SPHINGIDES. • 363 



manner. The two modes of development produce, in the adult 

 larvae of Aglia and Sphinx, a somewhat superficially similar adult 

 animal. The only existent larva that we know that even very distantly 

 approaches (but certainly does not satisfy) the possible ancestral 

 Sphingid form is that of Bombyx nwri, where we have i, ii, 

 iii, iv and v in generalised position, occurring side by side with 

 a wart - like condition of i and iii in the early stadia, a 

 highly developed condition of i on 8th abdominal, and swollen 

 thoracic segments. The Sphingid egg, larva and pupa, all show 

 characters that prove that Sphingids were never developed either 

 through Lachneids, Dimorphids, Attacids or Bombycids, per se, 

 but, retaining some of the characters of each, their ontogeny 

 abundantly proves that they have had a common origin with these. 

 The impossibility of direct derivation from either of these groups is still 

 further intensified by the imaginal structures — antennae, maxillae, 

 and frenulum. The special form of the Sphingid antenna closes 

 the superfamily absolutely, and, unless one is willing to admit a 

 series of discontinuous variations and redevelopment of lost organs 

 as exemplified by the frenulum and maxillae, this direct derivation 

 is obviously impossible. Poulton's position of attempting to prove 

 that Sphinx has been evolved from Aglia through Ceratomia 

 and Smerinthus, appears altogether untenable. Here we get an 

 attempt to derive the most specialised Sphingids from the most 

 specialised Attacids, whilst the generalised Sphingids (Eumorphids, 

 etc.) and generalised Attacids (Hemileucids), where structure, 

 if not adult appearance, would be expected to approach somewhat, 

 are not brought into consideration. Bacot's paper on the subject (Ent. 

 Record. , vii., pp. 227 — 230, 246— 248) should be referred to as there are 

 many interesting facts therein. Among others he notes the following 

 resemblances between Dimorpha versicolora and the Sphingides as : 



Larva. — First skin. — (1) Caudal horn only faintly suggests that of the 

 Sphingides. It more nearlv represents that of Eutricha qnercifolia. Second 

 skin. — (2) Change in shape of head : S?nerinthus (pcellata). (3) Dark medio- 

 dorsal line : Mimas tiliae (in second skin). (4) Oblique stripes : as in Sesia (Hemaris) 

 (mentioned by Poulton) . (5) Strength of the 7th oblique as compared with the others : 

 Smerinthus. In further support I would point out that the oblique stripes are 

 caused (in D. versicolora) chiefly by the absence of the shagreen tubercles on the 

 stripes, and though in Smerinthus the reverse is the case with the stripes, the 

 borders, at a certain stage, are caused in the same way (i.e., by the absence of 

 the shagreen dots). (6) The fact that the lateral lines are coloured, while the 

 oblique are due at first chiefly to structural arrangements : Smerinthus. (7) 

 Small, thickly scattered, shagreen tubercles, each bearing one slightly bifid hair : 

 Smerinthus and Sphinx. Third skin. — (%) Anal flap outlined with yellow: 

 Smerinthus. (9) Darker bordering to oblique stripes : Smerinthus (? stage). 

 (10) Ventral and anal prolegs developed much after the manner of Smerinthus 

 (mentioned by Buckler). Last stage. — (11) Retractile head when fullfed : 

 Eumorpha eipenor (mentioned by Buckler). (12) Change of colour before 

 pupation : Sphinx ligustri and Mimas tiliae (mentioned by Buckler). (13) Only four 

 stages in ontogeny (three moults) (mentioned by Buckler) : Amorpha populi. 



Pupa. — (14) Shape and general aspect approaches M. tiliae. (15) Cremaster 

 (in some characters): M. tiliae. (16) Movement in cocoon ; the pupa forces its 

 way partly out : Eumorpha eipenor (Tutt and Buckler). (17) Spines on free and 

 following segments : E. eipenor. 



Imago. — (18) Shape of wings (roughly ouly) : Smerinthus. (19) Scheme of 

 coloration (roughly only) : Smerinthus. (20) Discoidal lunule on ibrewings ( ? 

 too general to be of much use). 



The peculiarities and apparent differences may be tabulated 

 as follows : 



