152 PROF. W. H. JACKSON ON THE 



somites of the larval T'. lo have each, as is usual, a pair of spiracles, the ninth has none, 

 the tenth bears the anal valve and anal prolegs. The distinctness of the ninth somite 

 from tlie tenth was hinted at by Lyonet. The posterior ring, he says, appears " a la 

 verite souvent, comme ici, compose de deux ; mais pour conserver I'uniformite, il con- 

 vient de ne le considerer alors comme un seul anneau subdivise" (' Traite Auat. de la 

 Chenille qui ronge le bois de saule,' 17G2, p. 21 of tlie "description anatomique "). 

 Later on, in treating of the musculature, he speaks of the last ring, his ninth, as con- 

 sisting of an anterior and posterior part. His figures demonstrate the separateness of 

 the two portions. Sir John Lubbock assigns nine somites to the abdomen of Pygcera 

 {Phalera) hucephala, as is clear from jip. 183, 184 of his memoir relating to the muscles 

 of the caterpillar of that moth (Trans. Linn. Soc. xxii. part iii. 1858), but it does not 

 appear that he regarded the tenth as a separately existing somite. A ghmce at 

 PL XVI. fig. 29 proves at once that the ninth somite in Vanessa lo is completely 

 separated from the tenth, and that it has bundles of muscles, fewer in number than the 

 eighth, but strictly conformed to the same type. The latter statement would not be 

 exactly true of the tenth somite. Attention may also be drawn to the existence on the 

 ninth segment of this caterpillar of a latcro-dorsal spine, and a tubercle with a small 

 hair-like spine, representing the second row of spines of preceding segments ; see 

 PI. XVI. fig. 21 and its description. W. Miillcr has drawn attention to a similar 

 fact in the caterpillar of Acnjea pellenea as showing the double character of the so-called 

 twelfth (/. e. ninth abdominal) segment (Spengel's Zool. Jahrbiicher, i. 1886, pj:). 529, 530). 

 In the pupal condition (PL XVI. figs. 22, 28) the same number of somites may be 

 traced as in the caterpillar, and it is a curious fact that Kirby and Spence (Introduction, 

 iii. 1826, p. 251) assign ten somites to the pupa, though not to the caterpillar. The two 

 figures of pupfB here given are contrasted in certain respects. Pig. 22 is that of a pupa 

 killed at the moment the susj)ended larva had begun to cast its cuticle, the operation 

 being completed afterwards by dissection. It is slender, and its somites very distinctly 

 separated from one another. Pig. 23 is that of a chrysalid more than one day old ; it 

 shows a stouter abdomen. The intersegmental membranes, whicli are of considerable 

 extent in this species, ai'e completely folded inwards and hidden ; the rings of the somites 

 are thvis lirought into close contact. This contraction in length and the greater 

 stoutness of body are caused by the excessive shortening of the muscles soon after the 

 assumption of the pupa state. The points to be noted in l)oth figures alike are (1) the 

 presence of eight pairs of spiracles, the first pair hidden by the wings in fig. 23, as is 

 generally the case, whilst the last or eighth pair are, as usual, abortive, a fact evident on 

 comparing PL XVI. figs. 27 and 26 with fig. 28 and the description given*; (2) the 

 perfect distinctness of the ninth somite ; (3) the replacement of the anal valve by the 

 cremaster (cr.) ; and (4) the formation of the sustentor ridges {s.r.) and sustentors of 

 Riley. The distinctness of the ninth somite in the pujia is also shown in the side 



* Heroic! states that the male pupa of J'icrls hnissiae lias eight abdominal somites, the female seven or doubtfully 

 eight ; that both sexes have alike six stigmata, the first somite always being devoid of one. He is in error in all 

 these statements. See his ' Entwickeluugsgoschiehte dcr Schmctlerlinge,' pp. 34, 00, (i'J and note, 84, and p. xxiii 

 of description of plates. 



