346 AMERICAN SPIDERS AND THEIR SPINNINGWORK. 



this sensitiveness of the entire skin, instead of limiting the perce})tion to 

 the eyes alone? 



There is indeed another theory which may be suggested, namely, that 

 the color surroundings of tlie spider, in some manner not now exjilicable, 

 so rapidly influence the organism of the creature that a change of color 

 is produced in harmony with its environment. Can we suppose, in this 

 case, that the spider possesses tlie power to influence at will the chromat- 

 ophores or j^igment bodies, so that they may change her color with chang- 

 ing siteV 



There is another explanation of the above peculiar habit of Misumena. 

 Many insects are strongly attracted by yellow colors, and as insects are / 

 the chief food of spiders, it is natural that the familiar resorts 

 Seekmg gf insects should be the places most affected by spiders. That 

 nsec insects have such attraction to colors has already been shown, 



and that they are drawn to yellow colored flowers has been fully 

 established by Miiller in his remarkable volume on Alpine flowers. ^ This 

 author gives a table recording the numerous visits of various insects to 

 flowers of different hues ; and a study of the table shows that butterflies, 

 bees, flies, and gnats, and other insects manifest a strong preference for 

 yellowish white and for yellow flowers. With such a fact as this in view, 

 we may, perhaps, conclude that the habit of Thomisus and Misumena to 

 frequent flowers of the character above described, resolves itself into the 

 well known instinct of all animals to seek their food in those resorts 

 where the supply is most abundant and accessible. This explanation does 

 not, of course, exclude the fact that the spider, in seeking such favorable 

 site, may be guided by its sense of color, but it reduces it to a suboixli- 

 nate rank. 



VII. 



Walckenaer ^ advanced the idea that the form of the cocoon corresponds 

 with that of the abdomen of the mother. This is in some measure cor- 

 rect, for the abdomens of spiders have most frequently an oval 

 Relation shape, and this is substantially the shape of the cocoon. But 



p when one comes to compare the shapes of the abdomens of in- 



coons dividual spiders with the shapes of their cocoons, the excep- 

 tions are so numerous and decided that no such generalization 

 can be accepted. 



The same author suggested that some correspondence exists between 

 the color of the cocoon and that of the mother's abdomen. The facts, 

 however, at least as far as American spiders are concerned, will not sus- 

 tain this theory, except in a general way. For example, the colors of 

 the abdomen of Argiope cophinaria are yellow, black, white, and brown. 



' Miiller, Alpen Blumeu, page 487. ' Apteres, Vol. I., page 147. 



