442 Annals of the South African Museum. 



long iii ? ; anterior lateral eyes a short diameter from the anterior- 

 border of the carapace. Anterior row of eves strongly procurved,. 

 medians round and smaller than the oval laterals. Posterior row 

 practically straight ; medians smaller than the anterior medians, long 

 oval, and touching the laterals, which are considerably larger, but 

 themselves a little smaller than, and narrowly separated, from, the 

 anterior laterals. Sternum slightly longer than broad ; labium less 

 than twice as broad as long, rounded apically and niuticous. Chelicera 

 with teeth on inner border only. Legs moderately long and robust ; 

 tibiae of 1st two pairs not thickened in $ and only slightly in 2nd 

 legs of <$ ; order of legs in $ 4, 3, 2, 1, in ? 4, 3, 1, 2. Tarsi not 

 scopulate, claws with a single row of 5-7 teeth in <^ and 8-10 in the 

 ? . Abdomen dark, without stripes or lighter marking as in Ischno- 

 thele. Superior spinners keeled on the under surface ; longer than 

 the abdomen, considerably longer than the carapace, and resembling 

 those of Ischnothele. Tarsus of $ palp short ; palpal organ piri- 

 form, elongate, curved, and reaching beyond the tibia. First leg 

 normal ; 2nd leg bearing on the under surface of the tibia a median 

 projection curving upwards and armed with 2-3 teeth ; tibia consider- 

 ably longer than broad ; metatarsus bearing a small sub-basal tubercle, 

 armed on the side or base with a stout spine. 



On examination of the types of E. coffer, Poc., and T. australis, 

 Pure., they appear to be identical and to belong to the genus Allothele. 

 Thus the specimens referred by me in Ann. S. Afr. Mus., vol. 17, 

 p. 120, to T. australis, now become Allotliele caffer. 



The $ 9 unfortunately appear to lack clearly denned specific 

 characters, and in the absence of the <$ <$ cannot with certainty be 

 separated into species ; size and markings vary to a certain extent, 

 but the relative proportions do not vary sufficiently or with the neces- 

 sary constancy to be specific. The eyes also vary, but only the ? ? 

 from Cookhouse (B 2402) seem at all distinct; in these specimens 

 the median eyes are very small, also the abdomen is necked (in alcohol) 

 with numerous lighter spots; provisionally they are classed as A. 

 australis, though with the advent of the $ they will most probably 

 prove to be a distinct species. 



The following species could only be separated from A. australis on 

 the <$ characters. 



ALLOTHELE TERETIS, n. sp. (Plate XXVIII, figs. 1 A-C). 



Specimens. 1 $,2 ? ?, and 2 juv. (4021, Types), M'fongosi,, 

 Zululand (W. E. Jones, 2/18). 



