British Oligocene Ants. 89 



Subfamily Camponotinaje, Forel. 



Tribe F R M I C i N i, Foiel. 



Genus Leucotaphus, nov. 



(XevKos, white, rcKpo?, tomb.) 



Diagno'iif. A Formicine with small iiead, very small clis- 

 coidal cell, and loiior cubital cell. The witig-s are similar to 

 those of Formica and Acanthomyops, but the discoidal cell is 

 much smaller in proportion and the cubitus and radius veins 

 join each other at the apex of the cubital cell — not a little 

 before it, as is usually the case with Formica and the subgenera 

 Donisthorpea, ChthonoJasins, etc., oi Acanthomyops. 



Genotype. Lej)tothora.v gumetensis, Cockerell (1915). 



Leucotaphus gumetensis (Cockerell). 

 Syn. Leptothorax gumetensis, Cockerell, 1915, p. 485, pi. Ixv. figs. 4, 5. 



Cockerell had before him only a not qnite complete fore 

 wing and part of another (whicli he considered a variety) ; 

 and, as he himself writes, " This seems to be a Leptothorax, 

 but I have only the wings to judge from." Unfortunately 

 the fragments described by Cockerell will have to be the 

 holotypes of this species and variety. Many of the specimens 

 in the large series I have studied are nearly complete. Most 

 of the winged specimens present the lateral aspect, others the 

 dorsal, with the wings expanded. The wings are generally 

 complete and the neuration very distinct. The species is, of 

 course, a Camponotine, and not a Myrmicine, there being 

 only a single joint to the pedicel, which bears a scale, as in 

 Formica, Acanthomyops, etc. 



L. gumetensis closely resembles Formica primitiva, Hecr 

 (1850), from the Oeningen beds; but, apart from the generic 

 distinction, it is much smaller. 



I have seen some eight workers, nearly all of them 

 being on the same piece of rock as winged sjiecimens. The 

 head is small, the scale distinct, and some parts of the legs 

 are present, but the general outline is not very clear. The 

 length is 2-2-7 mm. (PI. V. fig. 8^). 



There is also what I consider to be the cocoon on the same 

 b'ock as a winged ant of this species [(b. 71), I. 93-13 

 (PI. V. fig. 8a)]. It measures 3 nun. in length and is shaped 

 as in Acantliontyops. Wheeler tound cocoons witii Formica 

 and Acanthomyops in the Baltic amber. 



The variation in measurementsof these ants is as follows : — 

 Length 2'iJ-o'c> mm.; fore wing, long 2-8-3'7 mm,; hind 



