COLOR AND THE BEHAVIOR OF ARTHROPODS 103 



he studied ought to be re-examined by another investigator. 

 The evidence as it now stands cannot be said to prove that 

 decorator crabs choose colors for concealing themselves which 

 harmonize with their surroundings. 



The crab-spider. — Emerton ('02) says: " Whether spiders 

 prefer flowers colored like themselves is an unsettled question; 

 at any rate, Misumenas of all colors and both sexes have been 

 found on white fiow^ers." From the results set forth in the present 

 paper there can be little doubt that the majority of Misumenas 

 are to be found on flowers colored like themselves. Thayer ('09) 

 figures Misumenas as an example of "obliterative coloration," 

 and in experiments described in this paper it is shown that its 

 colors usually harmonize with the background. The question 

 is, whether the presence of a majorit)^ of yellow spiders on yellow 

 flowers and of white spiders on white flowers, is due to color 

 changes in the spider itself, or to the selection of a particular 

 background by each individual, or to some other influence. 



McCook ('89-'93, ^^^- 2, P- 341) says the color of spiders may 

 be influenced by a variety of factors, among these moulting, 



' advancing age, gestation, muscular contraction, sex difference 

 and excitement may be mentioned as being sometimes important. 

 Ne\'ertheless he says (vol. 3, p. 51), that there are no authentic 

 cases of rapid color changes in American spiders, and (vol. 2, 

 p. 271) because the color changes of Misumena are so slow, he 

 says, "we are therefore compelled to the conclusion .... that the 

 spider sought the flower and settled upon it, either accidentally 

 or by choice." Beddard ('92, p. iii) mentions a rapid color 

 change described by Heckel, in a spider which belongs to the 

 same family as Misumena. This species was Thomisus onustus, 

 which he says has three color varieties in the flowers of Con- 



* volvulus amensis, and two other colors in other flowers. These 

 varieties correspond closely to the flowers and Heckel maintained 

 that these spiders could change their colors in three or four days. 

 ■ Beddard says, however, that the evidence for color change was 

 not by any means conclusive. Davenport ('03) mentions a 

 light colored sand-spider which became gradually darker when 

 placed on grass. 



In the knowledge of the writer, the experiments described in 

 the present paper are the first in which spiders have been allowed 



