244 JAMES WATERSTON. 



Ny AS ALAND : Port Herald, bred from pupa of a Lymantriid moth {Dr. J. E. S. Old). 



Type, a 2 in the British Museum, one of a series of 1 (J, 20 $ 5- 



Several species closely related to T. balteaUis have recently been described by 

 Prof. Silvestri, viz. : gijfardi (1913), oxyurus (1913), giffardianus (1915) and dacicida 

 (1915). Silvestri's species are all parasitic on fruit-flies {Dacus, etc.). From 

 gijfardi, giffardianus, and dacicida, halteahis (J differs in the chaetotaxy of the 

 funicle, in the relative length of the first funicular joint, in the size of the scapal 

 sense-organ and its fringing bristles. With giffardi I have been able to make direct 

 comparison (a series from Aburi ; identification confirmed by Prof. Silvestri) and 

 there appear to be numerous minor differences between the species. The $ of 

 balteatus may likewise be known by the first funicular, the radius, etc. Of the 

 African Tetrastichus, balteatus probably stands nearest to dacicida, but the 

 proportions of the hind tarsal joints appear to be different. In going over the 

 collection in the British Museum an even closer ally was found in Tetrastichus 

 tachos, Wlk. (Ann. Mag. Nat. Hist, ii, p. 352, 1839), from England. 



A note on the antennal characters of T. tachos may be given. 



Tetrastichus tachos, Walk. ^.- — Antennae (fig. 7 a) : length over 1 mm. ; scape 

 with sense-organ covering four -fifths of the ventral edge ; eight bristles fringing 

 the sense-organ ; outside bare, inside with sparse, evenly set bristles which tend 

 to form a regular subventral row, especially near the apex. Each funicular joint bears 

 near the base a whorl of tubular bristles (8-9) on the outer and dorsal aspects ; the 

 first pair of joints equal and slightly shorter than the second pair, which are also 

 equal. The sensoria are fewer in number and shorter, and with longer flanges than 

 in balteatus. Win^s. There are more bristles on the radius than in balteatus 

 <figs. 7 6, 8 6). 



Syntomosphyrum phaeosoma, Wtrst. (figs. 7, 8). 



Since describing this species (Bull. Ent. Res., v, p. 370, 1915) I have seen 1 ^ and 

 5 5 $ bred from the cotton leaf-roller {Sylepta derogata), at Dedza, Nyasaland, by 

 Mr. E. Ballard. In these examples the first two joints of the funicle are a little 

 lighter than in the type (bred from the same host, in N. Nigeria), but the agreement 

 is otherwise complete. I take this opportunity of describing the species in more 

 detail and of noting the differences between it and S. glossinae, Wtrst. Both species 

 have the mesonotum densely pilose on the mid lobe, while the first of the scatellar 

 bristles occurs before one-half from the suture. There is a short, stout bristle 

 immediately above the spiracular emargination on the pronotum, and the 

 submarginal bears three bristles. The propodeal spiracle lies inside the 

 notopleural ridge. 



^. Antennae pale, clear yellowish brown from scape to nearly end of funicle ; 

 last funicular joint obscured and club darker.* Head and thorax non-metallic 

 purplish black, a little shining ; eyes reddish chocolate. Wings clear, veins very 

 light brown. Fore coxae entirely, and mid coxae (above), concolorous with thorax ; 

 hind coxae transparent, light brown ; legs elsewhere yellowish brown — at most 

 the fore femora with a smoky streak. Abdomen with the peduncle narrowly 



* In tlie type the 1st and 2nd joints are also smoky. 



