306 Mr. G. A. K. Marshall on 



attitude when I first looked at him. I 

 therefore killed him. 



Judging from Experiment VI. the inability to change 

 can hardly be attributed to weakness caused by want of 

 food. 



[In relation to the above-recorded experiments it is im- 

 portant to know the habits and natural food of the Mantis, 

 and if possible to determine the species. Mr. Marshall 

 kindly sent a specimen of an identical, or at any rate very 

 closely-allied species, tosfether with the following notes. 

 — E. B. P.] 



" Umkomaas Mouth, Natal ; Sept. 3, 1897. — The Malvern 

 species of Mantis is one of the largest out here, and I 

 selected it as I knew it to be almost entirely a butterfly 

 feeder. It frequents chiefly Acacias and their allies, and 

 catches the Oharaxes which come to suck the gum. Un- 

 fortunately they are scarce at Malvern, and I could not 

 procure a single specimen during the winter, for I had 

 hoped to make the very experiment you suggest, viz. 

 feeding exclusively on Acrteas or L. elirysvpims. However, 

 I caught two small ]\Iantises on my arrival here (Umkomaas 

 Mouth) yesterday, but I have not as yet even seen an 

 Acrma." 



^'Malvern, Oct. 7, 1897. — I nm not quite certain whether 

 the Charaxes-eating Mantis sent is specifically identical 

 Avith the one that died from Acr,va diet. I thought it 

 was the same in the pupal stage, but the imago of the 

 latter has the upper wings entirely green, with a small 

 yellowish spot about tlie middle. Unfortunately I have 

 not been able to get one." 



[The Mantis sent (captured at Malvern, Sept. 1897) 

 was PolysirUota, caffra (Westwood), of which the type is in 

 the Hope Collection, Oxford.— E. B. P.] 



VIII. Experiments with Fsendocreobotra wahlhcrgi, Stal, 



female. 

 1897. 

 Sept. 26. I captured at Malvern a full-grown female 

 Ocellated Mantis, which ate a specimen 

 of Acr/va enccdon during the day. 

 „ 27. Gave Mantis two A. cncedon. She ate the 



thorax of one, rejecting the remainder. 

 „ 28. The remaining enccdon died to-day. Re- 

 moved it, and put in one Acriea ncohide 



