NO. 1571. DRAGONFLIES OF BURMA AND SIAM— WILLIAMSON. 271 



to forsake the water for the trees and bushes, where they rest inactive 

 and inconspicuous till the reappearing sun brings them again to the 

 stream. Continued cloudiness ma}^ cause them to leave the river, 

 scattering far and wide over fields and woods. Moreover, species of 

 this subfamily are well known to have a brief seasonal range in 

 temperate regions, and the same thing is probably true in the tropics. 



Because of the scarcity of positive records for Burma and Lower 

 Siam of species of the subfamilies under discussion in tliis paper, a 

 slightl}^ different treatment from that emploj^ed in the paper on the 

 Calopteryginse has seemed desirable. In this paper I have given dis- 

 tribution and brief notes on all the species known in the oriental 

 fauna. From this it must not be supposed that I expect subsequent 

 collecting to show a large percentage of these species to occur in 

 Burma and Lower Siam. On the contrary, I believe a number of 

 species at present undescribed will be revealed. 



Throughout the })aper the halftone figures of wings are of arbitrary 

 size and give no idea of the relative size of the wings in various species. 

 The figures representing thoracic color pattern are diagrammatic, all 

 drawn over the same outline, and give no idea of the difl'erent forms 

 and sizes of the insects themselves. The figure illustrating venational 

 nomenclature is still more diagrammatic. (See figs. 1-3.) The 

 remaining figures are drawn to scale — the same for all — and give an 

 idea of the relative sizes in different species. 



KEY TO THE ORIENTAL GENERA OF THE SUBFAMILY GOMPHIN.^ (iMAGOES). 



ORDER ODONATA (Neuroptera Odonata, Para- 



neuroptera). 



aa. Front and hind wings dissimilar in shape, the latter usually broader at base; the 

 quadrangle of the suborder Zygoptera « divided to form the triangle and super- 

 triangle. Males with one inferior abdominal appendage which, however, may 



be deeply bifid or rudimentary Suborder ANISOPTERA 



b . Antenodals of first series mostly coinciding with those of the second series ; triangle 

 of front wing with its long axis at right angles to the length of the wing, and 

 t riangle of hind wing with its long axis parallel to the length of the wing. 



Family Libellulid.e 

 bb. Antenodals of first series not coinciding with those of the second series, excepting 



in the case of two, which are thickened Family jEshnid.e 



c. Radial and median supplements present; triangle of front wing at least as elon- 

 gate as triangle of hind wing; M2 paralleling Mj at least as far as the stigma. 



Head globose. Lateral abdominal carinee present subfamily .eshnin.^ 



cc. Head transversely elongated ; eyes separated or meeting at a single point only. 

 Lateral abdominal carinse wanting. 

 d. Radial supplement developed ; triangle of front wing at least as elongate as 

 triangle of hind wing. Median labia llobe divided. . .PctaHa and allies, 

 probably worthy of subfamily rank ; (not regional. ) 

 dd. Radial and median supplements not developed ; M.2 ijaralleling R^. 



"See Proc. U. S. National Museum, XXVIII, p. 167. 



