200 DWIGHT E. MINNICH 



affected only by distance stimuli, and these were identical for 

 all the substances employed, viz., water vapor. Other explana- 

 tions must, therefore, be sought. 



Let us consider this difference in responsiveness to sodium 

 chloride more fully. The results of tables 5 and 6 are based on 

 too large a number of observations for us to suppose that either 

 nullifies the significance of the other. The result in each table 

 undoubtedly gives a correct idea of the responsiveness of animals 

 under those particular conditions. The only possible con- 

 clusion, therefore, seems to be that the responsiveness to a 1 M 

 solution of sodium chloride may vary from time to time with the 

 physiological state of the animals. This explanation becomes 

 more plausible when the data from which tables 5 and 6 are 

 compiled are analyzed with respect to individual specimens 

 instead of the entire group. It then appears that while a few 

 butterflies may vary in the course of a single experiment in their 

 responses to sodium chloride, the great majority are constant, 

 either responding or failing to respond, alike in every trial. The 

 correct explanation, therefore, seems to be that in experiments 

 2f and 3f (table 5) there were a number of butterflies which were 

 unresponsive to NaCl, while in experiments 2g and 3g, due to 

 some physiological change, a number of these individuals had 

 become responsive. Besides this normal variability, possible 

 effects of the operation must not be overlooked. In practically 

 all animals there was a loss of blood. It is true this was very 

 slight in most cases. Nevertheless, it is entirely conceivable 

 that a factor of this sort might considerably increase the sensi- 

 tivity to a salt solution. 



In experiment 2g, table 6, and, to a less extent, experiment 

 3f, table 5, the number of responses was about the same for each 

 of the five substances. In both of these experiments the period 

 of inanition undergone by the animals was greater than was the 

 case with the other experiments of the same series. Thus, in 

 experiment 3f the animals were 50 to 98 hours old, while in 

 experiments If and 2f they were but 51 to 75 hours old. The 

 specimens in 3f, therefore, may have been as much as twenty- 

 three hours older than specimens in If and 2f. In the case of 



