ARTHUR DEWITT WHEDON 383 



species instantly fall into two groups — those in which the varia- 

 tion in length between males and females is normal {Lestes ungui- 

 culatus, forcipatus, etc.) and those in which the male exceeds the 

 female b}^ at least 4 mm. and often as high as eight millimeters 

 (Lestes rectangulans, eurinus and perhaps vigilax). Sufficient 

 material is not at hand to determine how much of this variation 

 is due to large and small specimens and how much to actual 

 variation in the abdominal length, the thorax remaining con- 

 stant. Lestes rectangularis seems to have the widest range. 



It is especially essential to note at this point that in some 

 species the female is consistently longer than the male : Ischnura 

 vertigalis, one of the •commonest, most wideh' distributed, and 

 successful of the Zygoptera is very constant in this respect. The 

 male abdomen varies from 20 to 22 mm., that of the female 

 from 23 to 25 mm. This is true of Ischnura posita also. In 

 Lestes this is, naturally, uncommon, but even here the male of 

 L. uncaius is slightly shorter than his mate; L. forcipatus is often 

 so. Anomalagrion hastatum and Chromagrioyi conditum are like 

 Ischnura. The Enallagmas tend toward equality between the 

 sexes, there being but few cases known to the writer of variation 

 in favor of the female. The Argias are normal with but one or 

 two exceptions among American forms. 



The North American species of Anisoptera show, fi-oni lower 

 to higher groups, a procession from females markedly longer 

 than the males to equality or even to longer males. The 

 subfamilies may be characterized as follows: 



Gompiiinae: In form and, in many genera, in size the Gom- 

 phinae possess the most highly modified abdomen of the Odonata. 

 The females of Ilogenius and Gomphits quite generally exceed 

 the males in length. In some closely allied genera the sexes 

 seem equal. The ratio of thorax to abdomen is also greater, 

 indicating that the females are not simply the larger specimens. 



Aesiimnae: There are few if any exceptions to the normal con- 

 (Htion of male longer than female in this group. The reverse 

 may at times be true in Epiaexhna Jicros. Walker (1912) gives 

 measurements of all the North American si)ecies of Acshva de- 

 rived from a large collection of specimens. In tlu> case oi Acshnn 

 irnifata only does there seem a distinct dilTciencc in favor of the 



TRANS, am. ENT. SOC, XLIV. 



