TRACHEAL GILLS OF ADULT INSECTS 



477 



The adult Pteronarcys is nocturnal, flying only at dewfall or in the night, and 

 Mr. Barnston observed it when on the wing, " constantly dipping on the surface 

 of the water" ; by day it hides "in crevices of rocks which are constantly 

 wetted by the spray of falling water, under stones and in other damp places." 

 It may thus be compared with the Amphibians, Necturus and Proteus, whose 

 gills are retained in adult life. A similar large Chilian Perlid (Diamphipnoa 

 lichenalis Gerst.) differs in completely lacking the thoracic gills, though there 

 are four pairs on the abdomen, i.e. a pair on each of the first four segments. In 

 this form the number of individual filaments in the largest tufts may amount to 

 about 200. 



Another Perlid (Dictyopteryx signata) is said by Hagen to have two pairs of 

 gill-tufts on the under side of the head ; the first pair situated on the base of the 

 submeatum, the second on the membrane connecting the 

 head and prosternum. 



Kolbe states that in the imagines of Perla maryinata 

 and P. cephalotes on the hinder edge of the thoracic stig- 

 mata arise three very small chitinous plates, which, on their 

 under side and on the edges are beset with numerous short 

 white filaments. These completely correspond to the fila- 

 ments of the tuft-like larval gills. Persistent anal gills also 

 occur in the imagines of Perla. 



In Nemoura lateralis and cinerea the tracheal gills are 

 A differently disposed. On each side 



of the anterior edge of the proster- 

 num arise delicate tightly twisted 

 filaments, like those of the larva. 

 (Einfuhrung, p. 536.) 



Hagen also states that in the 



* 1 



FIG. 457. Under side of Pteronarcys regalia, showing the situation of the gills (g, b,f) and 

 the sternal orifices : A, a branchial filament showing the direction of the current of blood ; c, d, 

 trachese. , end of the abdomen enlarged. After Newport. 



dragon-fly, Euphaja, the gills of the nymphs are retained in the imago, and 

 Palnie'n remarks that in ^Eschna the rectal gills of the nymph persist in the 

 imago, though not used for respiration. 



Palme'n gives an instance of a caddis-fly (Hydropsyche, Fig. 448) retaining its 

 gills through the imago stage, but they are unfit for respiration, as they are 

 minute and shrunken. 



A walking-stick (Prisopus flabelliformis') found in the mountains of Brazil 

 has the remarkable habit, according to Murray, of spending "the whole of the 



