30 KNIGHT DUNLAP 



general hypotheses, from which it proposes to make deductions 

 to be tested by further observations and experiment. 



The material here copied from The Johns Hopkins Circular, 

 by permission of The Johns Hopkins Press, is reprinted verbatim 

 with the omission of some of the footnotes, excepting that the 

 term reaction is substituted for "reflex" and reaction arc for 

 "reflex arc." This change is necessary because of the confusion 

 which has resulted from the attempt to wrest "reflex" from its 

 older and established meaning, and is one which I hope will be 



A 1 



Special Sense 



Organs 



Brain and 

 Cord 



Muscles 



generally adopted. The term "reflex" should be used to desig- 

 nate a certain class of reactions only, namely, those which are 

 least variable, since the flavor of that meaning persists in clinging 

 to the term hi spite of the attempts of various authors to give it 

 a wider definition. 



More changes in the wording would be appropriate to the con- 

 temporaneous discussion of the topic, but do not seem essential. 

 If I were writing this part of the paper anew, I should discard 

 the term "sensation," since it does not seem possible to give it a 

 fixed definite meaning in psychology or physiology, and in spite 



