The Odonata or Dracjonflies of Soiifli Africa. 375 



Cordulinae, as far as known to the writer by personal observation 

 and as far as precise records exist, have in common a peculiar manner 

 of flight. Their motion is particularly sustained, even more so than 

 in Aeschninae, and the same individual may be, with sufficient patience, 

 observed for an almost indefinite time planing to and fro without a 

 moment's pause, and they stand — this particular motion is more deve- 

 loped in them than in any other group of dragonflies — often for a long 

 time on the same point. Their particular manner of motion makes 

 some of them an easy prey, as the collector will know perfectly where 

 the insect will turn up within a short time and be within reach of his 

 net. But he must not wait to see his Cordulina taking rest ; he 

 might wait for half an hour or more and then see it fly suddenly 

 away ; the very regularity of the motion makes a stroke at the flying 

 insect pretty efficacious. But this is not the rule ; in most species 

 the flight is not only sustained but also exceedingly swift, and the 

 insect, moreover, defiant, and not ready to leave the open water or 

 some high level over a forest road. This may be j)articularly the 

 case for the MacromiaAiim of the subfamily, as proved in Mr. 

 Williamson's interesting report. Material from the exotic regions, 

 where dragonflies have rarely lieen coUecteid liy specialists, is scarce, 

 and descriptions are often made from a few, or even from single 

 specimens. 



There is no other group of large, conspicuous and lieautiful forms 

 amongst Odonata where our knowledge is so poor and so fragmentary 

 as for the il/rtcro»n'a- group of Cordulinae. Only the North American 

 species are now well known, and even these only since a few years. 

 The material at our disposal for the present paper is fragmentary. 

 The existence of a greater number of species in the temperate part of 

 the country does not seem probable, but a few more may be expected 

 from these parts where tropical conditions of climate prevail. 



Corduline nymphs of the Macromia groups are similar to libellidine 

 nymphs (see before) in structural details, but characterised by 

 their exceedingly long legs and a cylindrical horn-like process on 

 the frons. They may be looked for in standing as well as in running 

 waters. 



MACEOMIA (Rambur, 1842). 



The African Macromia, all or part of them, may at some later 

 date be separated from the main body of the genus ; a fraction has 

 even been placed in a subgenus (genus according to R. Martin) 

 Phyllomacroinia by de Sc'lys. The characters given for Pliyllomacromia 



