and on the Number of Abdominal Segments, in Insects. 181 



and in which the tenth bears the anal appendages — namely, the 

 LibellulcE. In these insects what is generally counted as the first 

 abdominal segment is, however, a posterior part of the nieta- 

 thorax — separated, it is true, by a deep incision and a softer 

 membrane capable of some extension*, but not by a complete 

 articulation, from the anterior portion of it. There are two 

 reasons which seem to settle this point beyond doubt. In the 

 first place, this apparent segment is destitute of spiracles, the 

 first of the seven conspicuous pairs of spiracles lying in the con- 

 nective membrane of the dorsal and ventral arcus of the apparently 

 second segment f. A first abdominal segment destitute of spira- 

 cles is without any analogy in the Insecta, while, on the con- 

 trary, it is a characteristic feature of the metathorax that it never 

 has spiracles. A second argument in favour of this opinion is, 

 that the apparently first segment is developed, during the meta- 

 morphosis of Libellula, in proportion to the growth of the wings. 

 In the young larva we count only nine abdominal segments, of 

 which the ninth bears the anal plates ; in a larva of moderate 

 size which has already undergone some moultings, and where 

 the future wings appear on the back of the thoracic segments, 

 that part begins to show itself; even in the pseudopupa, previous 

 to its last transformation, it is still but little developed ; and it 

 is only in the perfect insect that it completely assumes the shape 

 of an abdominal segment. The metathorax there attains an 

 unusual development, but a development quite in proportion to 

 the posterior wings, which are, in the Libellula, of even a larger 

 size than the anterior ones, and which are moved during flight by a 

 system of proper muscles ; while in other orders, as, for instance, 

 in Hymenoptera, they are moved by the same system of muscles 

 as the anterior ones, being connected to them, during flight, by 

 the hooks of their anterior margin. 



According to this view, the penis of the Libellula is situated, 

 not on the second, as usually stated, but on the first ventral 

 segment ; and the outlets of the generative organs are, in both 

 sexes, on the eighth, and not on the ninth, ventral segment — 

 those of the male in the middle, those of the female at the base 

 of it. In this group the posterior part of the eiglith and the 

 whole ninth ventral segment intervene between the vulva and 

 the anus, the latter occupying, as usual, the end of the ninth 

 segment, which bears the anal appendages ; and in this group, 

 also, the nine ventral half-segments are all developed. 



Having thus ascertained that in insects the abdomen is as 



* A similar soft membrane exists between the anterior and posterior 

 portions of the mesothorax (scutellum and postscutelluui) in the Hyraeno- 

 pterous genus Eupelmus. 



t Hagen, Stett. Entom. Zeit. 1853, p. 319. 



