420 FRANK W. BANCROFT 



This may perhaps be made clearer by a mechanical analogy. 

 Let us imagine a spring which holds a vessel into which water is 

 flowing. Now we may say that the water acts on the spring in 

 two ways: (1) It produces oscillations in the tension of the spring 

 if the changes in the amount of water in the vessel are made with 

 sufficient rapidity. (2) It exerts a constant pull due to the total 

 amount of water in the vessel ; and it continues to exert this con- 

 tinuous pull, which is different from the effect produced by the 

 changes, even when the amount of water in the vessel is changing 

 and the vessel may be oscillating in response to these changes. 



Other criteria for deciding the nature of the stimulus are em- 

 ployed by Mast in his paper on the firefly ('12, p. 271, 270). He 

 says : ' ' The males do not orient when exposed to continuous illu- 

 mination. " ''These results, together with the fact that orien- 

 tation in the male ordinarily does not begin until after the flash 

 of light which induces it has vanished, demonstrate conclusively 

 that the process of orientation and the direction of locomotion 

 after orientation are not regulated by the continuous action of the 

 light, and that these reactions are consequently not in accord with 

 Loeb's theories of tropisms according to which orientation is, as 

 he puts it ('06, p. 135) a function of the constant intensity.' 

 Stimulation which results in orientation in the fireflies studied is 

 unquestionably due to changes in light intensity much as it is in 

 Stentor, Euglena and many other similar organisms. " Here the 

 criteria that were apparently employed in deciding the question 

 are: 



1. Lack of orientation to continuous illumination. 



2. Short duration of orienting light. 



3. Orientation occurs after the orienting illumination has 

 ceased. 



Of these criteria the last two are entirely without value in 

 deciding this question for Blaauw has shown that for Avena the 

 Bunsen-Roscoe law holds for light duration of .tto , s^i) and even 

 ToViy of a second. Moreover, in many of his experiments the 

 orientation regularly took place after the light which caused the 

 orientation had ceased to act. And yet Blaauw proved by means 

 of the Bunsen-Roscoe law that in this case the orientation was a 



