The antennae of the Perlidae are long, very flexible, and composed of a very large number of joints. The parts of the mouth vary a good deal. The mandibles and maxillae are usually rather small, and all the parts of the mouth are of feeble consistence or even membranous; the maxillary palpi are, however, well developed and exserted from the mouth, five-jointed. The labium is short and but little conspicuous. The mandibles in some forms are almost membranous, but in other genera they are firmer and are toothed. The labium is composed of a very large mentum, beyond which is a large piece, usually undivided, bearing the four terminal lobes; the three-jointed palpus is seated on the side of the large middle sclerite, which is no doubt of composite nature. Considerable variety as to the lower lip prevails. The head is broad and flat; there is an indistinctly-indicated clypeus, three—more rarely two—ocelli, and on each side an eye neither very large nor perfect. The prothorax is free, and has a flat, margined notum. The meso- and the meta-thorax are large, equal segments. The pro-, meso-, and meta-sternum are large pieces; between the first and second, and between the second and third there is an intervening membrane. The metasternum is much prolonged backwards, and has on each side a peculiar slit; similar orifices exist on the other sterna (Fig. 254, o). Newport, who has examined them in Pteronarcys, says that they are blind invaginations of the integument; he calls them the sternal or furcal orifices.[[321]] According to this naturalist these very peculiar openings pass into the body "as strong bone-like tubes, diverging from the axis to the periphery of the body in the immediate vicinity of some of the principal tracheae, but that they do not in any way communicate with them, as they terminate abruptly as caecal structures." He thinks them analogous with the endo-skeleton of other Insects; a view which cannot be considered sufficiently established. Laboulbène states[[322]] that when Perla parisina is seized and placed on its back, it does not move, but emits a liquid at the base of the articulation of the legs. This suggests that it may come from these sternal orifices. The abdomen consists of ten dorsal plates, the first being short, and of nine ventral; the dorsal plates are much more ample transversely than the ventral. Frequently the hind body is terminated by two long, many-jointed cerci, looking like antennae. The coxae are small, not prominent, and are directed outwards. The legs are slender, the tibiae often grooved. The tarsi are three-jointed, terminating in two claws and a more or less distinct pad. In the genus Isopteryx an auditory organ has been described as existing in the legs, in a position similar to that of the analogous structures in Termitidae and Blattidae. The wings when closed repose flat on the back, and fold and overlap so that only one is seen (Fig. 251); in this state the costal portion of each front wing is turned downwards, so as to protect to some extent, the sides of the body.
Fig. 251.—Perla maxima. (After Pictet.)
Fig. 252.—Perla sp., nymph, showing tracheal gills. Pyrénées orientales.
The early stages are known, but have not been described minutely, and there appears to be very little information as to the youngest life. All the species are, when immature, aquatic in their habits; the larvae greatly resemble the perfect Insects in form, though differing in not possessing wings and in the ocelli being merely opaque spaces. They have rather large compound eyes; the future wings are represented by lobe-like prolongations—varying in length according to age—of the meso- and meta-notum. In the Nemourae the cerci are absent in the imago though present in the young. The larvae of Perlidae are carnivorous and are able to swim well, the legs being provided with abundant swimming hairs; they, however, as a rule, prefer to walk at the bottom of the pool, or on rocks or boulders in the water they live in.
One of the most peculiar features of the Perlidae is their respiratory system. Unfortunately the greatest differences of opinion have prevailed on various matters in connexion with this subject, and there are several points about which it is not possible at present to express a decided opinion.
Fig. 253.—Tracheal gill and portion of a trachea of Pteronarcys. (After Newport.)
The larvae have no stigmata; it appears to be generally agreed that there is in them no means of admitting air to the tracheal system by means of orifices. Some breathe entirely through the integument, the process being aided by the accumulation of tracheae at the spots where the breathing orifices should be, and where the integument is more delicate. Others, however, possess gills in the form of protruded bunches of filaments, connected with tracheae in the manner shown in Fig. 253. These filamentous branchiae occur in numerous species of the family, and are situate on various parts of the body, but many species are destitute of them in genera, other members of which possess the filaments. In some Nemourae instead of bunches of filaments there are tubular projections on the prothoracic segment; and in Dictyopteryx signata similar structures occur even in the cephalic region, Hagen stating[[323]] that there exists a pair on the submentum and another on the membrane between the head and the thorax. In the imago state, stigmata are present in the normal fashion, there being two thoracic and six abdominal pairs. In several species the filaments persist in the imago, so that in these cases we meet with the curious condition of the coexistence of branchiae with a well-developed and functionally active system of spiracles; this is the more curious because the creatures usually have then nothing to do with the water, it having been ascertained that in these cases the species live out of the water as other terrestrial and aerial Insects do. These instances of persistence of branchiae during the aerial life have been the source of some perplexity; the condition was shown to exist in Pteronarcys by Newport, and has since been demonstrated in various other forms. Newport believed that the imago of Pteronarcys breathes by means of the gills, although it lives out of the water and possesses spiracles; and he informs us that Mr. Barnston observed the Insect when on the wing "constantly dipping on the surface of the water." Hence Newport concluded that Pteronarcys in the winged state is "an amphibious animal." That a winged Insect should live in the air and yet breathe by means of gills would be truly extraordinary, and there can be little doubt that Newport's idea was erroneous. Hagen[[324]] was able to examine living imagos of the species in question. He found that they avoided the water, and though he placed some individuals therein, yet they did not use the gills. He also informs us that the branchiae have, during life, a shrivelled appearance, indicating that they are not functionally active, but are merely useless organs carried over to the imago from the previous instar, in which they were truly the means of obtaining air. Hagen also ascertained that the spiracles of the imago are in a normal state, being adapted for breathing, even as far back as the seventh abdominal segment.