196 Annals of the South African Museum. 



beyond mid-length of the sector ; r on .R., just beyond the fork ; cell 

 \st Mo small, about as long 1 as the basal deflection of Cu^, which is 

 inserted at the fork of M. 



Abdominal tergites pale brown, sternites yellow. Ovipositor with 

 the tergal valves very slender, almost acicular, subfleshy, delicately 

 pubescent, slightly upcurved before their tips ; sternal valves pale, 

 very flattened and compressed, approximately as long as the tergal 

 valves. 



Habitat. South Africa. 



Holotype, $, Krantzkop, Natal, November, 1917 (K. H. Barnard). 



Paratopotype, 9 



Type in the South African Museum. 



GEN. PLATYLIMNOBIA, Alexander. 

 PLATYLIMNOBIA PUMILA, sp. n. 



Size small (male, length 2'6 mm.) ; general coloration brownish-black 

 including the hypopygiurn ; antennae with but nine flagellar segments, 

 the last more elongated than the others. 



Male, Length, 2'6 mm. ; hind leg, femur 2-8 mm., tibia 3"2 mm. 



Eostrum and palpi dark. Antennae black ; scapal segments large ; 

 flagellar segments (Plate IV, fig. 22) nine in number, oval-cylindrical, 

 rather crowded, the terminal segment a little narrowed, elongate, about 

 as long as the two preceding taken together. Head with the eyes small, 

 widely separated ; ommatidia very coarse. Head dark, sparsely dusted. 



Thorax dark, dusted with grey. Pleura paler. Legs with the coxae 

 and trochauters rather light yellowish-brown, especially the posterior 

 pair ; remainder of the legs black. 



Abdomen dark brownish -black ; hypopygium large, black. 



Habitat. South Africa. 



Holotype, $ , Hottentot-Hollands Mts., Caledon, Cape Colony, 1917 

 (K. H. Barnard). 



Type in the South African Museum. 



This interesting species is undoubtedly related to the genotype, 

 Platylimnobia barnardi, Alexander, but is only about one-half as large 

 (in barnardi the middle leg shows the following measurements : femur 

 6'4mm., tibia 6 mm.) and almost entirely blackish in colour. The 

 antennae are but 11-segmented. In the material of P. barnardi I 

 believed that I could distinguish 16 segments, there being 14 flagellar 

 segments ; the material has since been returned to the South African 

 Museum and I cannot confirm the statement at this time, but would 

 indicate the possibility of a mistaken observation. P. barnardi has the 



