The Bionomics of South African Insects. 509 



become a broad oval in shape. A pair of white spots (as 

 well ns a white line towards the outer border of the 

 posterior half of the elytra) has also arisen on the anterior 

 part of the elytra in A. omoi^lata, var. mellyi (Fig. 26), and 

 it is of deep interest to note that these spots at a little 

 distance or when tlie insect is moving would resemble the 

 entirely different thoracic spots of fhoracicaiixv more closely 

 than these latter resemble the entirely homologous spots 

 of pctcrsi. The comparison of Fig. 23 with 22 and then 

 Avith 26 will bear out this conclusion. Perhaps one reason 

 for the development of the remarkable structures in which 

 the spots are situated in thoracica (they are placed in the 

 concavities of special lateral outgrowths of the thorax) 

 may be in order to favour the synaposematic approach to 

 an arrangement like that of mellyi; for by this means it 

 is possible for the spots to attain approximately the same 

 size and shape, and at the same time to retain an interval 

 between them which corresponds to that obtaining in the 

 very different position upon the elytra. The concavities 

 exist however in a much smaller form in A. maxillosa, and 

 the white markings in them are inconspicuous. 



In A. nimrocl from West Africa (Fig. 25) we see a further 

 development of the mellyi pattern in the appearance of 

 another pair of spots on the posterior region of the elytra, 

 while in A. sex- guttata., from India (Fig. 24), this appear- 

 ance is combined with the two spots of thoracica, thus 

 building up a warning pattern of remarkable simplicity 

 and effectiveness, being an almost exact negative of the 

 six of dominoes. The success of the aposeme is much 

 enhanced by the approximate equality of the shorter 

 intervals between the spots of each pair, and the longer 

 ones between the spots of one pair and the corresponding 

 spots of the next. The great breadth of the thorax 

 permits this symmetry in one direction without the 

 development of outgrowths like those of thoracica, while 

 symmetry in the other direction has been rendered possible 

 because the middle pair of spots occupies a more posterior 

 position on the elytra than does the corresponding pair of 

 nimrocl or mellyi. It will be seen by a glance at Figs. 23 to 

 25 that the addition of the spots of Fig. 23 to the thorax 

 of Fig. 25 would produce a very inferior warning pattern 

 as compared with that of sex-guttata (Fig. 24). 



Mr. Marshall gives the following account (1902) of the 

 habits of these formidable beetles : — 



