Odonata from New Caledonia. 49 



by decided liypertrophyj frequently occurs in the case of the 

 first and third of them. Both the basal subcostal cross-vein 

 and the hypertrophied antenodals occur in all the four species 

 from New Caledonia, not even excepting the small, open- 

 veined one, S.fenella, 



Those two characters emphasise the close relationship 

 subsisting between the Synthemini and the TEschnidee. 

 Indeed, >S. miranda, S. montaguei, and 8. flexicauda may be 

 regarded as the most archaic Corduliinpe yet discovered, and 

 the nearest to the ancestral TEschnid or ^Eschnid-like stock. 

 lu the presence of cross-veins in the median space, they 

 remind one more particularly of the Chlorogomphinte, and 

 the wings in that subfamily exhibit the same kind of sexual 

 dimorphism as in Sijnthemis in respect of the complexity of 

 the anal loop, as well as the width of the wings. Furthermore, 

 the males of Chlorogomphinte possess the peculiar tibial keel 

 which is found alone in themselves and the CorduUinse. It 

 was characteristic of De Selys that his unerring instinct 

 immediately led him to compare Synthemis miranda with 

 Chiorogomphus magnificus. Tillyard has drawn attention to 

 the close similarity which the nymph of Synthemis bears to 

 that of Cordulegaster, but it would not be surprising to find 

 that it will present at least an equally great resemblance to 

 Chiorogomphus or Orogomphus, whenever a nymph of one of 

 those genera becomes known. 



Synthemis regina * is the true representative in the 

 Australian fauna of S. miranda and its New Caledonian allies. 

 For one thing, it is the nearest to them in point of size. 

 Then, the anal loop in its hind wings consists of two 

 enclosures in the male and three enclosures in the female 

 sex, as in ^S'. miranda. Furthermore, the resemblance to that 

 species extends to important abdominal characters, such as 

 the anal appendages and dorsal spine of the male and the 

 ovipositor of the female. The existence of such a clear link 

 between the three species before us and the more typical 

 members of Synthemis seems to render it inadvisable to 



* Si/ntlu'7m's regina, in both its sexes, was described by De Selys from 

 " (^iieeusland " material iu the " Mus(5e brit. et collect. jMacLachlan." 

 The well-preserved male iu the National Collection, ticketed "N.8.W.," 

 and caiTyiug- De >Sel3's'.s identiticatiou-label, \ regard as the holotype, and 

 have marked it accordingly. I have done this, notwithstanding the dis- 

 crepancy iu the locality, and the presence in the MacLachlau Collection 

 of au incomplete male labelled " (Queensland " (on white paper) and (in 

 De Selys's handwriting) " Synthemis regina de Selys cJ'' (on piuk paper). 

 The allotype is undoubtedly the female in the same private collection, 

 carrying white and piuk labels inscribed in the same way (except for the 

 changed sex symbol) as the paratype male. 



Ann. <fc Maq. N. Hist, Ser. 9. Vol. viii. 4 



