SOUTH AFRICAN BAGWORMS. 161 



as yet very imperfectly kuown, and some careful breeding 

 remains to be done to fill up the gaps. 



Systematic ^Position. — The species belongs to the 

 family Psychidse, siib-fam. Psychinte, and was described 

 by Janse (loc. cit.) under the name of Acanthopsyche 

 alba. The type was captured at light — a very unusual 

 occurrence — and it is the only species of this group which I 

 have ever noticed to fly so late as to come under the influence 

 of light attraction. 



The type and other specimens known to Janse were males; 

 the female has not been described. 



Food-plants. — "The larva feeds on acacia or wattle or 

 both" (E. E. Piatt, in lit.). Mr. Piatt and the writer found 

 the species abundant on a thorn tree (Acacia sp.) near the 

 Race-course at Durban, and I also found it attached to the 

 trunk of an Eucalyptus in Mr. Todd's garden at Phoenix, 

 Natal. In the latter case there were no native acacias near, 

 but the garden was surrounded by wattle plantations, and as 

 it is very improbable that the bagworm feeds on Eucalyptus, 

 we must infer that it had fed on the surrounding wattles. 

 However, in captivity the specimens obtained from the thorn 

 tree at Durban refused to eat wattle leaves ; neither did the 

 first instar larvaj take to them. We may therefore conclude 

 that the species is very particular about its food-plant, and an 

 individual larva, or group of larvas from the same brood, 

 does not readily make a change. On the other hand, there are 

 probably some races which have adapted themselves to 

 wattle and other plants, and a brood probably chooses only 

 those food-plants on which the parents throve. This lack of 

 adaptability on the part of the young larva may explain the 

 relative scarcity of the species. 



Of other food-plants the following may be mentioned : in 

 October, 1917, the writer found isolated specimens, practically 

 full-grown, on Cupressus macro car pa and on privet 

 (L i g u s t r u m v u 1 g a r e) , both commonly used for hedges. In 

 neither case was there any leguminous tree or shrub in the 

 immediate neighbourhood from which the larv« might have 

 VOL. 4, PART 1. 11 



