496 Mr. II. Campion on some 



eut is direct observation in the field. Further information 

 could probably be gained by the examination of ihc contents 

 of the alimentary canal in newly-caught specimens. Another 

 mode of enquiry has been suggested by Professor H. Maxwell 

 Lefroy, who has written on Indian Dragonflies and their 

 prey (Journ. Bombay Soc. xx. pp. 236-258, 1910). He 

 says : " In the field one sees dragonflies sitting on a con- 

 venient plant or support and darting off every now and then 

 on the cl ase. Belo\v such a point, to which the same 

 dragonflies comi back constantly, one finds their excreta." 

 A study of these excreta, undertaken by the same authw, 

 revealed the presence of remains of Orthoptera, Aculeate 

 llymenoptera, Lepidoptera, Coleo[)tera, Diptera, and 



lllivncliotM. 



In connection with the study of predac.'ous insects gene- 

 rally, Professor E. B. Poulton has published sixteen illustra- 

 tions' of the kind of prey selected by Dragonfiies as food 

 (Trans. Ent. Soc. London, 1906, pp. 398-101). The 

 following records will serve to supplement tiiose illustrations, 

 and they are here presented in the same convenient form. 

 'The captors and prey from Nyasaland and British East Africa 

 cited in Table I, (pp. 498-501) were obtained by Mr. b. A. 

 Neave, while visiting those countries on behalf of the Imperial 

 IBureau of Entomology. I am indebted to Mr. Guy A. K. 

 Marshall, the Director of the Bureau, for his kindness in 

 allowing me to study this material, as well as some other 

 examples of a similar kind sent from Uganda by Dr. b. D. 

 JI. Carpenter (Sleeping Sickness Commission ot the Royal 

 Society^ The cases collected in Essex and Surrey by my 

 brother' and myself have been already published m our 

 annual reports upon Biitish Dragonflies, but they are now 

 brou.rht together and incorporated with the original records 

 from^Africa. I have considered it advisable to separate the 

 cases of cannibalism— if this term can be rightly employed 

 when the captor and prey do not beh.ng to one and the same 

 j^pecies— from the instances in which Diagonfiies have sought 

 their food, more legitimately, among insects of other orders. 

 My reason for doing so is that cases of this description, 

 where one Dragonfly hunts another, are quite as germane to 

 an enquiiy as to what kind of animaU prey upon Dragonflies 

 as they are to the matter at present under consideration. 



The whole of the African material mentioned in the 

 followin<r Tables has been presented to the I5riti..h Museum 

 (Naturaf History) hy the Imperial Bureau ot Enluniology. 



In view of the vvell-known fact that, both in collections 

 and in the field, the males of must species of Dragoutlica are 



