INSECTA TRANSVAALIENSIA. 87 



forming a fork with Ih at the base ; \c ahsent ; vein 5 from the middle of discocelhdars, or ritrehj from 

 just below upper anrjle of cell* Hind wing with ttuo internal veins; vein 5 from the centre of the 

 discocelhdars or rarely absent; 8 free from the base, curved, and running close along the stibcostal 

 fiervurc, or joined to it by a bar." 



Packard, in dealing with the Nearctic Notodontidfe, remarks, "the larvae of this family 

 vary greatly in form and ornamentation for a group of such moderate numbers " ; and in a 

 synopsis he enumerates fifteen distinct larval characteristics.! The same impression has 

 been stated by Weismann : "In fact, in the whole order Lepidoptera there can scarcely be 

 found associated together such diverse larvse as are here placed in one imago-family." I 



As regards the food-plants of the South African species, there is still much to be 

 recorded. According to Packard, from what is known of the life-histories of the Californian 

 and Campestrian Notodontians, their prhicipal food-plants in that region are the poplars and 

 willows found along the river-courses of that dry area ; others feeding on the scrub oaks of 

 the plains and foot-hills. 



Edw. Newman proposed a tribe, or division, which equalled this family, under the name 

 Cuspidates "(in science Cuspidatce), from the singular form of the caterpillars, which 

 frequently end in a sharp point"; but, as he further remarks, "in this, as in many other 

 instances, the name of the division does not strictly apply to all of the contents of the 

 division." § Some have gone under the name of " Pvomineuts," owing to the larvae being 

 usually without prolegs on the hindmost segment, which is thus carried erect. 



There had long been an opinion that some of the grotesque attitudes of these 

 larvs possessed a protective signification. Under the genus Cerura (p. 92) we have 

 recorded the observation of a distinguished entomologist, that those larvae represent the 

 Puff-Adder among the Bombycine caterpillars, as the larvae of Chcerocampa do among those 

 of the Sphingida;. Hermann Midler has made similar suggestions as to the larvae of 

 another well-known European Notodont, Stauropus fagi. When sitting on a twig m its 

 natural position, the head and first five segments are held erect, and the greatly lengthened 

 legs of the second and third segments held outstretched ; thus, when seen from the front, 

 the whole aspect of the insect, both in form and colour, is most Spider-like, and when 

 alarmed it immediately raises its four long legs and moves them irregularly, after the 

 manner of a Spider attacking its victim. This Spider-like appearance is further suggested 

 as a special protection against ichneumons, which, according to the experience of 

 H. Miiller, are particularly afraid of Spiders, while Fleddermann, an experienced breeder 

 of insects, never found the larva of 5'. fagi to be attacked by ichneumons. When observed 

 from the rear, there is nothing to be seen but the erect, hard, shield-like surface of the 

 last segment surmounted by two black horns, and presenting an appearance totally unhke 

 that of a caterpillar.il Miiller even found a further suggestion, when observing on the side 

 view of the larvae a small depression just below the spiracles on the fourth and fifth segments, 

 which gives the appearance of a caterpdlar which has been stung by an ichneumon. This 

 suggestion has been adopted by Poulton as "another form of mimetic resemblance— the 



* Except ill CypUanta, an Indian genus. I Mem. Nat. Acad. Sci. vol. vii. p. 23 (1895). 



' Studies in the Theory of Descent,' Eng. transl. p. 443. § ' British Moths,' p. 204. 



' Kosmos,' Nov. 1879, p. 123, and Proc. Ent. Soc. Lond. 1880, p. 111. 



t '' 



