COLE. — IMAGE- FORMING POWERS OF VARIOUS TYPES OF EYES. 349 



we are safe in assuming (1) that the image of the larger light is spread 

 over a larger area on the retina, or better, that a larger number of vis- 

 ual elements receive light, and (2) that the intensity of the stinndus act- 

 ing on any one visual element is very much less than that which acts 

 on the one element, or very few elements, which are stimulated by the 

 small light. Analogy with the reception of stimuli by sense organs in 

 general makes the assumption reasonable that the difference between 

 even a very weak light and no light falling on a visual element may 

 have a much more stimulating effect upon the animal than the same, 

 or even a greater, difference in the amount of the light at higher inten- 

 sities. Such being the case, we should expect an animal to react more 

 strongly to that stimulus which fell upon the larger number of visual 

 elements — that an animal normally positive, for example, would be 

 more strongly positive to the large light than to the small, and similarly 

 that a negative animal would tend more often to move away fi-om the 

 larger than from the smaller luminous area. Parker's ( : 03) results give 

 evidence that in Vanessa such is really the case. In the following pages 

 are recorded the results of experiments made under more uniform and 

 more precisely determined conditions upon a variety of animals, some 

 negative and others positive, some with eyes and others without. 



The kinds of animals experimented upon were not taken entirely at 

 random, but species whose reactions to a single light were definite and 

 well known were selected as far as possible. The number of animals 

 from which selection could be made was greatly limited owing to the 

 optical dilficulties with animals living in water. Only those were used, 

 consequently, which could be studied in the air. In addition to the 

 animals Avith Avhich a considerable series of experiments was finally 

 made, a large variety of animals were tested in a more or less complete 

 manner, to determine their suitability for the work. These included a 

 number of insects, such as the water strider (Hygrotrechus), an elater, 

 a small species of cockroach, and certain kinds of bees and flies, sev- 

 eral myriapods and spiders, the European slug {Limax maximus), the 

 horned toad (Phrynosoma), and two species of salamanders {Plrthodon 

 (jhitiwmm and P. er)ithronotii^). Some of these did not appear from 

 their movements to be responsive to light, others were inconstant in 

 their responses, and still others were too inactive for the purposes 

 of this work. Those from which more or less satisfactory data were 

 obtained are as follows : 1. The common dungworm or earthworm 

 {Alhihihophom foet'ida [Sav.]). 2. A large land planarian (yy/)w//«w 

 l-eireuse Moseley). 3. The mealworm (larva of Tenehrio molitor Linn.). 

 4. The sow bug (Oniscu.t a.<f^/lns Linn.). 5. The cockroach {Periphnn^ta 

 americana Linn.). 6. The mourning-cloak butterfly {Vanessa anti- 



