428 Annals of the South African Museum. 



Prothorax about one-half as long as the head, distinctly transverse 

 and shaped as in P. gracilis, Bagnall. Spine at each posterior angle 

 small ; posterior marginal pair minute, and two or three similar 

 setae at each anterior angle. Pterothorax broader than wide across 

 fore coxae, and one and three-quarter times the breadth of the 

 prothorax ; mesothorax strongly transverse, and the metathorax 

 with the sides strongly arcuate and narrowing to base of abdomen. 

 Wings reaching nearly to the apex of the seventh abdominal 

 segment, broad ; tinged with yellow and iridescent in a strong 

 light ; cilia shaded with brown. Legs with tibiae shorter and 

 stouter than is usual in the Idolothripid group ; fore femur slightly 

 thickened, inner edge straight and outer edge curved ; fore tibia 

 only slightly longer than the femur and fore tarsal tooth obsolete; 

 hind legs the longest, and with tibia decidedly longer than the 

 femur. All femora with a few inconspicuous spines and tibiae with 

 rows of fine setae ; hind and intermediate tibia with one or two 

 strong spines at tip within and also a few moderately long 

 bristles. 



Abdomen long, more than three-quarters the length of the whole 

 insect; segments subtransverse excepting the ninth, which is 

 elongate and cylindi'ical and much narrower than the preceding 

 segment. Tube exceptionally long and slender, curved upwards, 

 and gradually narrowed towards tip ; as long as the preceding five 

 segments together, at least twelve times as long as breadth at base, 

 seven times as long as the ninth segment, and more than two and 

 one-half times the length of head ; surface roughened, but smooth 

 near tip, sparingly setose, terminal hairs weak. Abdominal spines 

 stout, those on ninth segment only slightly longer than the others. 



P. caudatus may be easily recognised from P. gracilis, Bagnall, by 

 its colour and larger size, the subtransverse abdominal segments, 

 the much longer tube, the comparatively shorter spines of the ninth 

 abdominal segment, and many other distinctive characters. 



Ty2)e. A single dried and carded example, apparently female, in 

 the South African Museum, Cape Town. 



Hab. Knysna, Cape Colony. 



Note. — The specimen was obtained from among somewhat dry leaves of Curtisia 

 fraginea, forwarded by Mr. E. J. O'Connor of the Cape Forest Department, 

 in order to ascertain whether the numerous galls covering them were caused by 

 Cecydomiid flies, gall-making Hymenoptera or Aphidae, or simply by fungi. The 

 production of these galls was found not to be due to fungi ; and certainly not to the 

 action of terebrantious Hymenoptera. These galls might have been produced by 

 Aphidae, but the presence of Panurothrips caudatus leads one to conclude with 

 some amount of reason, that the galls were produced by the Thrips. 



