344 AMERICAN SPIDERS AND THEIR SPINNINGWORK. 



spider was then admitted to the cage and, after liaving become suflficiently 

 domesticated, was gently driven into a specially colored com[)artment, say 

 the bine. It was then left without interference to select such position 

 as it might prefer in any one of the four differently colored comi)artments. 

 When the spider had changed its position and remained therein a suffi- 

 cient time to indicate a preference for the color under whicli it rested, it 

 was again disturbed and moved to another color. If, for example, it set- 

 tled within the red compartment, it was transferred to tlie yellow, and so 

 on, a record being made of the various changes and preferences. This 

 process was continued during several days, in which several hundred ex- 

 periments were made. As a result it was found that among all 

 p^ „ '^ the spiders experimented with 181 preferred the red, 32 the yel- 

 Eed. ^^^' 11 blue, and 13 green. The preference of the spiders for 



red was thus decidedly marked, resembling, although in a more 

 marked degree, the preference of ants for tlie sanie color, as demonstrated 

 by Sir John Lubbock's experiments,' which ajjpear to have suggested those 

 of Professor Teckham. 



A test case was made which gave a striking result, quite in confirma- 

 tion of the experiments as above described. An individual of Lycosa 

 nigroventris, which had shown a strong preference for red, choosing that 

 compartment 33 times out of 41, was temporarily blinded by covering 

 its eyes with paraffine. When put within the cage it was found that 

 the spider remained quiet in whatever compartment it was placed until 

 it was driven out. If placed in the blue compartment, with its eyes 

 as close as i)ossible to the red, it showed no inclination to enter, al- 

 though this color had before proved so strongly attractive. Its prefer- 

 ences, or rather its locations, during tJic resulting experiments, arc recorded 

 as follows : — 



Preferences after blinding : Red G, yellow (>, blue 6, green 5. 



Preferences before blinding : Red 33, yellow 5, blue 0, green 3. 



Such results leave scarcely any room for doubt that in some way the 

 spider had been influenced by a color sense, since, wliile it possessed 

 normal vision it exi)rc.sscd a most decided preference for the red color, 

 but when temporarily deprived of vision settled indifferently and about 

 equally in all the colors represented in the series, there being no stronger 

 preference for red than there had been in previous experiments for the 

 blue compartment, which it had entirely shunned. These results seem to 

 justify the conclusion that there exists a color sense in certain spiders. 



It is to be remarked, however, that in all the cases recorded, and ap- 

 parently in all experimented upon, the individuals were chosen from the 

 Lycosids alone. These spiders undoubtedly have a keen sense of sight, 

 although I am inclined to think that in this respect they are inferior to 



Ante, Bees, and Wasps," page 189. 



