1891.] EXAMPLE OF PROTECTIVE MIMICRY. 463 



There is little doubt that we have to do with a palatable insect 

 much relislied by insect-eating foes, which defended itself, like the 

 great majority of its allies, by Protective Resemblance (Procryptic 

 Colouring), in this case by a likeness to leaves. The green colour and 

 compressed body were probably evolved in response to the need for 

 concealment. But as the increasing acuteness of foes, also subject 

 to natural selection, enabled them to see through a disguise which is 

 so universally adoi)ted by palatable insects, it became of advantage to 

 certain hard-pressed furms to resemble something which was posi- 

 tively objectionable to their enemie-i rather than merely useless and 

 uninteresting. And in this case the transition from Protective 

 Resemblance to Protective Mimicry (Pseudaposematic Colouring) 

 would be especially easy, i'or it would be brought about bv com- 

 paratively insignificant modifications of colour and form. Such an 

 easy transition into so marvellous a metliod of defence is of course 

 due to the accident that specially dctended insects generally 

 associated with pieces of leaf are well known in the same locality. 



My friend Col. Swinhoe has objected to this interpretation that it 

 is extremely difficult to believe that the habit of the ant has been 

 continued sufficiently long for the change to have been produced in 

 the Homopterous insect. But the habit seems to be very ancient, 

 inasmuch as it is common to a group of species inhabiting tropical 

 America, and presumably dates from the time of their origin in a 

 common ancestor. Furthermore it has been shown that, detailed as 

 the resemblance certainly is, it was probably prepared for by an 

 earlier resemblance to leaves alone. 



It must also be remembered that several species of the genus 

 Stegaspis are known in tropical America, and that the published 

 fi-ures and descriptions of these show that the same shape and 

 outline are characteristic of the whole. It is therefore extremely 

 probable that Mr. Sclater's interpretation ap|-lies to a group of 

 closely allied Homopterous insects inhabiting a district where the 

 various species of leaf-carrying ants are well known and abundant. 



In the preparation of the Plate accompanying this paper I have 

 to thank Miss Kelly for kindly lending me some sketches of living 

 Cooshie Ants bearing leaves ; these sketches, drawn upon the spot, 

 were of especial value. Miss Horman Fisher also made for me a 

 careful drawing of a leaf-carrying Ant in the Natural History 

 Museum. Mr. H. M. J. Underliill has prepared the coloured figures 

 from this material. 



DESCEIPTION OF PLATE XXXVI. 



Fig. 1. X about 6'5. Immature form of an unknown species of Stegaspis 

 {MemhracidcB) as seen from the right side. Drawn with the camera 

 lucida. 

 2. A representation of the Stegaspis (on the right) beside the leaf-carrying 

 Ant (on the left). The" fragment, of leaf carried by the latter is 

 generally gnawed along its upper edge. The figure of the ant was 

 chiefly adapted from Miss Herman Fisher's drawing of a specimen in 

 the Natural Historr Museum. 



