Front matter (continued 2)

Family Acroceridæ.

Acrocera longirostris, n.s. (Plate VII., fig. 4).

An extraordinary and very rare species, occurring amongst white rata^[10] blossoms in February. At present I have only taken three specimens, viz., two in Wellington and one in Nelson. The transformations of all the Acroceridæ are as yet unknown.

Family Syrphidæ.

Syrphus ortas (Plate VII., fig. 3, 3a larva, 3b pupa).

Very common everywhere from September till May, or even later, when specimens may be often seen basking in the winter sunshine. The larva (3a) is a most useful insect to gardeners as it destroys an immense number of aphides, those noxious little insects that commit such fearful ravages on many valuable plants (see Hemiptera, page 120). In general appearance this larva resembles a small green slug, with the skin much wrinkled, and bearing at its extremity a short thick tube, which is probably the respiratory apparatus, the four lunate holes situated at its apex being no doubt the spiracles. These insects grow very slowly, occupying several weeks to attain maturity. Their mode of capturing the aphides is very curious, and is, briefly, as follows:—The larva lies in the midst of a number of aphides, and it occasionally happens that some of them crawl over it. On feeling an aphis touch its back the larva instantly darts out its long, pointed head and strikes its prey with the apex, which is enveloped in a quantity of very sticky mucus constantly ejected from the mouth. On the aphis being thus captured the larva withdraws its head into the hinder segments of its body and devours all the juicy portions of the aphis, whose dry skin is afterwards thrown aside. When full-grown it slowly shrinks up and changes into the pupa shown at Fig. 3b. In this state it is not protected by any kind of cocoon, but lies amongst the refuse of the aphides, near the stem of the plant. The fly emerges in a fortnight or three weeks' time, and is very fond of hovering over and sucking honey from the flowers, but the females may be often noticed running about plants, probably in search of a suitable place to oviposit.^[11] For an account of Scolobates varipes, a species parasitic on the present insect, I refer to page 39.

Family Syrphidæ.

Eristalis cingulatus (Plate VII., fig. 2).

This conspicuous insect occurs occasionally in glades in the forest about January, but is by no means common. It is very fond of the white rata flowers, where it may be taken, if anywhere. Its life-history is at present unknown, but no doubt resembles that of the following insect.

Family Syrphidæ.

Helophilus trilineatus (Plate VII., fig. 1, 1a larva, 1b pupa).

This fine species occurs abundantly in all damp situations throughout the summer. Its larva may be found in stagnant pools and is often met with in the mud at the bottom of ditches. Its posterior segments are enormously elongated, forming a telescopic breathing apparatus, composed of two tubes, the smaller of which is capable of being more or less extended at the will of the larva, which is thus enabled to adjust the length of its breathing tube, according to the depth of water or mud in which it happens to reside. This peculiarity has given all these larvæ the name of rat-tailed maggots. The other segments are very stout, each being furnished with a pair of minute feet, and the head is also provided with two small appendages which are supposed to be the outlets through which the exhausted air is discharged by the larva. When mature this insect leaves the water, forming a small oval cell in the neighbouring moist earth, in which it lies with its long tail folded along the breast. The skin then gradually hardens, and it is finally transformed into the pupa shown at Fig. 1b, the conical pair of breathing-tubes on the thorax being slowly protruded from two hardly perceptible warts, whilst the telescopic apparatus shrinks up, its functions being at an end. A variable time, dependent upon the season, elapses before the perfect insect makes its appearance, but prior to this occurring, a large circular plate, forming the thorax of the pupa, is thrust off, thus assisting the escape of the fly, which immediately ascends a plant, or other convenient object, to dry and expand its wings (Fig. 1). In the perfect state it delights to hover in the air, darting away with great rapidity on the approach of any enemies. It also frequently enters houses, where its presence is at once betrayed by a peculiarly shrill noise made while flying. The sexes of this insect differ chiefly in size, the female (Fig. 1) being about twice as large as her companion.

Closely allied to this species are Helophilus ineptus, and H. hochstetteri. The former is slightly smaller than H. trilineatus and may be at once distinguished by its tessellated orange-yellow and black abdomen. It is rather local, but extremely abundant wherever found. The latter has a superficial resemblance to some of the smaller blowflies (Musca), but may be readily known by its large brownish-red scutellum.^[12] It is the commonest of the genus and may be found in great numbers throughout the summer amongst veronica and other flowers.

Family Muscidæ.

Miltogramma mestor (?) (Plate VII, fig. 5).

A conspicuous species, found occasionally on forest-clad hills round Wellington. The life-history is at present unknown, but its larva is very possibly parasitic in some large Lepidoptera.

Family Muscidæ.

Nemorea nyctemerianus (Plate VII., fig. 6).

This little fly is seldom met with in the perfect state. Its larva is parasitic on the caterpillar of Nyctemera annulata^[13], the eggs being deposited on the moth larva at an early age. The caterpillar grows and eats in the ordinary way, until it has assumed the chrysalis state, when the maggot eats its way out and changes into a dark-brown pupa. In this condition the parasite is protected by the web which was previously constructed by the unfortunate caterpillar for its own use. The perfect fly appears in about six weeks' time, its great agility and large white scales rendering it very conspicuous.

Family Muscidæ.

Eurigaster marginatus (Plate VII., fig. 7).

Another parasitic species, its larva inhabiting the caterpillars of various noctuæ which it destroys just before they change into the chrysalis state. The pupa of the parasite lies in a small oval cell constructed in the earth by its larva. A variable number of these maggots are found associated in one host, the smaller caterpillars only harbouring a single individual, while a large larva will frequently contain three or four. This species has been bred from the following Lepidoptera: Mamestra composita, M. ustistriga and M. mutans. It also occurs in the curious Œceticus omnivorus, being found in the cocoons of that moth in numbers varying from two to eleven, or even more, and it is especially interesting, as it is in turn destroyed by a small species of Pteromalus already noticed among the Hymenoptera (page 37). The perfect insect occurs occasionally on flowers throughout the summer.

Family Muscidæ.

Calliphora quadrimaculata (Plate VII., fig. 9).

This is the large blue-bottle fly of New Zealand and is found everywhere in great abundance. Its larva feeds on decaying flesh and is of a dirty yellow colour, measuring, when full-grown, about seven lines in length. The pupa is buried at a considerable depth in the ground, the larva having descended before changing. The duration of this, and in fact of all the stages of the insect, depends entirely upon the temperature, but the females invariably deposit eggs, even during the hottest weather, and are never ovo-viviparous like the next species, and several others of the genus.

Family Muscidæ.

Sarcophaga læmica (Plate VII., fig. 10).

Another extremely abundant species having a similar history to the last, but its powers of development are very much accelerated owing to the larva being positively born alive. The females hover over meat and other suitable substances, depositing a number of minute wriggling maggots thereon, not infrequently to the great disgust of some hungry individual, who perhaps is making his dinner off a mutton chop which the fly has selected as a home for her offspring. These larvæ are all produced from distinct ova, which hatch before being laid, as I have often proved, by removing them from the insect's abdomen, and watching the young larva emerge from a minute elliptical white egg, covered with a thin leathery skin. Every one who has travelled in New Zealand must have noticed that, in the wildest spots, these insects assemble in large numbers as soon as any meat is uncovered, thus not only showing their universal distribution throughout the country, but also that they possess a very keen sense of smell.

Two British species at least, allied to this genus, have been introduced into New Zealand, viz., Musca domestica and Musca cæsar. The former is probably a world-wide insect, every ship teeming with it, but the latter is at present rather scarce and is usually found in the neighbourhood of farm-yards, where the larva feeds on cow-dung. The perfect insect may be at once known by its brilliant green colour.

Family Muscidæ.

Cylindria sigma (Plate VII., fig. 14).

A curious species, occurring occasionally in damp situations in the forest where it may be noticed leisurely walking over the leaves of various shrubs. It is very sluggish and may often be captured between the fingers without the aid of a net. Its life-history is at present unknown, but the larva probably feeds on fungi. The pretty little insect depicted at Fig. 11 may be found in similar situations but is not so common.

Family Muscidæ.

Phora omnivora, n.s. (Plate VII., fig. 15, 15a pupa).

This minute species may be found in large numbers nearly all the year round. Its larva is parasitic on a great variety of insects and is also not infrequently met with among decaying vegetable matter. Its habits are, therefore, very varied. When parasitic in the Lepidoptera it usually selects the noctuæ, destroying a great number of many of the commoner species^[14]. The infected caterpillars usually turn into chrysalides some time before the little maggots emerge, but this is not invariably the case, the parasite often destroying the larva at a comparatively early stage. The pupæ are buried in the earth, near the remains of their host, and are light brown in colour, with the segments much more distinct than is usual (Fig. 15a). From these the perfect flies proceed in about a month's time. The occurrence of this insect as a parasite in Coleoptera is not common, but I know of one instance in which a number of these little flies were produced from a pupa of Uloma tenebrionides (Plate II., Figs. 2, 2a, 2b), which I was rearing at the time (page 29). In this case it is difficult to understand how the female contrives to deposit her eggs in a horny beetle larva which lies safely hidden in its narrow tunnel in the middle of a large log of wood. Among bees this is a most destructive insect, its larva being parasitic in their grubs, and thus greatly reducing the population of the hive, which is finally ruined by the wholesale destruction of its honey when the flies emerge. Driving the bees into a fresh box would, no doubt, be frequently beneficial in these cases, but it is to be feared that bee-keepers will have much difficulty in contending with this insect. Its sexes are readily distinguished by their size, the female being considerably the larger.

Family Muscidæ.

Coelopa littoralis (Plate VII., fig. 13).

Extremely abundant on the sea-beach. Its larva feeds on decaying seaweed, burying itself in the sand before changing. The perfect insects often congregate in such vast numbers on some of the rocks that it is necessary to run past them in order to avoid being positively suffocated by the countless multitudes which fly up into one's face. This insect must be regarded as the New Zealand representative of the well-known dungfly of England (S. stercoraria), which many of my readers will recollect has a similar habit of assembling in great numbers.

Family Œstridæ.

Œstrus perplexus, n.s. (Plate VII., fig. 12).

This species is mentioned here as it is the only New Zealand exponent of a very important and well-known family of Dipterous insects. I am at present quite ignorant as to its life-history which would, no doubt, be very interesting. The only two specimens I possess were taken at Nelson, some four years back, so that it appears to be very rare.

The two remaining groups of the Diptera are of very limited extent. The Pupipara include a few anomalous species, in which the young are not deposited until they become pupæ, thus undergoing all their transformations within the body of the parent, while the Pulicina comprise the well-known fleas, which are probably identical with the European species. They are placed by many authors in a distinct order termed the Aphaniptera.

CHAPTER VI

This Order includes the well-known Butterflies and Moths which are the first insects to arrest attention on account of their beautiful colouring and conspicuous appearance. Some of the families are fairly numerous in New Zealand, but the diurnal section is decidedly poorly represented, our total number of butterflies being limited to fifteen, of which one (Diadema nerina) has unquestionably been introduced from Australia, although it will doubtless shortly effect a permanent settlement in the Nelson district, where several specimens have recently been observed. Among the others only four species can be called at all common, the remaining twelve only occurring in certain favoured localities. Of the moths there are a large number, chiefly belonging to the Geometridæ and Micro-Lepidoptera, many of which are very interesting. Of the life-histories of the latter, however, I regret to say there is little known at present, the attention of naturalists having been hitherto chiefly occupied with the larger and more conspicuous species.

Group Rhopalocera.

Family Nymphalidæ.

Argyrophenga antipodum (Plate VIII., fig. 1 type, 1a var.).

Passing over the local but conspicuous Danais plexippus, about which so much doubt exists as to its origin in this country, we come to A. antipodum, one of the most curious and interesting butterflies found in New Zealand. It occurs in great abundance amongst the tussock grass on the plains in the South Island, but becomes an alpine species further north. I have taken a very peculiar form (Fig. 1a) on the "Mineral Belt" near Nelson, but can find no record of its appearance in the North Island at present. Its larva is as yet unknown, but in all probability it feeds on tussock grass, a fractured pupa having been found attached to that plant by Mr. G. F. Mathew in January, 1884. Two other closely allied species are Erebia pluto and Erebia butleri, both strictly alpine insects, occurring in the South Island at elevations ranging from 4,000 to 6,000 feet.

Family Nymphalidæ.

Vanessa gonerilla^[15] (Plate VIII., fig. 2, 2a underside, 2b 2c larvæ, 2d 2e pupæ).

One of our most beautiful butterflies, found abundantly throughout the country from August till May. The larva feeds on the New Zealand nettle, where it may be taken in great plenty by careful searching. The caterpillar joins several of the leaves together and forms a sort of tent, in which it lives secure from all enemies. While young, these insects are of a uniform dull brown colour, with two faint lines on each side, but as age advances they become very variable. The two extreme forms of variation are depicted at Figs. 2b and 2c, the dark-coloured variety being by far the commoner. When full-grown, this larva suspends itself by the tail to a small patch of silk, which it has previously spun on the under side of a leaf. In this position it remains for about twenty hours, when it begins to twist and distend the lower portions of its body, thus causing the skin to eventually break on the back of the thoracic segments, when the soft green pupa may be seen through the rent. The insect now works the skin upwards by violent wriggling motions until it is gathered in a crumpled mass round its tail, the old rent extending on one side almost up to the silken pad to which it is suspended. Through this rent the tail of the pupa is brought and firmly anchored in the silk by a few vigorous strokes, the insect hanging meanwhile to the skin which has not been quite cast off on the reverse side to the rent. When thus firmly attached to the silken pad, the pupa shakes itself entirely free, whirling itself round and round until the old skin is dislodged from the silk and falls to the ground. The two usual varieties of pupæ are shown at Figs. 2d and 2e, many of them being more or less ornamented with metallic gold or silver spots. The butterfly emerges in a fortnight or three weeks, and is common from February till April in most situations, but the greatest numbers are to be found in the spring months. These hybernated specimens appear as early as August, and some of them survive till the end of December or beginning of January, when the earliest of the new ones are just emerging. In fact it is not infrequent at this time to take both hybernated and recent specimens together. This species is a great traveller, and may be often seen flying over the tops of the trees at a great rate. It shows a singular indifference to shadow, and is constantly flying out of the sunlight into shady places in the forest, probably in search of the food-plant of the larvæ. The two other species of Vanessa are V. cardui, a periodical insect only distinguished from the "Painted Lady Butterfly" of England by the blue centres in three of the black spots on its hind-wings, and V. Itea, a lovely butterfly found in the northern portions of this island, of which I have at present only taken three specimens.

Family Lycænidæ.

Chrysophanus salustius (Plate VIII., fig. 3 ♂, 3a ♀, 3b larva).

This is the commonest of our Butterflies, and is found in great abundance throughout both islands from November till April. It is double brooded, and is consequently most abundant in the early summer and in the autumn, few of these merry little insects being seen at midsummer. The most forward individuals of the second brood usually emerge about the middle of March, but the butterflies are very irregular in their appearance at this season. The young larva (Fig. 3b) is much thickened anteriorly, the head being concealed from above by the large thoracic segments. Its colour is pale green, with a pair of long, erect bristles on each segment, a large number of shorter ones being situated on the ventral surface, and behind the head. After the second moult, a brilliant crimson dorsal line is noticeable, but beyond this I have no record, as my larvæ unfortunately died just after completing their third moult. Up to this time they had fed but sparingly on the dock, eating minute holes in the leaves and clinging to them with great firmness. It is much to be regretted that their subsequent history could not be followed, especially as I only succeeded in obtaining the eggs on this one occasion, although I frequently kept females in captivity with this object. Three other species of Chrysophanus occur in New Zealand, viz., C. feredayi, common round Nelson, and chiefly distinguished by the olive-green under-surface of its hind-wings; C. enysii, which is occasionally met with amongst forest, and may be at once known by its broad black markings and pale yellow colour; and C. boldenarum, a little insect uniting the "Coppers" with the "Blue Butterflies," and found in great abundance in certain river beds and shingly places. The western side of Lake Wairarapa is one of the best localities I know of for this curious little species.

Family Lycænidæ.

Lycæna phœbe.

This is the common blue butterfly of New Zealand, which may be observed in great numbers along the roadside on a hot summer's day. Its larva must be very abundant, but has hitherto escaped attention, owing, probably, to its small size. The perfect insect is on the wing from October till May.

Group Heterocera.

Family Sphingidæ.

This family is represented in New Zealand by the splendid Sphinx convolvuli, an insect I am at present unacquainted with.

Family Hepialidæ.

Porina signata (Plate IX., fig. 2).

Common throughout the summer, when it may be taken in great numbers round lighted windows during any mild evening. The larva is as yet unknown, but is in all probability subterranean in its habits, and feeds on the roots of plants. A large Hepialus larva I once discovered under a stone, whilst looking for Coleoptera, was very likely referable to this insect, but as it unfortunately died shortly afterwards it is impossible to speak with any degree of certainty at present. Two closely allied species are P. umbraculata, and P. cervinata. The former is rather smaller than P. signata and of a more uniform brown, with a white stripe in the centre of each fore-wing, surrounded with darker colouring. The latter is one of the smallest of the family, its size at once distinguishing it from any of the rest. In colour it is pale brownish with numerous black and white markings, varieties occasionally occurring much suffused with the darker colour. It is rather local, but may be found abundantly in the Manawatu district.

Family Hepialidæ.

Hepialus virescens (Plate IX., fig. 1 ♂, 1a ♀, 1c larva, 1b pupa).

This gigantic insect is seen occasionally in the forest during the early summer. The larva (1c) tunnels the stems of living trees, feeding entirely on wood which it bites off with its strong mandibles. The plant most usually selected by the caterpillar is Aristotelia racemosa, called by the settlers "New Zealand currant," from its large clusters of rich-looking black berries, which appear in autumn. Other food-plants are numerous, the black maire (Olea apetala) and manuka (Leptospermum) being among those more frequently chosen.

This larva, for the most part, inhabits the main stem of the tree, its gallery always having an outlet to the air, which is covered with a curtain of dull brown silk, spun exactly level with the surrounding bark, and consequently very inconspicuous. These burrows usually run down towards the ground, and are mostly two or three inches from the surface of the trunk. In some instances the larvæ inhabit branches, in which case, if the branch is of small dimensions, the tunnel is made near the centre. These remarks only refer to galleries constructed by young larvæ, as the tunnel made by the insect prior to becoming a pupa is of a very complicated character and merits a somewhat detailed description. It consists of a spacious, irregular, but shallow cavity, just under the bark, having a large opening to the air, which is entirely covered with a thin silken covering, almost exactly the same shape and size as the numerous scars which occur at intervals on the trunks of nearly all the trees. Three large tunnels open into this shallow cavity: one in the centre, which runs right into the middle of the stem, and one on each side, which run right and left just under the bark. These are usually very short, but sometimes extend half-way round the tree, and occasionally even join one another on the opposite side. The central tunnel has a slightly upward direction for a short distance inwards, which effectually prevents it from becoming flooded with water; afterwards it pursues an almost horizontal course until it reaches the centre of the tree when it appears to suddenly terminate. This, however, is not the case, for, if the gallery floor is carefully examined a short distance before its apparent termination, a round trap-door will be found, compactly constructed of very hard, smooth silk, and corresponding so closely with the surrounding portion of the tunnel that it almost escapes detection. When this lid is lifted a long perpendicular shaft is disclosed which runs down the middle of the tree to a depth of 14 or 16 inches, and is about six lines in diameter. At the bottom of this the elongated pupa (Fig. 1b) sleeps quietly and securely in an upright position, the old larval skin forming a soft support for the terminal segment of the pupa to rest on. The upper end of this vertical shaft is lined with silk, which forms a framework on which the trap-door rests when closed. The lid itself is of a larger size than the orifice which it covers, and this makes it extremely difficult, if not impossible, to force it from the outside, whilst it fits down so closely to the aperture as not to be readily lifted. The object of this most ingenious contrivance is, in all probability, to prevent the ingress of insects, large numbers of spiders, slugs, and various Orthoptera being frequently found in both central and lateral tunnels, but they are quite unable to pass the trap-door. The galleries of different individual larvæ are all wonderfully alike, the only differences observable being in the length of the perpendicular shaft and the direction of the horizontal burrow, which is sometimes curved. These variations are usually caused by the presence of other tunnels in the tree, which the larva invariably avoids, although how it can ascertain that it is approaching another tunnel before actually reaching it, is hard to understand. As development progresses in the pupa, it becomes darker in colour, especially on the wing-cases, which in some individuals show the future black markings of the moth, as early as two months before emergence. Others remain quite white and soft, the green wings suddenly appearing through their cases a fortnight or three weeks prior to the bursting forth of the imago. Previous to this change the pupa works its way up the vertical tunnel, lifts the trap-door, which yields to the slightest pressure from within, and wriggles along the horizontal burrow until it reaches the air, the last three or four segments only remaining in the tree. The thoracic shield then ruptures, and the moth crawls out and expands its wings in the ordinary way, resting on the trunk of the tree until they are of sufficient strength and hardness for flight.

The perfect insect, although it must be common, is very rarely seen. It is best reared from the pupæ, which can be often successfully cut out of their burrows and kept amongst damp moss until they emerge. It appears to be much persecuted by birds, as we often observe its large green wings lying about on the ground.^[16]

The curious "vegetable caterpillar," which is usually referred to this species, probably belongs to one of the larger subterranean larvæ of the family.

Family Bombycidæ.

Nyctemera annulata (Plate IX., fig. 3 ♂, 3a larva, 3b pupa).

This abundant species is usually mistaken for a butterfly by the uninitiated owing to its diurnal habits and conspicuous colouring. Its larva feeds on various plants, the most usual being a light green kind of ivy with yellow flowers, but its original food no doubt consisted of the "New Zealand groundsel" (Senecio bellidioides), on which it may now be occasionally taken in wild situations. Its general colour is black, with interrupted dorsal and lateral lines, the ventral surface and connecting membrane between the segments being slate-coloured. In younger larvæ there are also several slate-coloured lines extending the whole length of the insect, and thus dividing the black into squares. Round the middle of each segment, at its greatest circumference, a variable number of brilliant blue warts are situated, and out of these dense tufts of long black hair take their rise. There are, however, no warts along the ventral surface. This description applies very well as a rule, but the larva is subject to many slight variations. It remains in this state for nearly three months, or more, according to the season, and is very common, numbers being found on the different plants which constitute its food. The pupa (Fig. 3b) is of a shining black colour, with many longitudinal rows of small yellow blotches on the abdominal segments; there is also a stripe of the same colour at the tip of the wing-case. It is enclosed in a slight cocoon, formed of a mixture of silk and hair, and is attached near the ground to any firm object. The moth emerges in the course of a month or six weeks. It is very common, being found profusely in the neighbourhood of its food-plants, and appears in the greatest numbers during the early morning hours in the middle of summer.

For an account of a Dipterous insect, parasitic in the present species, I refer to page 59.

Family Psychidæ.

Œceticus omnivorus (Plate X., fig. 1 ♂, 1a ♀, 1b larva, 1c ♂ pupa).

This insect is very rarely seen abroad, but can be easily reared from the larva, which feeds on manuka and other plants throughout the year. When very young, and in fact immediately after leaving the egg, it constructs a wide spindle-shaped case, principally composed of silk, with a few small fragments of leaves, &c., attached to the outside. It has a large aperture in front, through which the head and anterior portion of the larva are projected, and a much smaller one at the posterior extremity, which allows the pellets of excrement to fall out of the case as they are evacuated. The body of the enclosed caterpillar is of a light straw colour, the head and three first segments being dark brown, with numerous white markings. The abdominal segments are considerably thickened near the middle of the insect, rudimentary prolegs being present on the third, fourth, fifth, and sixth segments of the abdomen. The anal prolegs are very strong, and are furnished with numerous sharp hooklets, which retain the larva very firmly in its case. As it grows it increases the length of its domicile from the anterior, causing it gradually to assume a more tubular form, tapering towards the posterior aperture, which is enlarged from time to time. The outside is covered with numerous fragmentary leaves and twigs of various sizes, placed longitudinally on the case, and frequently near the anterior aperture, the materials, owing to their recent selection, are fresh and green. The interior is lined with soft, smooth silk of a light brown colour, the thickness of the whole fabric being about the same as that of an ordinary kid glove, and so strong that it is impossible to tear it, or indeed to cut it, except with sharp instruments. The size of the case when the caterpillar is mature varies considerably, ranging from 25 to 30 lines or more in length, and about three in diameter, the widest portion being a little behind the anterior aperture (see Fig. 1b).

During the day the larva closes the entrance and spins a loop of very strong silk over a twig, the ends being joined to the upper edges of the case on each side; in this way it hangs suspended, the caterpillar lying snugly within. I have often known a larva to remain thus for over three weeks without moving, and afterwards resume feeding as before; this probably occurs while the inmate is engaged in changing its skin. At night the larvæ may be seen busily engaged: they project the head and first four segments of the body beyond the case, and walk about with considerable rapidity, often lowering themselves by means of silken threads; the only locomotive organs are, of course, their strong thoracic legs, which appear to easily fulfil their double function of moving both larva and case. If disturbed, these insects at once retreat into their cases closing the anterior aperture with a silken cord which is kept in readiness for the purpose, and pulled from the inside by the retreating larva. This operation is most rapidly performed, as the upper edges of the case are flexible, and thus fold closely together, completely obstructing the entrance. When full fed, this caterpillar fastens its case to a branch with a loop of strong silk, which is drawn very tight, preventing the case from swinging when the plant is moved by the wind, and also rendering the insect's habitation more inconspicuous, by causing it to resemble a broken twig. The anterior aperture is completely closed, the loose edges being drawn together and fastened like a bag. The posterior end of the case is twisted up for some little distance above the extremity, thus completely closing the opening there situated. It is lined inside with a layer of very soft silk, spun loosely over the sides, and partly filling up each end. In the centre of this the pupa lies with its head towards the lower portion of the case, the old larval skin being thrust backwards amongst the loose silk above the chrysalis. In this stage of existence the extraordinary sexual disparities, which are so characteristic of the family, manifest themselves, the male and female pupæ being very widely different in all respects. The former is figured at 1c, the female pupa differing from it in the following particulars. It is much larger and more cylindrical in shape, the abdomen occupying nearly the whole of the body, and consisting of nine visible segments, the terminal one being obtusely conical. The head and thorax are very rudimentary, more resembling those of the larva than the male, all the appendages being, however, reduced to hardly visible warts. In colour it is pitchy black and shining, and its length is about ten lines. This insect remains in the pupa state during the winter months, viz., from May till September. When about to emerge, the male chrysalis works its way down to the lower end of the case, forces open the old aperture there, and projects the head, thorax, and upper portion of the abdomen, the pupa being secured from falling by the spines on its posterior segments, which retain a firm hold in the silk. Its anterior portion then ruptures, and the moth makes its escape, clinging to the outside of its old habitation, and drying its wings. It is probable that the female insect does not leave her case, communication with the male being no doubt effected through one of the orifices, and the eggs afterwards deposited inside. On one occasion I found a case full of eggs, containing the shrivelled body of the female and her old pupa shell, which would seem to confirm the above opinion. The perfect insects are drawn at Figs. 1 and 1a. The male (1) is extremely active, dashing about the breeding cage with great rapidity when first emerged, and rapidly beating his wings to tatters; but the female (1a) closely resembles a large maggot, all the appendages being completely rudimentary, except the two-jointed ovipositor at the end of her body; she is incapable of any motion, except a slight twirling of the abdomen, which takes place while the eggs are being laid.^[17]

Family Noctuidæ.

Leucania nullifera (Plate XIII., fig. 3, 3a larva).

This large, though dull-looking insect, is occasionally taken at light during the summer and autumn months.

The larva feeds on the spear-grass (Aciphylla squarrosa), an abundant plant on the coast hills near Wellington. It devours the soft central-growing point, and its presence in a tussock can be at once seen by a quantity of pale-brown "frass," visible at the bases of the leaves. The formidable spear-like points with which this plant is armed must afford the caterpillar considerable protection from enemies. As a rule a single specimen only is found in each clump of the grass, so that the female probably deposits her eggs singly. This larva is full-grown about August, and may be found feeding in the plants during the autumn and winter.

The pupa state is spent, in an earthen chamber, amongst the roots of the spear-grass, and the moth emerges during the summer.

This species occurs at considerable elevations. I have seen it as high as 4,000 feet in the Nelson province, where its food-plant may also be found.

Family Noctuidæ.

Leucania atristriga (Plate X., fig. 2).

Abundant among various blossoms during the latter end of summer, being one of the last of the Noctuæ to disappear in the autumn.

The larva probably feeds on grasses, but I have not yet met with it.

The illustration (Fig. 2) is taken from the male insect, the female differing only in having her abdomen rounded at the tip, a sexual distinction which holds good throughout the family.

Family Noctuidæ.

Erana graminosa (Plate X., fig. 5, 5a larva).

This beautiful insect occurs commonly on the white rata blossoms (Metrosideros scandens) round Wellington during March and April, at which time it may be readily taken just after dark with a lantern and killing-bottle. The larva (Fig. 5a) feeds on the mahoe (Melicytus ramiflorus) in the spring and autumn. It remains concealed in crevices in the bark during the day, not infrequently selecting the deserted burrows of wood-boring beetles as a secure retreat from its enemies. When full grown it is olive-green, the colour being lighter on the ventral surface and between the segments. A row of ill-defined, feathery, black markings extends down the back and sides and there are also two tolerably conspicuous ochreish spots on every segment except the last. The head, legs, and prolegs are reddish-yellow, and the whole insect is more or less spotted with black. Younger larvæ differ in being of a light yellowish-green, with very pale yellow dorsal and lateral lines. A row of black warts, emitting a few bristles, extend round each of the segments, while the head is pale ochreous with a few black dots.

When full-grown this larva descends to the ground, and forms a slight cocoon in the earth round the roots of the tree, where it is transformed into a very stout, ruddy-brown-coloured pupa, somewhat paler on the wing-cases. The moth emerges in two or three months' time. Its colouring renders it so inconspicuous amongst moss that I have frequently lifted a handful of the latter out of the breeding cage, and only discovered that the insects had emerged by their falling from the moss on to the table. A very noticeable peculiarity in this species is the presence of a fringe of long hairs in a fold on the anterior margin of the fore-wing. This organ emits a fragrant perfume, and is confined to the male sex (Fig. 5). Only one or two other instances of this kind are at present known among the New Zealand moths.

Family Noctuidæ.

Mamestra mutans (Plate X., fig. 7, 7a larva, 7b pupa).

This extremely abundant species occurs almost without intermission during the whole of the year. The sluggish larva (7a) feeds on plantain, and is best obtained by overturning logs and stones, when it may be discovered among the grass and other plants growing round their edge. Its head is pale green, with two broad black stripes, and is clothed with numerous short bristles; the four succeeding segments are of a ruddy-brown colour, considerably wrinkled, the remainder being light green, suffused with a dull, pinkish hue towards the dorsal surface. The markings consist of a triangular black spot on each side of the second to eighth abdominal segments, and a cloudy lateral line of the same colour; the legs and prolegs being pale green, and the whole insect more or less marbled with black. This description and the figure on Plate X. exhibit the usual peculiarities of the larva, but in some individuals the markings there indicated are quite obsolete, and the insect is of an almost uniform pale-green colour. When mature, this caterpillar sometimes constructs a slight cocoon amongst moss, on fallen trees, but more often buries itself in the usual manner, the moth appearing in a few weeks' time. Nearly all pupæ collected at random in New Zealand will be found to give rise to either this species or the one which immediately follows (Mamestra composita). The perfect insect is most abundant in the spring and early summer, but may be found fluttering round lamps on any mild night throughout the year. The sexes differ considerably: the female is greyish white, with faint brown markings, while the male is dull reddish-brown, with the markings considerably darker (Fig. 7). His antennæ are also slightly pectinated, those of the female being quite simple.

Family Noctuidæ.

Mamestra composita (Plate X., fig. 3, 3a larva).

Very common during the spring and autumn in all open situations.

Its pretty larva (Fig. 3a) feeds on various grasses, and threatens in time to do considerable damage to pastures. The head and dorsal surface of the first segment are dark shining green, with one or two obscure white markings; the rest of the body is ornamented with a number of parallel brown, white, and orange lines, which render the larva very inconspicuous when amongst the grass. Sometimes it occurs in great numbers, nearly every blade of grass having its caterpillar; in fact this was almost the case in the Wairarapa valley in the summer of 1886, when the larvæ must have produced a marked effect on the paddocks. When full-grown this caterpillar changes into a light chestnut-brown pupa, which lies on the surface of the ground amongst the vegetable refuse. The perfect insect appears in about a month's time, and if the evening be mild may be seen flying with great rapidity at dusk; it may also be readily captured at light. The figure (3) represents the male insect, the female differing only in her simple antennæ.

Family Noctuidæ.

Mamestra ustistriga (Plate X., fig. 6 ♂).

This handsome insect is rather uncertain in its appearance, but is occasionally taken quite unexpectedly at rest on tree-trunks or palings in the daytime. Specimens may also be captured while feeding on the white rata blossoms early in March, where they occasionally occur among the hosts of other Noctuæ. The larva, which feeds on the honeysuckle, is of a pale brown colour, with two obscure darker lines on each side, the under-surface being light slate-colour. The pupa state is spent in the ground, and many fine specimens may be reared from chrysalids picked up while gardening, &c. The sexes of this insect differ considerably in colour: the male is of a pinkish grey with black markings, while the female is of a uniform pale grey, and considerably smaller.

Family Noctuidæ.

Heliothis armigera (Plate X., fig. 4, 4a larva).

This conspicuous insect occurs in great abundance during certain seasons, but is very irregular in its appearance, it frequently happening that only two or three specimens are noticed in a whole year. It is generally seen flying in the daytime, when it delights to suck honey from the flowers of the Scotch thistle, a plant which much overruns the forest lands when first cleared. The larva (Fig. 4a) is a very handsome caterpillar, of a dark brownish black colour, ornamented with yellow subdorsal and lateral lines and numerous streaks and dots of the same hue. The ventral surface is a rich yellowish brown, and the subventral line white, the spiracles being white with black rings; a reddish blotch also adorns each of the three thoracic segments. It feeds voraciously on geraniums, tomatoes, peas, and many other garden plants, where it often commits the most serious ravages. About the end of April it is full-grown, when it descends to the ground and buries itself two or three inches below the surface. In this situation it is shortly transformed into a pupa, remaining in that state until the following summer, when the moth appears. The sexes of this insect differ considerably, the male having the fore-wings of a ruddy-brown colour, sometimes inclining to orange, while in the female they are pale ochreish; both sexes are, however, subject to considerable variation, and the figure (4) is taken from a rather dark male specimen.

Family Noctuidæ.

Plusia eriosoma (Plate X., fig. 8, 8a larva).

An abundant species round Nelson, where almost any number may be taken hovering over flowers on a still summer's evening. In Wellington it occurs occasionally. The larva (Fig. 8a) is a pseudo-geometer, having twelve legs, and thus showing a strong affinity with the next family. In colour it is pale green, darker on the dorsal surface than elsewhere. A white line runs down each side, and the whole insect is covered with black dots and bristles. The colouring of different individuals varies in intensity, and a fainter white line, above the usual one, exists in some specimens. It feeds on beans, geraniums, and many other imported plants, and is doing much good in the Nelson gardens by the havoc which it is committing among the Scotch thistles—weeds equally injurious to the agriculturalist and the gardener, not only crowding out useful plants, but rapidly exhausting the soil in which they grow. Formerly this insect must have fed exclusively on the New Zealand nightshade (Solanum aviculare), on which plant it may still be occasionally found in the forest, where no imported species are available, but, like many other caterpillars in this country, it is forsaking the native vegetation for the European. When full-grown, this larva spins a slight cocoon of white silk, which is generally placed between two leaves. The pupa is of a shiny black colour, the membrane between the segments being reddish-brown. The moth emerges in about three weeks' time. The figure (8) is taken from a female insect, the male being readily distinguishable by two large tufts of hair situated at the end of his body and often very conspicuous. In some cases the wings of the female are considerably lighter than in the illustration, but otherwise the species does not seem to vary. It is the New Zealand representative of the English "Silver Y Moth" (P. gamma), no doubt familiar to many of my readers.

Family Geometridæ.

Declana floccosa (Plate XI., fig. 1, 1a larva).

I have started the Geometridæ with Declana because it exhibits a great many more points in common with the Noctuidæ than does the genus Acidalia, which latter is placed at the head of the Geometridæ by some modern Lepidopterists, chiefly, I believe, on account of neuration, a character which if taken alone cannot but produce the most unnatural divisions. The present insect is one of the commonest of the genus, and may often be observed throughout the whole summer resting on the sheltered sides of trees and fences, occasional stragglers being met with as late as the end of May. Its larva is a pseudo-geometer possessing twelve legs (Fig. 1a), and thus almost exactly resembling the caterpillars of the genus Catocala, belonging to the Noctuidæ; the curious filaments on each side of the insect making this likeness still more complete. It feeds on the "New Zealand currant" (A. racemosa), from which, individuals can be occasionally beaten during the spring and early summer. They are almost impossible to find by searching in the ordinary way, from a habit they possess of clinging firmly to the twigs, which they exactly imitate in colour. When full-grown this caterpillar constructs a small cocoon just below the ground, where it is transformed into a robust-looking pupa, from which the moth emerges in a month or six weeks' time. The sexes of this species may be readily distinguished, the male (Fig. 1) having the antennæ slightly pectinated, while those of the female are quite simple, and her body much more robust. The moth drawn at Fig. 1b has been reared from larvæ exactly resembling those of the present insect, of which it is consequently now known to be only an extreme variety. It was formerly ranked as a distinct species under the name of Declana junctilinea.

Family Geometridæ.

Chalastra pelurgata (Plate XI., fig. 2 ♂, 2a ♀, 2b larva).

This delicate species may be taken flying about the forest at night, from October till March, but is most abundant on the white rata blossoms during the latter end of summer.

Its caterpillar feeds sparingly on a delicate fern (Todea hymenophyllioides) which grows in dark glades in the forest, where the sun seldom or never shines. In colour it is generally dull brown, with a row of green or pale brown lunate spots on each side; on the ventral surface the colour is darker, except on the thorax, where it is green, the legs being also green. There are in addition numerous fine, wavy lines down the back and sides of the larva, and the dorsal surface of the thoracic segments and ventral prolegs are bright reddish brown (Fig. 2b). These larvæ are, however, very variable; in many the "lunate" stripes are much longer, having a diagonal direction, and thus extending up the sides of the insect towards its dorsal surface, while others have the ventral surface dark green, and additional markings of more or less importance.

When full-grown it spins a loose cocoon of earth and dead leaves, from which the perfect insect emerges in a month or six weeks' time. The sexes are widely different, both being figured on the Plate (Fig. 2 ♂, 2a ♀). I have noticed that at least four females occur to every male, which is a very unusual arrangement, the males being generally much the commoner among the Lepidoptera.

Family Geometridæ.

Ploseria hemipteraria (Plate XI., fig. 3, 3a larva).

A curious moth, occurring in some numbers at various blossoms during the summer evenings, but rather uncertain in its appearance. The larva (Fig. 3a) feeds at night on veronica, where it may be often found with a lantern, devouring the flowers and leaves. In colour it is light green with two yellow lines on each side, the dorsal surface being considerably darker, and almost blue. Specimens are not infrequently met with of a uniform dark brown, and the two conspicuous lateral lines are then reduced to a single obscure ochreous band. These caterpillars are very inconspicuous during the daytime, as they remain quite motionless for hours together, sticking straight out from the stems of their food-plant, which they closely resemble. The pupa is unusually robust, and possesses a sharp spine at its extremity. In colour it is pale olive brown, with a pinkish line on each side of the abdomen, the wing-cases being more or less suffused with pink. It is not enclosed in any cocoon, but may be found amongst the dead leaves round the stems of the veronica. The perfect insect appears in about three weeks' time. It is liable to be passed over for a faded leaf, the general outline and colouring of the wings rendering the insect very inconspicuous, especially amongst foliage. The specimens I have reared all closely resemble Fig. 3, so that this insect does not appear at all prone to vary.

Family Geometridæ.

Ploseria alectoraria (Plate XI., fig. 4; Plate XIII., fig. 7 larva).

One of our most variable moths, occurring occasionally amongst foliage during the summer, but most abundant on the white rata blossoms in February and March.

The larva feeds on Pittosporum eugenioides, where it may be sometimes found in October and November. It has a most wonderful resemblance to the buds of the plant, and can only be dislodged by vigorous beating. It is easily reared in captivity—in fact the female moths may often be induced to lay their eggs and the insect observed through all its stages.

The eggs are very flat, oval, and light green in colour, becoming brown at one end about five days before hatching.

The young larva is pale green with a dull yellowish head. It has no markings until after the first moult when a reddish dorsal line appears. As age advances the larva becomes darker in colour and is ornamented with a series of diagonal yellow stripes. The spiracles and antennæ are pink and very conspicuous. The legs and prolegs are very small, and the latter are bright red in colour; a fleshy process which projects from the last segment of the larva is similarly coloured. The whole insect is also speckled with yellow. When full-grown this caterpillar is very robust and measures about ten lines in length. The pupa is enclosed in a light cocoon formed of three or four leaves fastened together with silk. It is greenish brown in colour.

The perfect insect first appears in December. It may be observed during the whole of the autumn and occasionally in the winter. As the larvæ grow very slowly I am inclined to think that the females hibernate and lay their eggs early in the spring (Fig. 4).

Family Geometridæ.

Sestra humeraria (Plate XI., fig. 5, 5a larva).

This abundant species occurs in large numbers round Wellington, amongst brushwood, whence it may be often dislodged during the daytime, but is most readily procurable in the evening. The larva (Fig. 5a), feeds on Pteris incisa, a pale green fern, growing in many open spots in the forest to a height of three or four feet. Its general colour is dull brownish yellow, slightly darker on the back, and ornamented with a number of wavy yellow lines on each side. The ventral surface and legs are green and the head is dark brown; the whole insect being covered with numerous black dots and bristles. When disturbed these larvæ immediately drop to the ground, and coiling themselves up like small snakes, become very inconspicuous.

The pupa is buried in the earth about two inches below the surface, the insect remaining in this state during the winter months. The moths generally emerge about October. So far as my experience goes they are not subject to any notable variations. The specimen drawn at Fig. 6 is regarded as a variety of this species by Mr. Meyrick, but I myself believe it to be quite distinct, as among over a dozen humeraria larvæ reared in captivity, none of the imagines had the slightest resemblance to Fig. 6, although the caterpillars were all taken within a few yards of the place where such moths occurred.

Family Geometridæ.

Selidosema dejectaria (Plate XI., fig. 8 ♂, 8a ♀, 8b larva).

An abundant and conspicuous species, occurring throughout the summer, often noticed at rest on fences and trees during the day and always taken in great numbers on various blossoms in the evening.

The caterpillar is extremely variable, the colouring of different individuals being apparently much influenced by their surroundings; those specimens, for instance, taken from the pale green foliage of the mahoe (M. ramiflorus) resemble in colour the twigs of that plant, while others captured feeding on the white rata (Metrosideros scandens) are dark reddish brown. Fig. 8b is drawn from a larva found on the fuchsia, which, when in its favourite position, viz., sticking straight out from the side of a branch, is so much like one of the sprouting twigs that it absolutely defies detection. When full-grown this insect buries itself about two inches in the earth, where it shortly becomes a dark chestnut-brown pupa, lighter between the segments. The time required for the development of the perfect insect depends upon the season, larvæ which undergo their transformations in the spring developing much more rapidly than those that feed up in the autumn.^[18]

This insect is extremely variable, having been formerly divided into several distinct species; the two most usual forms are those shown at Figs. 8 and 8a, but every intermediate variety exists. The sexes are distinguished by the usual differences in the antennæ. My experience leads me to believe that the light varieties occur more frequently in the female than in the male sex, and also that the dark larvæ give rise to dark moths, and vice versâ, although a great many more specimens will have to be reared before these can be regarded as established facts.

Family Geometridæ.

Selidosema panagrata (Plate XI., fig. 7 ♂, 7a ♀, 7b larva).

One of our commonest moths, occurring in great numbers in the forest throughout the whole summer.

The larvæ (Fig. 7b) are extremely variable, the most usual colouring being that of the individual figured, but when very young they are all of a uniform green with a conspicuous white dorsal line; as age advances the caterpillars become dark olive brown of varying degrees of intensity in different specimens, some retaining a considerable amount of their original green colouring, especially those feeding on the kawakawa (Piper excelsum), whose hue consequently harmonizes with that of the plant. These larvæ often select a forked twig to rest in, where they lie curled round with the head and tail close together. They are very voracious, and are the primary cause of the riddled appearance which the leaves of the kawakawa almost invariably present. Other food-plants are the "currant" (A. racemosa), and the Myrtus bullata; those taken from the latter have a strong pinkish tint, and are consequently very inconspicuous amongst the young shoots where they generally feed. The burrows of Hepialus virescens are frequently utilized by the larvæ which feed on the "currant," as convenient retreats during the winter, a large number being often found in a single hole. When full-grown they descend to the ground and construct, on the under-side of fallen leaves, loose cocoons of silk and earth from which the perfect insects emerge in about a month's time. The autumnal larvæ, however, either hibernate or remain in the pupa state throughout the winter. This moth is even more variable than the last species (S. dejectaria), which it occasionally somewhat resembles. The sexes are very different, the colouring of the male consisting of various shades of warm brown (Fig. 7), while in the female the prevailing hue is slaty brown or even grey (Fig. 7a). Many specimens are much suffused with ochre and reddish-brown, while the stigma near the centre of the fore-wing, although sometimes almost obsolete, is often very conspicuous and black, white, or even yellow in colour. It would be of great interest to learn, by rearing a large number of these insects, whether the many varieties existing in the larval and perfect states could be traced to differences in food-plant, or some other external circumstance.

Family Geometridæ.

Selidosema productata (Plate XII., fig. 1 ♂, 1a ♀, 1b larva).

Abundant in the forest, where it may be dislodged from ferns and undergrowth during the day or captured flying about in the evening. Its larva is rather attenuated, and possesses a large hump on the second abdominal segment. In colour it is dark reddish brown, mottled with creamy white and pale green, and is sparsely supplied with a few isolated hairs (Fig. 1b). It feeds on the white rata (Metrosideros scandens), and when in its usual position—i.e., sticking straight out from a branch—absolutely defies detection. Specimens, however, may be readily procured with a lantern at night, when they may be found walking about and eating. The pupa state is spent in the earth, about two inches below the surface, the moth appearing in three or four weeks' time, this period, however, being extended in the case of autumnal larvæ, to as many months. It is extremely variable, scarcely two individuals being found exactly alike. The colouring, as in the caterpillar, is chiefly protective, consisting of a delicate tracery of browns and greys, which render the insect quite invisible when resting on the trunk of a tree, with its pale yellowish hind-wings concealed, a position it invariably assumes during the daytime (Fig. 1 male, 1a female). The curious and interesting "Tatosomas," with their enormously elongated bodies, are closely allied to the present insect; one of them (Tatosoma agrionata) being found in similar situations, although in much more limited numbers; as, however, I know nothing of their transformations, I am forced reluctantly to pass them by.

Family Geometridæ.

Hydriomena deltoidata (Plate XIII., fig. 1, 1a larva).

One of our commonest moths, appearing in great numbers during January and February, in all open situations. It is especially abundant on the fern-hills.

The larva (Fig. 1a) feeds on the plantain. It is very sluggish, and lives all through the winter, becoming full-grown in September, when it changes into a pupa, among the roots of its food-plant. In colour it is a uniform dark brown.

The moth is extremely variable, but the figure may be taken as representing a fairly typical specimen. It is a pretty insect, and may be often seen resting on fences with its fore-wings folded backwards and forming together a triangle, whence its name of deltoidata. Any unusual-looking specimens of this species should always be netted, in order to form a thoroughly representative series, as many of the varieties are very interesting. A rather uncommon and remarkable-looking form occasionally occurs, in which the dark central band of the fore-wings is completely divided near the middle.

Family Geometridæ.

Asthena schistaria (Plate XII., fig. 2, 2a larva).

This delicate little insect may be often taken at rest on fences and tree-trunks during the day, and is a conspicuous moth when flying in the evening, owing to its light colour. The larva (Fig. 2a), which feeds on the manuka (Leptospermum ericoides), is very ornamental. Its general colour is light green, with black dorsal and lateral stripes, and a series of diagonal markings bordered with crimson; the legs and prolegs are also crimson, and the segments are divided by brilliant yellow rings, a white line extending down each side of the larva. It is difficult to find, as it remains closely concealed amongst the dense manuka foliage, from which it can only be dislodged by vigorous and continued beating. The caterpillars allow themselves to fall a short distance, hanging suspended by a silken thread, which they rapidly ascend when the danger is passed. The pupa is rather attenuated, dark-brown, and much pointed at its posterior extremity. It is found buried about an inch in the earth, and the moth appears in a month's time. This insect varies much in intensity of markings. The males are generally considerably darker than the females, but are more certainly distinguished by their attenuated bodies.

The pearly white Asthena pulchraria occurs in October and April; it is a most beautiful insect, and may be found amongst the foliage of the kawakawa (P. excelsum), on which its larva will probably be found to feed.

Family Pyralidæ.

Scoparia hemiplaca (Plate XII., fig. 4).

This pretty little moth was reared from a larva found feeding amongst moss during the winter of 1885, but unfortunately I neglected to make a drawing until it was too late. Doubtless many of the other Pyrales we meet with in the New Zealand forest have similar habits, their larvæ probably feeding on different kinds of mosses. These can always be examined during the winter months, when the entomologist is usually in want of work, and thus much information may be obtained regarding this interesting but little-known family.

Family Pyralidæ.

Scoparia sabulosella (Plate XIII., fig. 4, 4a larva).

This is that extremely abundant, though dull-coloured little insect, that rises in such multitudes from every field before one's footsteps during the early summer.

Its larva (Fig. 4a) feeds on various mosses, forming numerous silken galleries amongst the roots in which it resides. These caterpillars are very active, and consequently rather difficult to obtain, as they move either backwards or forwards in their galleries with equal rapidity.

They feed during the whole of the autumn and winter, changing into pupæ about September, from which the moths emerge in a month or six weeks' time.

The habits of the numerous other species belonging to this genus and the closely allied genus Xeroscopa (Meyr) probably do not materially differ from those of the species here described.

Family Pyralidæ.

Crambus flexuosellus (Plate XII., fig. 5).

An extremely abundant insect, occurring in swarms over meadows during the summer, where it may be captured in the daytime or taken by hundreds at the attracting lamp in the evening. Its larva is at present unknown, but probably feeds on the roots of grasses.

Closely allied is Crambus tahulalis, found in similar situations, but appearing rather later in the season, the earliest specimens being met with about January, while C. flexuosellus is on the wing throughout the summer.

Family Pyralidæ.

Siculodes subfasciata (Plate XII., fig. 3, 3a larva, 3b pupa).

This curious insect may be occasionally taken flying round patches of Muhlenbeckia adpressa, which grows freely amongst brushwood in many parts of the country.

Its larva (Fig. 3a), is very stout and sluggish, resembling the caterpillar of an ordinary Pyrale in general appearance. It feeds in the stems of the creeper, causing large swellings therein, which readily betray its presence, and should therefore be cut off and kept until the moth emerges, as specimens obtained in this way are far superior to any captured in the open. The pupa is dark brown, and shining; it lies in the centre of one of the swellings, the larva having previously prepared a safe outlet for the moth in the form of a small burrow leading to the air, its extreme end remaining closed by a thin pellicle of the original bark, which effectually prevents the inmate's resting-place being discovered from the exterior (see Fig. 3b, the small circle marked * represents the outlet).

The perfect insect appears about December, flying rapidly in the hottest sunshine. It varies greatly, both in size and colour, some of the small males being very much suffused with dark brown, while the females usually resemble the figure (3), and are often more than twice the size of their mates. This insect is generally placed in a family called the Siculidæ, but I think without sufficient reason, and have therefore located it among the Pyralidæ, with which it has unquestionably a great affinity.

Family Tortricidæ.

Isonomeutis amauropa (Plate XIII., fig. 2, 2a larva).

This odd little moth may be occasionally seen basking in openings in the forest, and usually flies away with lightning speed when an attempt is made to capture it.

The larva lives under the scaly bark of the matai-tree, feeding on the soft, juicy inner bark and sap. In colour it is light yellowish white, darker on the back, some specimens becoming quite pink on the dorsal surface. When full-grown it encloses itself in a tough silken cocoon, covered on the outside with fragments of wood, from which the moth emerges in about a fortnight's time.

The sexes differ considerably in appearance, the male having much broader wings, and darker in colour than those in the female from which the illustration (Fig. 2) is taken.

This insect is probably single-brooded, as the larva may be found feeding in the trees during the whole of the winter.

Family Tortricidæ.

Cacoecia excessana (Plate XIII., fig. 5, 5a larva).

This is the commonest species of Tortricidæ in New Zealand, and may be found almost without interruption during the whole of the year.

The larva (Fig. 5a) feeds on a great variety of plants, the common manuka being probably the most usual food for the species when in a state of nature. It now, however, eats numerous European plants, including honeysuckle and occasionally the fruit of the apple, but further evidence is required on the latter subject before we can really consider it as actually injurious in that direction.

In colour this caterpillar is light green with a yellow line on each side, but varies considerably; it feeds between several rolled-up leaves, in which it is afterwards converted into a pupa whence the moth emerges in about three weeks' time.

The perfect insect is also excessively variable and is often more or less suffused with yellow. It is most abundant in the middle of summer, and may be taken at light, or in the daytime at rest on fences and trees.

Family Tortricidæ.

Ctenopseustis obliquana (Plate XII., fig. 6).

This little moth is occasionally noticed at rest on garden fences during the autumn. Its larva inhabits the interior of the peach, feeding on the kernel, which appears to exactly meet its requirements, the caterpillar being full-grown as soon as it has completely devoured the nut. Before assuming the pupa state this insect provides a ready means of escape for the future moth by drilling a small hole through the hard shell and pulp of the peach to the air; it also spins a slight cocoon inside the stone, the pupa resting in the place formerly occupied by the kernel, in which position it is often discovered. The only noticeable mischief produced by this insect is delay in the ripening of the fruit. In fact all the infected specimens which I have seen were quite hard and green, whilst other fruit from the same tree had reached complete perfection.

Family Tineidæ.

Endrosis fenestrella (Plate XII., fig. 7, 7a larva, 7b pupa).

This common species may be observed in almost any house in New Zealand, and is often mistaken for the dreaded "clothes moth" (Tinea tapezella), which it somewhat resembles in general appearance. Its larva (Fig. 7a) is very destructive, feeding on dried peas, amongst which it creates great havoc, drilling numerous holes through them and spinning a large number together, in the centre of which the caterpillar undergoes its change into a pupa (Fig. 7b), from which the moth emerges in about a fortnight's time. This insect should be destroyed whenever seen, as there is no doubt that much loss will be caused by its ravages in the future. It also infests bee-hives.

Family Tineidæ.

Œcophora scholæa (Plate XIII., fig. 6, 6a larva).

This dull-coloured insect is extremely abundant during the early summer.

The larva feeds on the roots of various plants, forming numerous white silken galleries in the earth where it resides. In colour it is dark chocolate-brown with a yellowish head and white markings. It is very large, considering the size of the future moth, full-grown specimens often measuring as much as 10½ lines in length. About the end of September these caterpillars are transformed into pupæ, and the moths emerge in a month or six weeks' time.

The perfect insect may be often disturbed amongst brushwood. It is very sluggish on the wing and usually drops to the ground, where it is very inconspicuous. It also has a habit of running into any crevice immediately on the approach of an enemy. This peculiarity is shared by the other members of the genus Œcophora, of which there are large numbers in New Zealand.

Family Tineidæ.

Semiocosma platyptera (Plate XII., fig. 8, 8a larva, 8b pupa).

This is one of the largest of the Tineidæ found in New Zealand, measuring fully fifteen lines across the expanded wings. Its larva (Fig. 8a) is abundant under the bark of dead henau trees (Eleocarpus dentatus), feeding on the soft inner surface, but leaving the hard wood untouched. In colour it is pale yellow, the head and prothorax are dark brown and corneous, and the remaining segments are provided with two horny warts, from which numerous hairs arise; its legs are all very small, and the caterpillar is considerably attenuated posteriorly; it is very active, wriggling about with great violence when disturbed.

The pupa (Fig. 8b) is enclosed in a compact cocoon, constructed of minute fragments of wood, firmly woven together with silk, and attached to the inner surface of the bark, where it may be soon found by careful searching, and the finest specimens may thus be easily reared in captivity.

The perfect insect appears about November, and may be often observed at rest on the trunks of trees; its pale hind-wings are completely concealed by the dark upper pair, which render its discovery very difficult. The sexes may be at once distinguished by their size, the males being much smaller than the female (Fig. 8) and usually lighter in colour.

CHAPTER VII

The Order Neuroptera, as here considered, is a very limited one, consisting only of the seven small families, which comprise the Lace-wings, Ant-lions, Caddis-flies, and a few others. It forms a most convenient passage from the insects undergoing a complete metamorphosis with a quiescent pupa, to those which are active during the whole of their life, as the larvæ are widely different from the adults, but the pupæ, although incapable of walking or eating, approximate very closely in structure to the perfect insects. I regret that my observations have been at present restricted to three families only, i.e., the Hemerobiidæ, Sialidæ, and Phryganidæ, which will consequently have to represent the entire series. I understand, however, from Mr. A. S. Atkinson, that a species of Myrmeleontidæ (Ant-lion) is not uncommon round Nelson, and doubtless future investigation will reveal insects belonging to the other families.

Family Phryganidæ.

Oxyethira albiceps (?) (McLach.) (Plate XIV., fig. 3, 3a larva, 3b pupa).

This insect occurs in the neighbourhood of ponds and streams during the summer. Its larva may be found commonly in the green, slimy weed floating in large masses on all stagnant waters. Being very small it is rather difficult to detect, and is best procured by washing a small quantity of the weed in a saucer of water, when the little insects will be at once seen walking about at the bottom. On examination with the microscope the case will first arrest attention, being of a most unique structure. Its shape is best described as closely resembling that of a minute pocket-flask, very much flattened at the lower end and almost transparent. Its surface is slightly corrugated, and the neck of the flask constructed of a much denser material than the body. It is open at both ends, the posterior end being perforated by a long shallow slit, which extends for nearly the whole width of the case, thus admitting a free circulation of water round the larva, which is also able to turn round and project its head and anterior segments through the lower aperture, thus occupying the reverse position to that shown in the illustration (Fig. 3a). It is, however, prevented from actually leaving the case by its abdomen, which is too large to be withdrawn from either end. The head and thorax of the larva are very horny in comparison with those portions permanently retained in the case, the legs being constructed to fold up into the smallest possible compass, a cavity existing in each joint for the reception of the preceding one—a structure which is almost universal among the caddis-worms. The two organs, situated on the posterior segments, are doubtless respiratory in their function, a large air-tube taking its rise from each and ramifying through the body in all directions. When alarmed these insects retreat into their cases with lightning rapidity, remaining concealed until the danger is passed. Their food probably consists of the green weed, although they are perhaps carnivorous, feeding on the rotifers and other animalculæ, which swarm in the water where they are found.

With regard to the method employed by the young larva in constructing, and subsequently enlarging, its case, I can give no positive information, although it is undoubtedly made of a viscous fluid, secreted by the insect, which hardens when exposed to the water; this secretion is no doubt analogous to the silk of caterpillars, which always exists in the form of a gummy fluid before being spun.

When about to change, the insect fixes its case down by four ligaments, two at each end, the extremities of these being firmly fastened to a stone; it then closes the small aperture, and constructs a curious arch-shaped partition, of dense material, a short distance from the broad end (Fig. 3b). In about a week's time the larva is transformed into a pupa, having the limbs, &c., free from the body but incapable of motion. The fixing down of the case prior to the change may be easily performed from each of the apertures, which are no doubt left open till the last for this purpose. Before the final transformation the pupa breaks through the partition at the broad end of the case and rises to the surface, the imago (Fig. 3) ascending a blade of grass to dry and expand its wings. The little exuvia of the pupa may be often noticed floating on the water, and the empty cases are very conspicuous on the sides of a glass aquarium, where the insects generally fix them down when in captivity.

Family Hemerobiidæ.

Stenosmylus incisus (Plate XIV., fig. 2).

This lovely insect is figured as an example of this family, being found occasionally in the New Zealand forest, but is rather scarce as a rule. I regret that nothing is at present known of its transformations.

Family Sialidæ.

Chauliodes diversus (Plate XIV., fig. 1, 1a larva, 1b pupa).

During still warm weather, from December till March, this large insect is frequently observed flying lazily over water at dusk, when it may be readily captured with the ordinary net. Its larva is aquatic, living under stones in running streams, where it devours large quantities of Ephemeræ and other insect larvæ, which are always abundant in those situations. It is very ferocious and will bite violently when disturbed, being furnished with a pair of powerful mandibles. The curious filaments on each side are gills, and it will be noticed that they are situated exactly where the spiracles of the perfect insect afterwards appear (see Fig. 1a).

This larva probably lives over a year, its growth proceeding very slowly, but mature specimens are not infrequently met with quite as large as the illustration. When full-grown it leaves the water and forms an oval cell in the mud, usually under a large stone; its gills then gradually shrivel up, and in ten days or a fortnight it is transformed into the curious pupa, shown at Fig. 1b, from which the perfect insect proceeds in about six weeks' time. The sexes of this species may be readily distinguished by their size, the male being considerably smaller than the female (Fig. 1), and possessing longer antennæ.

CHAPTER VIII

This Order, although including a comparatively small number of species, comprises some of the largest and most conspicuous insects inhabiting New Zealand, many of them reminding one of the denizens of the tropics in their gigantic size and striking appearance. They may be conveniently divided into the three following groups:—The Aquatic group, or those whose larvæ inhabit the water, including the Dragonflies, Mayflies, and Perlidæ; the Terrestrial group, including all the typical Orthoptera, Termites, and Mallophaga; and the Euplexoptera, including the Earwigs. We start our observations with the Aquatic group, as these exhibit the greatest affinity with the Neuroptera.

Aquatic Group.

Family Libellulidæ.^[19]

Uropetala carovei (Plate XV., fig. 1 ♂, 1a larva.)

This magnificent insect occurs in all swampy situations during January and February, when it may be seen dashing about with amazing rapidity intent on catching the various flies which constitute its food. Its curious larva is represented at Fig. 1a, the drawing having been taken from a singularly perfect exuvia, which I had the good fortune to discover, clinging to the stem of a fuchsia-tree in a swamp, the rent through which the perfect insect escaped having almost closed up. In this state it no doubt feeds on various aquatic animals, which it procures with a prehensile instrument similar in structure to the "mask" of British dragonfly larvæ, but much larger.

The female of this species may be at once recognized by the absence of the two peculiar leaf-like appendages at the anal extremity, from which the insect takes its name. Her abdomen is also much stouter. My experience leads me to believe either that she is very retired in her habits or else that there are at least six males to one female.

Closely allied, and much commoner than the above insect, is Cordulia Smithii, found almost everywhere, its rapid and continuous flight frequently taking it many miles away from any water. The specimen figured is a male (Plate XV., fig. 2), the female possessing a pair of slender sickle-shaped hooks, attached to the end of her body. She may occasionally be seen depositing her eggs in stagnant streams, the abdomen being violently beaten against the surface of the water during the operation. I have not yet met with the larva, which probably lives concealed in the mud. One specimen, taken near Lake Wairarapa, is remarkable in possessing a cloudy brown patch near the tip of each wing, but it is no doubt only a variety of the ordinary insect.

Family Libellulidæ.

Lestes colensonis (Plate XV., fig. 3, 3a larva).

Extremely abundant in all damp situations from September till May, being one of the last insects to disappear in the autumn. The larva is found under stones, &c., in every stream, feeding on various aquatic insects and crustaceans. When very young the wing-cases are scarcely discernible, but gradually become more distinct at each moult, until the larva assumes the form shown in the illustration (Fig. 3a), which is taken from a specimen about a week before the emergence of the perfect insect. In all these insects it would be much more convenient to regard the metamorphosis as consisting of only two stages, viz., larva and imago, as there is really no condition analogous to the quiescent pupa of other orders. The female is rather stouter than the male, which is the sex figured, and her abdomen is of a dull bronze colour, instead of metallic blue. The only other dragonfly found in my neighbourhood (Wellington) is the pretty little Telebasis zealandica (Fig. 4), which occurs in similar situations to the last, but is not quite so common. The male is of a brilliant red colour, the female being bronzy green, but she may be readily distinguished from the same sex in Lestes colensonis by her smaller size. The larva of this species is rather more attenuated than that of the previous insect, and is of course considerably smaller.

Family Ephemeridæ.

Ephemera, n.s., near Coloburus^[20] (Plate XVI., fig. 4, 4a larva).

The well-known mayflies are very extensively represented in New Zealand, hovering in swarms over running water during the summer evenings.

The larva of the present species (Fig. 4a) occurs abundantly under stones in rapid streams. It may be immediately distinguished from its numerous congeners by its large head and conspicuous black eyes. It is carnivorous, feeding on various small insects, chiefly those belonging to the present family, but in lack of these it will even devour individuals of its own species. It is consequently a most difficult insect to rear, and it was a long time before I succeeded in obtaining a single imago in captivity. When mature the insect leaves the water, and an apparently perfect imago escapes through a rent in the thorax in the usual way. In a few hours, however, a second moult occurs, the wings gaining additional size and beauty, and the anal setæ becoming very much more elongated than before (Fig. 4). This second change, which has so perplexed some entomologists, is merely an apparent departure from the general rule, a careful examination of the exuviæ of the dragonflies, and pupa shells of many other insects, revealing a delicate membrane within, which invests the imago, and is cast off at the same time as the harder external envelope. In the case of the mayflies, the retention of this internal membrane some two or three hours longer than usual, will fully explain its apparently unique metamorphosis.

Family Perlidæ.

Stenoperla prasina (Plate XVI., fig. 3, 3a larva).

This is the green gauzy-winged insect which we see flying feebly over running water, during the twilight, throughout the summer.

Its larva (Fig. 3a) is aquatic, hiding itself under stones, and devouring the unfortunate Ephemeræ found in similar situations. Towards the end of its career the rudimentary wings become very conspicuous, at which time it is a most interesting object. The curious appendages on each side of the abdomen are gills, which the larva is constantly vibrating, in order to obtain a fresh supply of aërated water. When mature, it ascends the stem of some aquatic plant, the skin becomes dry and brittle, and finally bursting, allows the perfect insect to escape, and in a few hours its wings are sufficiently hardened for flight. Several other species occur in New Zealand, one of the commonest being Perla cyrene, a black insect much resembling S. prasina, but considerably smaller; its larva may be occasionally found, and is at once known by its dark colour.

Terrestrial Group.

Family Psocidæ.

Psocus zealandicus, n.s. (Plate XVI., fig. 2, 2a larva).

During the hottest days in summer every one must have noticed numbers of minute active insects assembled on garden fences in groups, ranging from ten to fifty, immediately dispersing when disturbed. These are individuals of Psocus zealandicus (Fig. 2), a curious little species, closely allied to the renowned "Book Tick" (Atropos pulsatorium), whose ravages in museums and libraries need no description. Its larva (2a) may be found in the same situations as the imago, and often assembles in similar groups. Its food probably consists of rotten wood and other decaying vegetable matter, and in its later stages it is provided with wing-cases, thus differing from the Book Tick (A. pulsatorium), which remains apterous during the whole of its life.

Family Termitidæ.

Stolotermes ruficeps (Plate XVI., fig. 1 ♂, 1a ♀, 1b "soldier," 1c "worker").

The termites, or white ants, which occur in such great numbers in the tropics, are represented in New Zealand by several small species, the commonest in this neighbourhood being Stolotermes ruficeps.

This species inhabits rotten logs, excavating extensive burrows, resembling in a very humble manner the wonderfully elaborate nests constructed by the African and other species, about which so much has been written, and so much remains to be discovered. The present insect appears in the perfect state during January and February. It is seldom noticed flying about, but may be readily obtained by opening the nests, where a large number are frequently seen huddled together in the main galleries. At this time the community consists of three classes of individuals, viz., males, females, and workers, which last are in all probability nothing more than the larvæ. After pairing they shed their wings and return to the nest, the female becoming very much distended with eggs. About March she commences to lay. This is continued for several months, and during this time the female is queen of the nest. She resides in a capacious chamber, from which numerous galleries diverge in all directions, some extending as far as eighteen or twenty inches, but the most populous portion of the nest is contained within a radius of six inches from the queen's apartment. The "soldiers" (Fig. 1b) now appear in considerable numbers. They are chiefly stationed in the royal chamber, and furiously attack any intruders; but the workers which stream in and out, carrying the eggs from the queen, they treat with the greatest gentleness. I have never seen soldiers in a nest containing winged insects, nor indeed later in the spring than October, when they seem to have all disappeared. With regard to the nature of these individuals I am unable to supply any positive information, but it appears probable that they are abortive males, in the same way that the neuters of the bees and ants are abortive females. As none of these insects have yet been reared, many points of great interest remain to be discovered in connection with their economy, and a rigid investigation of a number of nests kept in captivity, is the only mode by which we can hope to become fully acquainted with the habits of this interesting family.

Family Blattidæ.

Periplaneta fortipes (Plate XVII., fig. 5).

Few people who cut up old wood remain unacquainted with this species for very long, its insufferable odour immediately betraying its presence independently of anything else. It is very common under the bark of rimu, henau, and other large trees, where specimens may be found in all stages of growth; the mature individuals only differing from the young in the matter of size and the possession of rudimentary wing-cases. I have never found the females of this species carrying their eggs, but have, on several occasions, discovered the closely allied, but smaller, Periplaneta undulivitta thus engaged under stones on the hills round Nelson. This is a much more agreeable insect to study than P. fortipes, not possessing the disgusting odour so characteristic of the latter species.

The only winged Blattidæ found round Wellington are Blatta conjuncta, and Periplaneta orientalis. The former (Fig. 6), may be occasionally noticed under the scaly bark of rimu and matai trees, but a sharp eye and hand are needed to effect a capture, the insect running with marvellous rapidity. The latter species I have not yet noticed, but as it is the ordinary "cockroach" of Europe its habits have already been amply described.

Family Mantidæ.

Tenodera intermedia (Plate XVII., fig. 2).

A local species confined, I believe, to the South Island, and occurring in some numbers round Nelson, where my specimens were obtained. It seldom flies, but crawls stealthily about the trunks of trees, in the hottest sunshine, capturing and destroying great quantities of insects, its green colouring and leaf-like form rendering it very inconspicuous to its victims. The purple spots on the tibiæ of this insect are very noticeable, and resemble small drums in structure, hence they are regarded by Mr. A. H. Swinton ("Insect Variety," page 239), as the organs of hearing. These curious drums may be also found in insects belonging to nearly all the remaining families of the Orthoptera, but, as we find no auditory organs occupying a similar situation in any other groups of insects, I think that Mr. Swinton's explanation of their function must be regarded at present as a somewhat doubtful one.^[21]

Family Phasmidæ.

Acanthoderus horridus (Plate XIX.).

The curious Stick Insects are familiar to most people from their remarkable similarity to the twigs of trees.

The present species is one of the largest, the mature insect frequently attaining a length of five inches. It is best taken at night, when it may be readily discovered, feeding on the leaves of shrubs, and suddenly becoming perfectly motionless when the lantern is turned upon it. The favourite plant for this (and indeed most of the species) is the white rata, upon which they are often seen in large numbers when the entomologist is collecting Lepidoptera in autumn. One of the commonest species found in this way is Bacillus (hookeri?) chiefly remarkable for its great sexual disparities, the male resembling a very slender stick about twenty-eight lines long, while the female is nearly half as long again (thirty-eight lines), and much more stoutly built. A more systematic investigation of this family is needed before we can pretend to correctly determine the various species, as there is little doubt that in other cases the sexes will be found quite as divergent. In addition to this the insects are most variable in colour, and their completely apterous character rendering the distinction between larva and imago a matter of considerable difficulty, it is very probable that some of the smaller species may be only immature specimens of the larger ones.

Stick insects are easily kept in captivity, and will not be found devoid of interest. They are great eaters, and grow with considerable rapidity, frequently casting their skin, a task of no easy accomplishment, which I once had the pleasure of watching in the case of a specimen of Acanthoderus prasinus which I had under observation for several months.

The insect first suspends itself by its hind pair of legs, keeping the others in the same position as when walking, the head is bent in, and the antennæ are placed along the breast, the long abdomen hanging over backwards. The skin then splits along the back of the thorax, and the head and thorax are gradually pushed out. The front and middle legs are immediately afterwards extracted, the long femora and tibiæ easily passing the sharp angles in the exuvia, owing to their complete flexibility. When these are finally clear, the insect reaches forwards with its fore-legs and draws the abdomen and hind-legs out of the old skin, which remains attached to the branch until dislodged by some accident.

During the spring months great quantities of little stick insects may be noticed on the parasitic ferns covering the tree stems in the forest; they are curious little animals, their antics when simulating inanimate twigs being often most amusing, and if the reader wishes to investigate a comparatively untouched branch of entomology he cannot do better than keep a number of these until mature, when he will doubtless contribute much to our scanty knowledge of this curious family.

Family Achetidæ.

Acheta fuliginosa (Plate XVIII., fig. 1).

This destructive insect is not indigenous to New Zealand, having been introduced from Australia into the Nelson district many years ago. Strange to say it has never been seen in Wellington, where specimens must be constantly landed amongst produce, &c., but appear to be unable to effect a settlement, owing, probably, to some peculiarity of the climate which renders the place unsuitable for them. The larvæ may be first observed about December, when they are often seen hopping about the vegetation. They are extremely obnoxious, devouring everything, and frequently entering houses, where they consume provisions, clothes, and even boots. During the summer of 1875 the farmers round Nelson were fairly eaten out by this insect, the cattle absolutely starving for the want of food, but since that time the pest seems to have gradually diminished, although it is still very injurious to many garden plants.

The illustration (Fig. 1) is taken from a female, the male wanting the long ovipositor. These insects appear in the imago state about March, and continue in great abundance until the end of summer, the cold weather which generally sets in about the beginning of May rapidly destroying them.

Family Gryllidæ.

Deinacrida megacephala (Plate XVIII., fig. 2 ♂, XVII., fig. 8 ♀).

This conspicuous species is especially interesting, as it may be regarded as the type of a very peculiar assemblage of apterous crickets, pre-eminently characteristic of New Zealand. It is very abundant round Wellington, and may be occasionally taken under logs, &c., but is best procured from the hollow stems of various trees, where it is found inhabiting the deserted galleries of wood-boring species—frequently enlarging them to suit its own requirements.

The plant most usually selected by these insects is the mahoe (Melicytus ramiflorus), whose stems may be often seen pierced with large holes. Out of these the insects emerge at night to feed on the leaves. To extract a number of specimens, without injury, requires considerable care, and is best performed with a small axe, which should be first used to cut in about three-quarters through the trunk, just below one of the holes. Another notch is then cut about a foot lower down, and the intermediate wood split off in long pieces, until the tunnel is laid bare. On approaching an insect the first thing seen are two red threads, which are the antennæ, laid back as shown at Fig. 8. A deep notch is then cut into the trunk, some nine or ten inches below this point, and the piece bodily wrenched off. If the individual thus treated is a male he will cling firmly to the log, elevating his hind-legs in the air and biting viciously at anything within reach, but the females, in the majority of cases, endeavour to escape and hide themselves under the leaves, &c., on the ground. Both sexes when irritated emit a peculiar grating sound, which may be often heard at night in the forest, and is produced by the friction of the femur against a small file situated on each side of the second abdominal segment. They can also leap a short distance, but not so far as many of the smaller species (Libanasa macropathus, &c.). They are evidently strictly arboreal in their habits, as they exhibit great skill in walking along branches, and will climb up a thin stick with wonderful rapidity.