352 COSTA RICAN EPIGOMPHUS (oDONATa) 



to prevent crossing between different "species." A comparison 

 of the figures on Plate XV, illustrating verticicornis and 

 armatus shows that, although the females of both species agree 

 in possessing strong postocellar tu]:)ercles (pod), they differ 

 markedly in the rear of the head and in the positions of the 

 superior ocular scars (soc) , while the appendages of their respect- 

 ive males are very different. The presumption would seem to 

 be that these two species would not cross, but we have as yet no 

 observations to control our speculations. On the other hand, 

 I can see no a priori reason why the male say of quadracies might 

 not mate with the female of suhsimiUs or of suhohtusiis, but of 

 course mere mechanical ability to pair is no assurance of the pro- 

 duction of fertilized ova. It may be that a long time will elapse 

 before we have positive answers to these questions, l^ut their 

 solution should not be of insuperable difficulty. 



