FAM. MEMBRACID.K ii 



African membracids, states (in correspondence) : « These ants {Plagiolepsis custodiens F. Smith) repre- 

 sent in South Africa our species of Formica and Lasiis and probably derive niuch of their food from 

 membracids and coccids. » Other South African ants attending Membracida collected by the writer in 

 the vicinity of Victoria Falls were kindly determined by M. H. St. John Donisthorpe of the British 

 Museum as Polyrhachis {Myrma) schistacea and Crematogaster {Scroccelia) castanea rufonigra Emery. The 

 conimonest species of attendant ant taken by the writer in the Malay Peninsula was the small 

 longlegged form Plagiolepsis longipes. In Java the only ant taken by the writer was Myrmecaria brunnea 

 subcarinata Sm. Dr. M. A. Lieftinck of the Zoological Museum at Buitenzorg also reports this same 

 species as attendant on Hypsauchenia recurva l-^unkh., and states as a footnote in one of the writer's 

 reports (igSSb) : 



« At Tjibodas I have frequently wathced small colonies of this membracid on the end- 

 shoots of a small tree of Talauma candollei L., an indigenous Magnoliacea cultivateti in the 

 garden. Both imagines and larvte were found on this tree at any time of the year, the larvse 

 being assiduously attended by ants {Myrmecaria subcarinata Sm.) for the sake of a sweet secre- 

 tion emitted from the extremity of their body. A colony of Hybandoides sumatrensis Funkh.. 

 also found at Tjibodas, but on a host plant unknown to me, was being attended by the same 

 species of ant. » 



Ants attending various species oi Ebhnl and Leptocentrus collected by the writer in India were 

 determined b}- Mr. Donisthorpe as Polyrhachis bihamata Den., and Dolichodcrus {Hypocliiiea) affinis 

 glabripes Fovvl. 



Of course all of tliese examples represent merely the commoner forms collected on varied occa- 

 sions and from scattered localities and doubtless include only a few of the species of ants which concern 

 themselves with the Membracida. 



The behavior of both the ants and the membracids is much the same wherever they are studied. 

 The ants stroke their charges with their antennas, whereupon the inembracids give off from the 

 anal tube a liquid that issues in bubbles in considerable quantity. The anal tube of the membracid 

 is capable of great evagination, especially in tli(! nymphs, iii which it is long and cylindrical and usually 

 tipped with a fringe of fine hairs. The honeydew is eagerly taken from the end of this tube by the 

 ants. In many species the adults as well as the nymphs are sought, and the ants seem to be as attentive 

 to one as to the other but the adults have not been observed to excrete the liquid to the same extent 

 as the nymphs. That the ants are well paid for their attendance can hardly be doubted when their 

 industry around the cougregatioiis of Membracida is noted. In many cases the hiding places of the 

 membracid are at once betrayed by the svvarms of ants present. It is not believed that the ants herd 

 or segregate their charges as in the case of certain of the Aphidce, but shelters for membracid nymphs 

 are not uncommon. 



The advantage to the membracid is evident by the protection given by the ants, which do not 

 hesitate to bite viciously the fingers of the collector who seeks to remove nymphs or adults fiom the 

 host. The ants have been observed also to attack spiders and attempt to drive awaj' Reduviidie in the 

 neighborhood of membracid colonies. 



It as been suggested in a preceding section that in some cases the ants may take advantage of the 

 punctiires made by the inembracids to procure sap. The best evidence of this is the fact that ants often 

 reinain gathered about the spot where the niembracid has fed after the latter has moved away, and 

 apparently they find something there to attract them. This may be explained, of course, by the theory 

 that anal fluicl fioin the membracid has been left on the plant, but it does not account for the fact that 

 the ants are often at the anterior rather than the posterior end of the insect. 



The feeding habits of the Membracidat seem in no way affected by the presence ofants, which 



