South African Insects. 187 



latus, Servllle = Capensis, Saiissure), an undescribed 

 species of Thomisus, an Epe'irid, a species of Zilla, and 

 a small motli {Acontia formosa), extremely variable in its 

 colonr, are always found in conspicuous positions, and by 

 the attitudes they assume are difficult to distinguish from 

 the real droppings. The little Mantis in particular, 

 though extremely active on being disturbed when fright- 

 ened or on the watch for prey, doubles all its limbs 

 together and bends back its abdomen in such a manner as 

 to appear quite inanimate. 



Floral imitators. — None of these are peculiar to the 

 Acacia, vmless it be a species of Longicorn beetle, and 

 I am inclined to think that this is not protected by its 

 yellow-coloTU'cd bands. Harpax spinocula, and another 

 species of Thomisus, extremely variable in coloiir, are 

 found on the flowers of this and many other yellow 

 blossoms of different orders, such as Compositce., Liliacece, 

 &c. The flower-packing larva I have already mentioned. 

 I may also state that the blossoms qxq often frequented 

 by small Lamellicorn beetles having very long hind legs 

 which project from the blossoms. Similar insects are 

 also found on many Compositce, and in the same situation 

 may be found spiders which thrust up their forelegs in 

 such a manner as to be hardly distinguishable from the 

 beetles. From the frequency with which I have found 

 these insects associated together I can hardly doubt that 

 the attitude assumed by the spider is imitative. 



Mimicry of ants by spiders of the genus Salticus. — 

 It is well known that among spiders of the genus Salticus 

 there is a group {Attus, Walck.), of which the resemblance 

 to ants is wonderfully close. As I am not aware that any 

 explanation of this curious similarity has been given, I 

 now make known my observations on some of the species. 

 An ant of the genus Crematog aster, and another belong- 

 ing to Camponotus, are each of them imitated by spiders, 

 and in both cases, but especially in the first, I ha^e noticed 

 the spiders mingling with the ants on apparently friendly, 

 or at least neutral, terms. The small black Crematog aster 

 is imitated by a spider of the same size and appearance, 

 being smooth and shiny. It was very curious to notice 

 the spider curving its long forelegs after the manner of 

 the ant's antennte, turning up its abdomen exactly like the 

 stinging Crematog aster, and occasionally halting, and 

 then again speeding along the branches. In fact, so close 



