Harper, Behavior of Corethra. 449 



Experimejit y. — The first experiment was repeated with larvae 

 which had been kept in the laboratory several weeks and it was 

 found that they reacted negatively to sunlight or artificial light and 

 collected in the shaded portions of the aquarium. Repetition of 

 Experiments 2 and 3 with them showed some manifestation of 

 positive geotropism in strong light but the response was weakened. 

 The movements of such animals are however quite sluggish and 

 little can perhaps be inferred from their behavior beyond its sub- 

 normal character. The negative light response was not universal 

 as a considerable proportion came to the lighted end and kept 

 striking against the glass. 



It may be asked, what interpretation is to be put upon the posi- 

 tive light response in its relation to the normal behavior. It is 

 not to be concluded from the fact that they move downward in the 

 daytime that the larvae remain quiescent. Doubtless they are 

 attracted by places of greater intensity, although prevented from 

 going upward by their geotropism. At the depth then at which 

 thfey remain in the daytime they may be inferred to seek the 

 optimum intensity, which at nightfall or in cloudy \^eather 

 includes movements upward as well. Their well developed 

 eucone type of eyes, which are very rare in larvae, are well 

 adapted for vision at the lower depths, and their pigmented air- 

 sacs may have some specially important respiratory function, as 

 MiALL suggests, enabling them to remain at these depths. Upon 

 this point the writer has nothing to offer. 



AUTOMATIC MOVEMENTS DUE TO GRAVITY. 



These animals are so balanced m the water by their air sacs 

 that a slight change in their specific gravity causes them to rise ■ 

 or sink. When hungry they become more buoyant, as may be 

 well seen in an aquarium full of them. A jar of Corethras will 

 soon clean out the supply of small Crustacea. Let a scarcity 

 of food arise and the behavior changes. They are all seen to be 

 going through the same performance. Each larva rises auto- 

 matically a little distance and then ducks down by one of its lash- 

 ing movements, which are the same as the type described, except 

 that the vertical component is a little greater than usual and is 

 always directed downward. This performance of automatic 

 rising and ducking down is repeated indefinitely. In the case of 

 each individual it is quite rhythmical, the rate seeming to vary in 



