130 CHAS. K. BRAIN. 



pores wanting ; dorsal pores not numerous or conspicuous, but with very long (one- 

 half width of pygidium) internal tubes ; ventral thickenings of integument extending 

 from median lobes practically wanting." 



This species was recorded " on thick, narrow, slightly oblanceolate leaves about 

 1| inches long, collected by Mr. C. P. Lounsbury, at Mitchell's Pass, S.A., Jan. 29, 

 1897." This species has not been observed again nor has any of the original material 

 been found in the collection. I have left this species in the sub-genus Hemiberlesea, 

 as described, for the present, but I think, from the illustration, that it should be in 

 Aspidiotus s. str. 



Sub-genus Selenaspidus, Ckll. 



The chief distinguishing characters of this sub-genus may be tabulated as follows : — 



Scale of adult $ generally large, flat, +. circular, with central exuviae, most 

 commonly brown in colour. 



Puparium of $ elongate, brown, with exuviae placed towards one end. 



Adult $ in all species (except S. silvaticus, Lindinger) with the cephalothorax 

 distinctly separated at the margins from the abdomen. Pygidium with three pairs 

 of lobes and numerous coinb-like plates. The median two pairs of lobes are usually 

 large and + tri-lobed. The outer (3rd) pair of lobes are normal and nearly as large 

 as the others in S. lounsburyi ; normal but smaller in S. schultzei ; reduced to a small 

 pointed projection in S. euphorbiae, Newst., and strong, + tusk-like and pointed 

 in the other South African species. The scales of S. lounsburyi and S. griqua, sp. n., 

 are unusual for this sub-genus. In the former it is + buff-coloured at first, moderately 

 convex, and usually bleached in old specimens. In S. griqua the scale is very convex 

 and the hardened, chitinous body of the female is also highly convex. 



The position of the scales on the host-plant, the age of the specimens, and the 

 condition of humidity and sunlight probably influence the scales of the species of 

 Selenaspidus in a more marked manner than in any other known Diaspine Coccids. 



Circumgenital glands are absent in the majority of species known, but there are 

 two groups present in S. articulatus and pertusus, sp. n., and four groups in S. celastri. 

 At maturity the chitin becomes very dense, and the characters of the pygidium are 

 often more or less obscured, and in some cases the body-wall has the appearance 

 of being perforated by large, more or less rounded holes. These perforations are 

 often arranged in definite order, but could never be mistaken for the tessellated 

 designs found in Pseudaonidia spp. 



The species known to occur in South Africa may be distinguished as follows : — 

 A. Cephalothorax distinctly separated at the margins from the abdomen. 

 I. Circumgenital glands absent. 



(1) 3rd lobes normal, not much smaller than 1 and 2 (fig. 100) lounsburyi (Marlatt). 



(2) 3rd lobes normal but much smaller than 1 and 2 (fig. 101) . . schultzei, Newst. 



(3) 3rd lobe obsolete, not represented by lobe or definite tusk (fig. 102) 



euphorbiae, Newst. 



(4) 3rd lobe strong, tusk-like ; insect small, flat, highly chitinous (fig. 107) 



pumilus, sp. n. 



(5) 3rd lobe strong, tusk-like ; insect convex, conical (fig. 105) griqua, sp. n. 



