( Ixxxvii ) 



The tenth segment is apparently always symmetrical in the Sphingidae with 

 the exception of Si'siiiuie, where it is asymmetrical in a great number of species. 

 The tenth segment of the Sesiinae is normally divided into a right and a left 

 lobe, the tergite {Xt) as well as the stemite (A>) (PI. XLII. f. 1—25). la 

 most species of Ses/ime the tergite remains symmetrical ; in the highest members 

 of the subfamily it nndcrgoes, however, peculiar modifications. Tiie nearly 

 symmetrical ninth and tenth tergites of Haemorrhagia and Cephonodes (PI. XLII I. 

 f. 6. 21) become twisted (PI. XLIII. f. 8. 11), and the left half finally nearly 

 obliterates, being represented only by a broad semi-detached swelling {I) at the base 

 of the right process (left in figures), as shown in f. 15—18 of PI. XLIII. Tlie 

 stemite of Haemorrhagia venata and Cephonodes janus has two unequal processes 

 (Fl. XLIII. f. 12. 14. 20); the right one of them disappears, as more fully 

 explained on p. 401, there being scarcely a trace left of it (PI. XLIII. f. 25, r), 

 the sternite having a single process. This ventral process becomes shifted 

 towards the left side in Ceplwnotlcs, lying finally in a jdane with the tergite, 

 and forming with the latter a horizontal forceps (PI. XLIII. f. 8. 9). 



In Eni/o japix (PI. XLIV. f. 7) the sternite has become sinii)le, bnt the 

 left process has not entirely disai)peared. In Himantoides, also with a simple 

 tenth tergite, the sternite has jireserved a remnant of the aborted left process 

 (PL XLIV. f. 8), which rcinnant appears as a tubercle on the left side of the 

 sternite beyond the middle of the latter. The sternite of Epistor (PI. XLIV. 

 f. 10. 11) is simple bnt asymmetrical at the end ; it is doubtless a development 

 from a symmetrically divided sternite, bnt there is as yet no species known 

 which shows an intermediate stage. In Calliomma parce and Leucorkampha 

 ornatus the left process of the sternite is shorter than the right one, while in 

 PnchjUa remme.ns{?\. XLII. f. 0) the riglit one is a little longer than the left. 

 In all these cases the sternite is in advance of the tergite in the asymmetrical 

 reduction. 



The right and left har[)es are conspicuously different in several St'sii/iae. 

 Pachylia darceta (PI. XLV. f. 20. 27) has on the left harjie a large curved 

 process, which is represented on the right harpe by a small tooth only. In 

 Aleiiron neglectum and Epistor luguhris we find the right harpe produced into 

 a long conical, elbowed process (PI. XLVI. f. 13), while the left harpe is very 

 short (PI. XLVI. f. 12). 'Differences which are not quite so conspicuous are 

 observed in the allied species and genera. Sometimes the claspers are obviously 

 different right and left — as, for instance, in some Epistor. The most remarkable 

 cases are again found in the genera Ilncmorrhai/iit and Ccplioiwdes. The latter 

 genus has no free harpes ; the right and left claspers are always different in 

 size and shape (PI. LI. f. 22—25 ; PI. LII. f. 1—3), the contrast being strongest 

 in C. trochilus. In all Cephnnodi's, as well as in all llaemorrhigia, the left 

 (right in figures) clasper is reduced or modified. The right and left harpes of 

 Haemorrhagia venata are jjractically the same (PI. LII. f. 5) ; in the other 

 species the left is always shorter than the right, the latter being often long 

 and club-shaped, whereas the left one is vestigial (PI. LI. f. 17—21 ; PI. LII. f. 4). 



