THE SENSES AND TMKIR ORGANS. 



I lie movement was not so decided. At night the direction was reversed, 

 and the in ignition set toward the library lights, and the colony settled as 

 near them as possible. 



This behavior was so often and uniformly observed that the conclu- 

 sion was (piite satisfactory, but I nevertheless made a series of experi- 

 ments which fully confirmed it. For example, one evening I found the 

 brood massed at two points (a and b, Fig. 287) on either side of a toy 

 wooden column, hanging in a dim light which fell from a lamp that 

 previously had been turned down. An oil lamp giving a bright light 

 was now lit, and so placed (beyond the point c) that one cluster (at a) 

 was in shadow, the other (at b) faintly illuminated. In twenty minutes 

 twelve individuals had passed over from the partly shaded spot (b) to 

 the illuminated point (c), and about half the group in the shadow (at a) 



; 



FIG. 287. Experiment to show the perception of light by young spiders. The group in shade were all 

 transferred to the shelf by lighting the lamp thereon. 



hiid crossed to the better light at b. The lamp was then removed to 

 the opposite side of the table, reversing the conditions of light, casting c 

 in the shadow, b partly in the shade, and throwing full light upon a. 

 Instantly a movement began among the spiders now in the dark at c, 

 who turned and ran rapidly along the lines communicating with the col- 

 umn. In less than two minutes only four of the twelve remained ; one of 

 these soon followed, and shortly the other three. In the meanwhile the 

 shaded group at b was being broken up by an active transfer to the illu- 

 minated section at a. , 



Another experiment gave the following results : I captured a female 

 Kpeira selopetaria at Atlantic City and placed her within a box to cocoon. 

 May 2('>tli, a cocoon was formed in the angle of the box, over which was 

 placed a triangular piece of sheeted spiimingwork attached at the three 



