8 University of Michigan 



three anterior and two posterior, in hamatus, we have another 

 venational character probably of specific value. 



It is assumed in this paper that the Cristalina female pre- 

 viously described (Paper Number 59, loc. cit.) is conspecific 

 with the allotype of ^, furcatus from Bejuma, as no charac- 

 ters for separating these two females have been detected. How- 

 ever, the Magdalena basin in which Cristalina lies and the 

 Orinoco basin in which Bejuma lies are widely separated, and 

 it is not impossible, though I believe it is improbable, that 

 when males from the Magdalena basin are available, they may 

 be found to differ from those of the Orinoco basin. In fact, 

 with such scanty material no prediction as to the number of 

 species occurring in each region is possible. 



In the allotype female of furcatus 'the postoccipital horns 

 are as figured for the Cristalina female (Paper Number 59, 

 loc. cit., Plate I, figure 13), but due to the teneral condition 

 .of the allotype the horns are bent and slightly crumpled against 

 the prothorax as the head is turned one-fourth around with its 

 dorsal surface to the side. 



The vulvar lamina is likewise identical in the two and is 

 very similar to that of hamatus (figure 11, loc. cit.), except 

 that in furcatus the branches are more slender, slightly longer, 

 the outer edges more nearly parallel, and they come off from 

 the base more abruptly, or, to express this last point dift'er- 

 ently, each branch at the base is narrower in furcatus than in 

 hamatus. The vulvar lamina of infans has not been figured, 

 but, as described, it is of the same general character as it is in 

 hamatus and furcatus. 



The known males of Archaeogomphus are beautifully sep- 

 arated by the outline of the first joint of the penis (seminal 

 vesicle) in posterior view, as shown in figures 3 and 4 accom- 



