38 MEAIOIKS OF THE NATIONAL ACADEMY OF SCIENCES. 



Cremastochry.sali.s. Melaiioi-erii, uiul Cirinu. It is present in S ('i//iii[/)//ir i-inut. hut alisent in 9 

 C. her'dla. In the Ileniileucida- it is usualh' absent, thouoh present in Coloradia. In Satuniiiila' 

 it is wautinfj, or verj' short, arisinj;- at the apex, and ahnost vestigial. 



It appears, then, that in the SphingidiO this vein, either atrophied or vestigial in the super- 

 family of Saturniides, has become revived, restored, and strengthened, and functions as one of 

 the important veins in the wing. Its presence is correlated with the narrowness of the whole 

 wing, the acuteness of the ape.x, and consequent greater strength of the fore wing, adaj^ting 

 the moth for swift, powerful flight. It is to be observed that the bonib^'cine character of but 

 three branches of the so called, or what was formerU' the median vein (now, according to 

 Enderlein. niedius 3, cubitus 1 and 2), are the same as in the superfamily Saturniides, while 

 the '"independent vein" (medius 2) is detached as in the saturniide, but this vein arises nearer 

 medius 3 than medius 1. in this respect differing from the position of this vein in Ceratocampiiue. 

 The anterior discal vein of the Smerinthiiuv also differs in its direction from that of the Cerato- 

 campinaj, being directed inward instead of outward, toward the origin of the hinder discal; this 

 produces a change in the shape of the discal cell, the outer side of the discal cell being parallel 

 with the outer edge of the wing instead of being at right angles to it, as in Ceratocampiniv. (See 

 PI. XLIII.) 



The forking of the base of the axillary vein also occurs in the CeratocampiniC. 



The hind wings of the Sphingidie differ from those of Saturniides in tkose characters which 

 seem to strengthen the wing; the bristle is present, so that both wings are locked together, and 

 an additional axillary (^4.,) is added, while the subcostal and radius is strengthened by a cross vein 

 arising from near the middle of the discal cell and anastomosing with the subcostal vein (II). 



Turning to the head with its appendages, we find that in the primitive Smerinthinai 

 there seem to be no positive diagnostic characters which separate them from the Ceratocampina*. 

 The latter, like all the Saturniida', have a large, long, broad, .somewhat triangular or sciitellate 

 clypeus, extending up and inclosing the antennal foramina. As will be seen by reference to 

 PI. XXXVI, tigs. G, 7, the .shape and proportions of the front of the head of the two Smerinthine 

 genera Marumba and Cressonia are nearly identical with those of the Ceratocampimv, as Eacles, 

 etc., except that the head is wider in front in the Sphinges. In the typical Sphinges with long 

 maxillre the head is much larger, the fi-ont more convex, probably owing to the enlarged tongue 

 and its nmscles. 



The antenna* of the typical Sphingidie are, as is well known, of a peculiar prismatic fusiform 

 shape, in the Macroglossina? with a terminal hook and no pectinations, but in Cressonia, which 

 has evidently retained the vestigial characters of its ceratocampid ancestors, with their l)roadly 

 pectinated antenna% the joints are doubly pectinate, or with two pairs to a joint, the pectinations 

 of the anterior pair being a little shorter than those of the basal pair. (PI. XXXVI, tigs. s. S^^) 



The maxilhv of Cressonia, though scarcely "obsolete," as usually stated, t)cing long enough 

 to form a roll when retracted, arc but little longer than those of Eacles and Citheronia. It is a 

 very striking fact that while in the Smerinthinie, which appear to us to be primitive rather than 

 degenerate forms, the maxillw are so small as to be of little or at least very limited use. It 

 should be carefully ot)served whether an\- of the Smei'inthina? extend their maxilhe and probe the 

 corollas of flowers, like the typical Sphinges, or if they use the tongue to simply suck the sweets 

 of flowers while resting on their leaves or petals. It has been observed that Citheronia re(/<i//s 

 will be attracted ))y and sip the sugar laid as bait on trees. Its tongue is short and feeble, that of 

 Eacles a little longer, but in neither case extending as far as the end of the palpi. 



When wo conii)are the small size and lack of development and use of the maxillie of Cres- 

 sonia with those of Phlegethontius, which attain a length of over 3 inches (SS mm.), gi-eater than 

 that of the entire Ijody, or that of the South American sphinx {J/acro.sila chientius), whose 

 tongue is said by Wallace to be 9^ inches in length," we see that within the lilnits of a .single 

 family an organ like the spiral tongue may by frequent exei'cise be greatly enlarged and other- 



« One from tropical Africa, Marriiniid (Xanlhopan), morgani, ia 7i inche.s long, according to Wallace (Natnral 

 Selection, p. 146), though Rothschild and .Jordan state that it is about 225 mm., equal to 8 inches. Revision of the 

 Ijepidopterous Family Sphingida;, p. 32. 



