146 BRITISH LEPIDOPTERA. 



Mill., "Cat. Lep. Alp.-Mar.," p. 118 (1872); Cum y Mart., "Cat. Lep. Bare," 

 p. 39 (1874); Curd, "Bull. Soc. Ent. Ital.," vii., p. ill (1875); Butl., "Trans. 

 Zool. Soc. Lond.," ix., pt. 10, p. 568 (1876); Kirby, "Eur. Butts, and Moths," p. 

 71 (1879) ; "Cat.," p. 664(1892); Frev, "Lep. Schweiz," p. 57 (1880); Weism., 

 "Stud. Theory Descent," transl. p. 201 (1882); Buck., "Larv.," ii., p. 42, pi. 

 xxv., fig. 1 (1887); Auriv., " Nord. Fjar.," p. 46 (1889); Mina-Pal., "Nat. Sicil.," 

 vii., p. "134 (1888) ; Barr., " Lep. Brit.," ii., p. 46, pi. 50 (1895) ; Meyr., " Handbk.," 

 p. 296 (1895); Lucas, " Brit. Hawk Moths," p. 99 (1895); Tutt, "Brit. Moths," 

 p. 27 (1896); Leech, "Trans. Ent. Soc. Lond.," p. 285 (1898); Bartel, "Pal. 

 Gross-Schmett.," ii., p. 98 (1900). Phrynus (by error for Phryxus), Stephs., 

 "List Br. An. Brit. Mus.," p. 28 (1850). Dilephila, Hamps., '" Moths of 

 India," i., p. 98 (1892); Kirby, "Handbook, &c," iv., p. 27 (1897). 



The genus is diagnosed ( Verzeichniss, &c., p. 137) by Hiibner 

 as follows : 



The forewings almost entirely white-nervured ; the body on the back and on 

 the sides chequered with black and white — Phryxus livornica, Pet.* {lineata 

 Fab.), P. caicus, Cram. 



In 1835, Stephens placed this species in his "Abstract of the 

 Indigenous Lepidoptera contained in the Verzeichniss Bekannter 

 Schmetterlinge" (Illus. Haust., app. p. 5) as the British exponent of 

 Hiibner's Phryxus. In 1850, he placed it in the same genus (List of 

 Specimens of Brit. Animals in Coll. of Brit. Museum, pt. v), but the 

 name is, by a printer's error, misspelt Phrynus. The genus has 

 been diagnosed (in litt.) at length by Kaye as follows : 



Forewing triangular, pointed sharply at apex ; exterior margin very slightly 

 convex, not indented. Cell extending half the length of wing, nervures 7, 8 on a long 

 stalk, 6 from upper angle. Cross-nervure very oblique and slightly curved ; nervure 5 

 slightly nearer 4 than 6; nervure 2 two-fifths from base ; lb greatly curved up after 

 leaving base and returning to extreme angle at tornus. Hindwing rather short and 

 broad, apex very bluntly pointed ; exterior margin recurved and bluntly pointed at 

 anal angle ; cell scarcely extending more than one-third length of wing ; subcostal 

 spur emitted at less than half the length of cell ; 6 and 7 from upper angle, 6 almost 

 straight. Discocellular flattened, S-shaped, with both curves about equal ; nervure 

 3 from close to 4, 2 from half the length of cell. Palpi blunt. Antenna; with 

 pectinations very weak, consisting only of a few hairs— Phryxus livornica. 



Weismann shows (Studies in Theory of Descent, p. 200) that 

 livornica and lineata (referred by Staudinger, Cat., 3rd ed., p. 103, 

 to the same species) fall in separate groups when considered from 

 the point of view of larval development. From this standpoint he 

 differentiates them as having : 



Open ring-spots f appearing on the subdorsal line on all the segments from 

 the nth to the 1st - lineata. 



Closed ring-spots situated on the subdorsal line -livornica. 

 Later, however, he shows (loc. ci/., p. 358) that he considers the 

 larva of lineata to be an ancestral stage on the same line of descent 

 as that of livornica, the former being the more generalised, the 

 latter the more specialised, member of the same genus. The pupa 

 of P. livornica differs from the other Phryxid pupae known to us 

 in being much more like that of Hippotion than they are. It has 

 a certain amount of antero-posterior flattening, whilst the other Phryxids 

 are fairly round ; it also shows some S curve owing to ventral lengthen- 

 ing of abdominal segment 4, the others being fairly straight ; the 

 maxillary keel is long and thin, extending round front of head, 



* This was the Accipitrina livornica of Petiver, 1702. The name, however, 

 has no standing on Petiver's authority, being antecedent to 1758, the earliest date 

 accepted for binomial nomenclature. 



f Fernald's description fpostea, p. 150) suggests that certain forms of this larva 

 may reach or even surpass that of livornica iu the development of its ling-spots. 



