OF WASHINGTON. 175 



series. Both mimic ants, and so it is interesting to compare 

 these two groups of cases and see how nature has arrived at 

 similar results by using different paths. The mimicking Attidae 

 belong to the genera Salticus, Synageles and Synemosyna. The 

 mimicking Drassidse belong to two genera which are somewhat 

 widely separated, one in the Drassinse, Micaria ; the other in 

 the Clubioninse, Thargalia. Of the five genera mentioned, 

 Synemosyna and Micaria most closely resemble ants. 



On comparing a Drassid with an Attid one will notice very 

 striking differences of important structures ; differences which 

 have led to their wide separation in modern classifications of 

 spiders. Yet these differences are masked and covered by cer 

 tain resemblances which are of no important value for classifi 

 cation, but which nature has found admirably suited for the 

 deceptions she wishes to produce. They are both ground 

 spiders wandering for their prey ; this is a necessary habit in 

 order that they may mimic ants. The plan of markings on the 

 abdomen consists, in both groups, of transverse bands ; in each 

 of these families there are, of course, plenty of exceptions. But 

 throughout the higher genera of the Attidse one can trace quite 

 accurately a tendency for transverse markings. The exceptions 

 are found in the lower genera, Marptusa, Menemerus, Itius. 

 This style of markings is also the predominating one in the 

 Drassidse. 



There is one great difficulty to overcome in the mimicking of 

 an ant by a spider ; that is, in the former the body is divided 

 into three portions, while in the latter it is composed of two. 

 But nature can overcome this by contracting the abdomen of the 

 spider, and in many cases this is done ; but she can also deceive 

 the eye by color. Given the tendency for transverse markings, 

 it is easy for a transverse interrupted white band to be devel 

 oped on the abdomen so as to give the appearance of a. con 

 traction ; this is especially deceiving when it is developed in 

 connection with, and at the same place, as a slight constriction. 

 Two or more bands may be present and this tends to destroy 

 the real contour of the body and to make it look narrower. 



In Synemosyna formica the deception may be said to be per 

 fected. Usually blackish at each end and slightly reddish in 

 the middle ; a constriction near middle of the cephalothorax 

 and another on the front half of the abdomen ; at each constric 

 tion an interrupted light band, slender light-colored legs. 

 Sometimes the spider is almost wholly reddish-yellow. Its 

 habits are not very much like an ant; it is quite slow and 

 deliberate. Synageles lacks the cephalic constriction and the 

 abdomen is broader, yet it resembles an ant very closely. It is 

 very active and its movements supply what it lacks in structure. 

 Salticus does not structurally so closely resemble ants, yet there 



