354 NOVITATES ZOOLOGICAE XXIII. 1916. 



specimens from various Algerian localities from 10 to 12 mm. In the type-specimen 

 of codeti the forewing is 10 mm. long. 



Somabrachys appears to me to be a genns at which the causes of the variation 

 in size could be tested without much difficulty. As a rule, succulence and abundance 

 of food, and a moist atmosphere have a favourable influence ou the growth of 

 insects. Not only are the species of Somabrachys polyphagons, but the individual 

 larva will accept different plants — for instance change from Lychnis to a thistle. 

 The caterpillars are not at all rare in the environs of Alger, and as the female lays a 

 large number of eggs in clusters, sufficient material for feeding experiments should 

 easily be procurable. 



The colouring is much affected by flight and exposure. Most of our specimens 

 caught at the lamp are so worn that the wings have become semi-transparent and 

 the thorax has assumed a very pale brown tint. The veins usually contrast with 

 the interspaces as dark lines, which, however, are not so prominent as in Group B. 

 The deeper or paler colouring is a very unsafe guide in separating Groups A, 

 B and C, and I also cannot discover any structural difference between those 

 specimens of Group A which are more uniformly deep brown, or pale brown, or 

 dark-veined. 



The antennae have longer branches in the present group than in B and C, 

 which is especially noticeable if the last segments are compared. But the difference 

 is not evident enough to be of much diagnostic value. Moreover, the branches are 

 by no means constant in individuals of the same wing-size. 



The structure of the head is characteristic. The frons is broader in the female 

 than in the male, which is owing to the eyes of the female being narrower than 

 in the other sex, but bears the same protuberances in both sexes. The central 

 projection being very easy to see with a lens, it affords the best means of recognising 

 a female as belonging to the present group. The triangular or trilobate apical 

 surface of the frontal process is rarely concealed by the long hair of the head ; if it 

 should be the case, the hair need only be moved aside with the help of a pin 

 or small brush in order to bring the structure into view. On the denuded head we 

 observe three elevations between the antennae and the mouth-cavity (the month- 

 organs are absent or vestigial) (PI. VII, fig. 1). A low transverse ridge (C) 

 placed in front of the antennae is an excrescence of the anterior edge of the 

 suture (S) which separates the epicranium from the large sclerite which in Lepi- 

 doptera forms the face or frons and is homologous to the clipeus of other insects. 

 Centrally from this sharp and low ridge forward a transversely rounded-convex swelling 

 gradually rises to form a large truncate prominence (PI. VII, fig. 1, P.c.), the frontal 

 process. Below this process, i.e. towards the mouth, the frons still remains convex, 

 being raised above the frontal edge of the eye, and ends at the mouth with a second 

 projection, an enlargement of the anterior edge of the clipeus. This oral process 

 (PI. VII, fig. 1, P.O.) is a transverse narrow ridge which is almost vertical on the 

 plane of the frons, being slightly bent upwards. It nearly reaches across the frons, 

 being much broader than in Groups B and C, and is usually reversed cordiform, 

 being somewhat widened apically, with the apical margin sinuate or bidentate. 

 Fig. 2 represents a side view of the frons. 



These three structures are variable individually, the most conspicuous of them, 

 the frontal process, also presenting very strongly marked modifications in size and 

 outline. As a rule the sides of the triangle are much longer than the base, but 

 sometimes the surface of the process is equilateral, apart from the irregularities of 



