

 the theory of inherited specialized instincts developed by natural selection 

 from widely diffused generalized instincts and most of us are inclined to 



