MIMICRY AMONGST THE BLATTID.E. ;^51) 



genus and of I'latyzosteria Br., are fond of sunnin<x themselves 

 on the tops of posts and tree-stumps. (Jommantler J. J. Walker 

 when collecting in Australia fret|uently encountered a species, 

 Cosmozosteriu lateralis Walk., which emitted so vile a smell that 

 lie always refrained from touching it. 'I'his species is russet- 

 brown in colour with some variable yellow markings on the 

 thoracic and abdominal tergites, and on each posterior angle of the 

 ninth abdominal tergite is a brilliant orange-red spot ; when the 

 insect is at rest these two spots ai'e almost concealeil, the ninth 

 tergite being somewhat I'etracted within the preceding one, but 

 on the approach of an enemy the apex of the ii.bdomen is elevated 

 and slightly di.stended so that the orange spots become conspicu- 

 ously displayed to view. A better example of a warning signal 

 associated with highly distasteful properties could not be found. 



Pi'of. Baldwin Spencer, writing in 'Nature' of July 28, 1892, 

 ]). 3U9, says : — " One morning, when Mr. Frank Connelly and 

 myself were digging for woi'uis, we accidentally cut in two a 

 cockroach. From between the segments in its back it poured 

 forth a milky- white fluid, possessing an odour so execrable a.no 

 pungent that it drove us from the spot." It is unfortunate tl at 

 the species was not identified, but I expect that it was one of the 

 Poh/zosteria group. 



Commander J. J. Walker says of I'eriplaiteta forttpps Walk., .a 

 .synonym of riatj/zosteria no^He-zealandkc. J^r., that it is " very 

 evil-smelling," but that its smell " is cjuite mild in comparison 

 with several of the Australian species." (Entom. Monthly Mag. 

 (2) XV. p. 70 (1904).) 



Dr. G. B. Longstatf recently took this species in .some numbers 

 in New Zealand, finding it, however, not in exposed situations, 

 but under logs and the bark of dead trees. Quoting from his note- 

 book Dr. Longstaif tells me that one specimen had a " moderate 

 cockroach odour, evanescent," of another that it had " a strong 

 peculiar fetor." The discrepancy may possibly be accounted for 

 by sexual differences (see rema.rks later on Eur)/cotis floridmia 

 Walk.). 



All the Australasian species of the Pob/zosterla section appear 

 to be conspicuous iu.sects. Many of them are shining black, a. 

 colour which is noticeable enough in Nature when associateil 

 with free exposure ; others are black edged with yellow, or with 

 red legs. The species of A namesia Tep})., are chestnut-bi-own 

 banded or margined with yellow. Cosmozosteriu zonata Walk., 

 is black, banded Avith j^ellow oi' orange. Many of the species of 

 Polyzoster'm Burm. ai'e bright with metallic colours, and Eu- 

 zosteria mitchelU Angas, with its bronzy dorsal surface, spotted 

 and barred with orange or yellow, its pale yellow venti-al surface 

 and sky-blue tibite, is the most gaudy cockroach yet discovere<1. 

 In the New World the Pohjzosteria section of the Blattinje is 

 represented by the genei'a Ev,rycotis 8tal, and Pelmatosilpha 

 Dohrn. The species do not appear to be so blatantly conspicuous 

 as their Australian relatives, but it seems likely that most, if not 



