LOCUSTIDS MIMICKING ANTS 257 



resemblance. In spite of this he asks, ' Is this imitation 

 an accidental freak of nature ? ' If Myrmecophana were 

 the only example of such resemblance the question might 

 fairly be asked, but in view of the numerous other equally 

 close resemblances to ants, produced in various ways, it 

 is quite unnecessary. The suggestion of an 'accidental 

 freak ' can never explain such close likeness, in appear- 

 ance, in movements and habits (so far as they are known), 

 in locality — a likeness, furthermore, to particular insects 

 in the environment, members of a specially successful and 

 aggressive group — a likeness not produced in one way but 

 in many different ways. To suggest an ' accidental freak ' 

 as the explanation shows an amazing credulity, only to be 

 explained by the bias which is ready to accept any inter- 

 pretation except that afforded by the theory of Natural 

 Selection. 1 



1 Several examples of the kind are now known, although some doubt 

 has been thrown upon the details of Brunner's interpretation in the case of 

 Myrmecophana itself, an insect of which the habits were entirely unknown. 

 Guy A. K. Marshall has captured a Locustid of the same genus in 

 company with ants upon a small bushy vetch, near Salisbury, Rhodesia 

 (Trans. Ent. Soc. Lond., 1902, p. 535, Plate xix, fig. 59). The un-ant-like 

 parts of the body were green and thus invisible against the leaves ; but in 

 the dead specimen they have faded to a pale yellowish tint much like that 

 described by Brunner on the corresponding parts of M./allax. In fact, 

 Malcolm Burr is inclined to think that the Rhodesian example is M./allax 

 and that Brunner was misled by a faded specimen. But even if this 

 particular example is to be explained in a slightly different manner — 

 obliteration by harmony with a green background instead of a glaring one, 

 several other instances have now come to light, and these are to be 

 explained precisely in the manner suggested by Brunner. Thus Colonel 

 J. W. Yerbury has shown me an Asilid fly — Promachus ioplerus (Wied.), 

 from Pari, with white patches on the sides of the basal abdominal 

 segments, leaving the appearance of a slender black stalk like that of so 

 many Hymenoptera. Again, R. Shelford describes four Bornean 

 Longicorn beetles of the genera Scytasis and Oberea as ' marked with a 

 large white patch of pubescence on the sides of the first and second 

 abdominal segments, which patches, when the beetle is seen in profile, 

 give an impression of a wasp-like waist, from the posterior end of which 

 the abdomen appears gradually to swell in size' (Proc. Zool. Soc. Lond., 

 1902, vol. ii, p. 238, PI. xix, figs. 13-15). The most interesting case, 

 however, is that of a common British Reduviid bug, Nabis lativentris, which 

 in its immature stages bears an ant-like appearance produced exactly in 

 the manner described by Brunner. It is remarkable that so interesting a 

 form of mimicry in such a common insect should have been undescribed 



