388 PROCEEDINGS OF THE AMERICAN ACADEMY. 



condition in Vanessa, but the uncertain elements which have been 

 mentioned must be taken into account before conclusions can be drawn 

 with any degree of confidence. 



Mention has been made of individuals which at times showed much 

 hesitancy in these reactions, and having turned toward one light 

 seemed disturbed by the other, which was now behind them. At such 

 times the head was usually held high, in the manner described by 

 Holmes (:05*, p. 312), when the light was moved to a position behind 

 the animal. It was found that by painting over the posterior half 

 of the eyes, so that the light was excluded on that face, the influence 

 of the light behind the insect could be prevented. After having their 

 eyes treated in this manner, even individuals which had previously 

 shown great hesitancy held their heads low, and once oriented toward 

 either light, crawled straight on in that direction without turning. 

 This was not resorted to, however, in any of the trials summarized in 

 Table VII. If the eyes were painted on their anterior halves, exactly 

 the opposite state of affairs was brought about. In this way individuals 

 which had previously shown no hesitation when once oriented toward 

 either light, were caused to turn first to one and then to the other, the 

 reason being that, as soon as the animal was facing one light, that one 

 could no longer be seen on account of the paint, while the other, shin- 

 ing on the unpainted posterior half of the eyes, caused the insect to 

 turn from the apparent darkness ahead to the light behind. In this 

 way it was kept turning about and about, and making no definite 

 progress in any direction. 



8. Pomace Fly {Drosophila ampeloiyhUa Loew). 



Upon this form and the succeeding one, only negative results were 

 obtained as regards the question of inferred image-formation. It was 

 expected from their general behavior toward light that these animals 

 would be eminently suitable for this investigation, and the failure to 

 obtain definite results with them was a disappointment. Long series 

 of experiments were tried both with these flies and with snails, in- 

 volving much time and labor ; a brief account of them is given here 

 merely to illustrate how certain unforeseen conditions may influence the 

 reactions of an animal in ways entirely unexpected. 



In his investigations upon the reactions of Drosophila, Carpenter 

 (:05) found that this little fly is always positive in its reactions to 

 light, at least in its locomotor responses. ^ It, like Vanessa, often 



' There are undeniably times when Drosophila may crawl away from the 

 source of light ; but this is probably in response to other stimuli, such as food, 



