The Bionomics of South African Inserts. 537 



Mr. Marshall both call attention to the cause which has 

 doubtless prevented the fact from being generally recog- 

 nized at an earlier date, viz. the changes which take place 

 in dried specimens of Mantispa. On this account, and 

 because of the important part played by movement, the 

 appreciation of the mimetic resemblance required the 

 study of the living insect. 



"Salisbury, Sept. 21, 1900.— The large South African 

 3Iantispa grandis is an excellent mimic, on the wing, of the 

 Belenogastcr wasps. I caught one at Malvern, on my way 

 home in 1896, which I gave to McLachlan. This insect 

 completely took me in ; it flew out of a loquat-tree which 

 I was beating, and I at once took to my heels thinking I 

 had struck a nest of these vicious wasps. Fortunately I 

 kept an eye on the insect, and, as it seemed to be a s])ecies 

 of Belenogastcr new to me, I followed it up and caught it, 

 when to my surprise and delight it proved to be only a 

 3Iantis2)a. Unfortunately in a dried specimen the resem- 

 blance is much spoilt by the shrivelling and discoloration 

 of the abdomen," 



.34. Convergent Groups of South African Hemi- 

 PTERA (G. A. K. M.) 



A. Black and Bed Lyg;void Group (Represented on 

 Plate XIX). 



Lvamidis f^Y^,^^^^ rivularis (fig. 44); L. elegans (fig. 46); L. crudclis 



■^^ \ (fig. 47) ; Grapho.stetluis servus (fig. 45). 



H/idtcviidcG Reduvius sj). (fig. 43). 



In this group I consider that the Lygseids form a 

 Mlillerian association, of which the Bechcvius is probably a 

 Batesian mimic. The former insects are very abundant, 

 occurring on many different plants, but the Bygmi are 

 especially fond of the balloon-like seed-vessels of Gompho- 

 carpus. The Bcduvius inhabits much the same stations, 

 though I have never seen it (to my remembrance) actually 

 in company with the Lygseids, and it is a decidedly rarer 

 insect. 



