432 Prof. J. O. Westwood's 



the insect, arrived at its perfect state witliin its cocoon, 

 has to bite off the end of the latter with its strong jaws ; 

 but in the Trichoptera a totally different and exceptional 

 process is effected, in consequence of the pupa being 

 furnished with a pair of horny mandibles, quite unlike 

 those of the larva state, and which are thrown off on 

 assuming the winged state in which the mandibles are 

 quite rudimentary.* With these mandibles, which vary 

 in form and strength in the different groups according to 

 the nature of the cocoons,t the pupa cuts its way through 

 the net-work mouth of its case shortly before assuming 

 the perfect state; at which period it is stated in my 

 Introduction, 1. c, p. 69, that "it assumes considerable 

 activity, swimming along, as I have observed, by means 

 of its two hind legs, which are strongly ciliated, and 

 crawling about by means of its four fore legs, which 

 become detached from the breast." I have, unfortunately, 

 no recollection of the species which I observed acting in 

 the manner thus described ; \ but in the joupa communi- 

 cated to me by Mr. Poulton, it is the middle pair of legs 

 which are fringed with long strong hairs, and which the 

 insect uses in swimming exactly in the same manner as 

 the Notonectce use their hind legs ; and M. Pictet also 

 states (" Phryganeides," p. 53) that it is the middle pair 

 of legs which are furnished with " deux rangees symme- 

 triques de poils forts et serres, disposes comme des barbes 

 de plumes; la nymphe se sert de ces pattes comme d'avi- 

 rons, pour nager lorsqu'elle veut eclore ; ces poils restent 

 a la depouille et n'existent point dans I'insecte parfait." 



* In my Introduction, 1. c, I have said that the mandibles of the pupa 

 are different in form from those of the larva and pupa, the latter word 

 having been inadvertently printed instead of imago. 



f The pupa before us agrees, in the simple structure of the mandibles, 

 with that of Sericostoma (Pictet Phr}'ganides, pi. 4, fig. 3). According to 

 Mr. M'Lachlan, " these mandibles appear to be stronger and moi-e notched 

 in those families of caddice flies that make fixed cases of stones, such as 

 the Hydropsycliides and BliyacopliiUd.es. In the latter there is an inner 

 cocoon to be cut through as well as the ordinary case or grating ; in all 

 others the pupa lies free in the case." 



I The following is a copy from my original note on the subject: — 

 " 10th May, 1829. Observed several specimens of a pupa in a pond on 

 Coombe wild, which at first appeared like the pupaa of large gnats. They 

 were swimming slowly about, seeking for some twig to creep up, and the 

 motion was performed by means of their two hind legs, the tarsi and tibiae 

 being finely haired. On taking them out of the water I perceived a slit 

 down the front of the back of the thorax, and after a little while (after 

 creeping about by means of their four front legs) I perceived that they 

 were Trichopterous pupas, which must have left their cases to change 

 to the winged state. The tail was furnished with two small sette." 



