NO. 2192. DRAOONFLIES, CALIFORNIA AND NEVADA— KENNEDY. 521 



I.iater Williamson observed Cordulegaster maculatus Selys of the 

 eastern United States while in the act of ovipositing, which he de- 

 scribed as follows: ^ 



Cordulegaster maculatus waa observed to fly down from trees, alighting on algae- 

 covered rocks in the stream bed. On portions of these rocks not co veered with water 

 they crawled about in an awkward, almost crippled manner, thrusting their abdomen 

 with much commotion into the algae beneath the water. 



An interesting summary of our knowledge of oviposition in Cordu- 

 legaster is given by Dr. C. Wesenb erg-Lund in Odonaten-Studien.^ 



It is interesting to note that this very special form of ovipositor 

 may not be of any great systematic significance. It occurs in all of 

 the few cordulegasterine Odonata known. It is found in the libel- 

 luline genus Uracis (see fig. 144, F, Uracis ovipositrix Calvert), the sev- 

 eral species of which occur in South America and have habits of ovi- 

 position similar to those of Cordulegaster except that they oviposit in 

 the mud about the water. More strange yet is tlie fact that a smgle 

 species of the large genus Sympetrum of the North Temperate Zone 

 has this same highly specialized ovipositor. This species, Sympetrum 

 cordulegaster (Selys) (see fig. 144, E), occurs in northeast Asia and 

 nothing is known of its habits. While the majority of the Sympe- 

 trums have no ovipositor and oviposit by washing the eggs from the 

 tip of the abdomen, one or two Asiatic species have extraordinarily 

 long vulvar laminae and form a sort of connecting series between the 

 numerous species of Sympetrum. with no laminae or very small ones 

 and tins freak species, Sympetrum cordulegaster (Selys) , with the vul- 

 var laminae longest and united into a monster ovipositor (see fig. 

 144, E), Sympetrum frequens (Selys), with small vulvar laminae (fig. 

 144, A) and Sympetrum eroticum (Selys), from Japan, with very large 

 vulvar laminae but these not united into an ovipositor (fig. 144, C 

 and D). Figure 144, B, is Sympetrum matutinum Ris, an intermediate 

 form. These widely scattered cases of this special form of ovipositor 

 are an excellent example of convergent evolution. 



As neither of the western species of Cordulegaster has been well 

 figured, I have shown in the figures some of the pecuHarities of struc- 

 ture and have represented the color patterns diagrammatically. 

 Diadema is sHghtly larger than dorsalis. A male and female of dia- 

 dema measure as follows: Male, abdomen, 64 mm., hind wing, 4S; 

 female, abdomen, 65 mm., hind wing, 55. Male and female of dorsalis 

 measure as follows: Male, abdomen, 55 mm., hind wing, 44; female, 

 abdomen, 57 mm., hind mng, 45. As is shown in the figures 135, 

 diadema, and 141, dorsalis, the appendages on segment 10 of the male do 

 not differ noticeably. Figure 136 shows the pecuhar armature of the 

 tibia in diadema which does not differ from that in dorsalis. Figure 1 40 

 shows the organs of segment 2 of the male dorsalis, which are very simi- 



1 Ohio Nat., vol. 7, 1907, p. 144. 



2 Mitthl. biol. Siisswasserlab. Hallerod (Denmark), No. 16, p. 187. 



