Uy K. J. TILLYARD. 711 



the Gomphince should show us that they also are not really 

 similar except in an approximate sense. The triangles of 

 Fetahira appear to me to be closer in formation to those of a 

 majority of the CordidiincB than to those of any other subfamily. 

 (c) Anal angle {$) and memhranule : angle very marked, mem- 

 branule very small, as in the Gomphince. 



Reviewing the above points, there seems to be no doubt that 

 Petalura is a far closer approach to the Gomphince than to the 

 GaloplerygidcE. And of the genera comprised in the Gomphiiue, 

 Ictinus should be singled out as the nearest approach to Petalnra, 

 particularly in the shape of the head and thorax, the length of 

 the pterostigma, and the dissimilar triangles crossed by one or 

 more nervules. Apart from these points, the differences between 

 Petalura and Ictinus are so insuperable that the family Petaluridoi 

 must stand. It should probably be placed at the end of the 

 division Aniso2)tera which the American authors use to include 

 the Libellididce and ^Eschnidce with their subfamilies. The 

 division Zi/goptera, of which the Calopterygidce are the first 

 family, will follow it in the linear order; but it must be borne in 

 mind that the gap between Petalura and these latter is far greater 

 than the gap between Petalura and Ictinus. 



While I was collecting in the Cairns district of North Queens- 

 land, during the summer of 1904-5, I was told of the occurrence 

 there at rare intervals of a dragonfly of such enormous propor- 

 tions that I scarcely credited the story. It was said to come 

 swooping down " like a bird," and local residents went so far as 

 to declare that " its bite would pretty well kill you." When I 

 captured Anax giittatus at Atherton I thought this was the 

 species refered to, but when I showed it to a Cairns resident he 

 declared that the one he had spoken of was far bigger than that. 

 I kept on the look-out, and a few days before I left Kuranda I 

 was rewarded by seeing an enormous dragonfly along the banks 

 of the River Barron. I was unable to capture it, but I could 

 see that it was a Petalura. A day later a local collector brought 

 me a female of the species, which he had captured in the bush. 

 It was in bad condition, but measured about 6J inches across the 

 56 



