joints—the third, fourth, and fifth the shortest, the sixth very large, ovate, with a small projection at the tip. Feet rather thick; digitules four, all fine hairs. Mentum large, conical, triarticulate; setæ very long. Abdominal extremity bearing two small tubercles, each with a long seta, and two or three spines; between the tubercles protrude two rather thick pencils of white cotton. Epidermis not thickly clothed with hairs, but on each segment is a single transverse row of small spiny hairs with some small simple circular spinnerets.

Male unknown.

Hab. On various trees in forests, Reefton district, New Zealand. The second stage mostly on Podocarpus totara or various species of Fagus.

This insect is easily distinguished from C. zealandicum by the snow-white shelly covering both in the adult and the second stage, and by the thick coating of hairs on the epidermis. The larva is much smaller than that of C. zealandicum, and, curiously, it differs also in being much less hairy, quite the contrary of the second stage. I do not think that this can be the female form of C. wairoense, because the peculiar brush of digitules which distinguishes the male of that species has no counterpart in C. pilosum; and as far as my experience goes such a character would in all likelihood be shared by both sexes. The female of C. wairoense and the male of C. pilosum may both be discovered some day.

Cœlostoma assimile, Maskell. N.Z. Trans., vol. xxii., p. 153. Plate VII., figs. 11–17.

In my paper of last year I gave a brief description of the second female stage of this insect; I am able now to complete it, and to add also the adult female and the larva.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Adult female reddish-brown, occupying a deep pit in the twig it lives on (mostly in the axils), the mouth of the pit being covered, with the mass of thick yellow wax, of irregularly globular shape, already formed by the second-stage female. The cavity burrowed out is frequently quite deep in the bark, and even seems sometimes to extend into the wood. The dimensions vary: some of the waxy coverings are only about 1/15in. across; others observed reach ⅛in. The wax is very hard. The adult female at first, after discarding the pupal skin, occupies the whole space, and is globular, with a diameter of about 1/16in., obscurely segmented; but at gestation it shrivels up into an extremely small shapeless mass, filling the cavity with red oval eggs, and becoming at last so shrivelled that it is extremely difficult to find her. Antennæ short, thick, atrophied; the joints apparently only five, but the fifth may be made up of several, so that the normal antenna would have

eleven joints, as usual in the genus: the third and fourth joints are very short; the fifth long, sub-conical, with three or four, spiny hairs at the tip: as the third and fourth joints are the thickest, the antenna is somewhat elliptical. Feet absent. Rostrum and mentum entirely wanting. Epidermis covered with very slender hairs, which are sparse on the cephalic and thoracic regions but very numerous on the abdomén, and also with many small, simple, circular spinneret-orifices. Spiracles sixteen, four of which, on the cephalic region, are very large. Anogenital ring simple, with a tubular organ leading to it, as usual in the genus; at each side of the ring is a sub-circular scar or orifice of the same dimensions.

Female of the second stage as described in my last paper (vol. xxii., p. 153). I omitted, however, then to state that from the anogenital ring and tubular organ there springs a long, white, slender filament. This filament is sometimes more than an inch long: it protrudes through the waxy covering, and at the extremity there is often seen a minute bubble of honey dew. The filament is excessively brittle, and pieces broken off are frequently observed on the adjacent twig. At the metamorphosis the exuviæ of this second stage remain in the cavity, and become flattened and disc-like: on breaking open the waxy test in late autumn, when the eggs have been laid, this pupal skin is usually the first object visible, covering the eggs, and, as the adult female (as stated above) has shrivelled up, the skin may be at first mistaken for the adult, but the presence of the very large rostrum and mentum at once shows the real state of the case. The filament already mentioned frequently remains protruding, and is found to be attached to this pupal skin.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Larva red, elongated, sub-elliptical, tapering somewhat posteriorly, active, sparsely covered with thin white cotton; length about 1/35in. Antennæ of six joints—the first five cylindrical, sub-equal; the sixth very large, ovate. Feet with rather thick femora and tibiæ; tibia shorter than tarsus; claw slender, with a tooth on the inner margin; no tarsal digitules, the lower pair long, fine, with a slightly-dilated tip. Mentum large, conical, triarticulate, with a few hairs at the tip. Epidermis bearing many simple circular spinneret-orifices, which are most numerous in a circular patch at the abdominal extremity, with the usual tubular organ bearing three rings of glands; at the extremity are two very small anal tubercles, each bearing a long seta, and from the tubular organ there protrudes between the tubercles a thickish pencil of white cotton.

Male unknown.

Hab. On Fagus menziesii, Phyllocladus trichomanoides, and Fagus fusca, Reefton district. It appears to be not

uncommon in that region. The waxy tests are very small and inconspicuous.

The finding of the adult female enclosed in the waxy mass formed by the female of the second stage was a surprise to me, being a distinct departure from the normal condition of things, especially in the group Monophlebidæ. It affords a good instance of the necessity, as I take it, of thorough examination of all stages before any absolute certainty can be arrived at, even as to the generic position of an insect. Whatever may be the rule amongst other orders and families of insects, Coccids present this difficulty to students: that one must be prepared at any time to find very distinct departures from generic, or even group, types, and to consider any character whatsoever as elastic and variable. Thus, for example, in the Dactylopidæ, the insect described in my last paper under the name D. obtectus departs entirely from the generic type in employing a portion of the plant it lives on as a shelter or “scale” for the adult female. So, again, in the Monophlebidæ, no genus or species has been reported hitherto in which the adult female is not, at least before gestation, free and active, or only covered with loose cotton. Cœlostoma assimile departs altogether from the types of the group and the genus in remaining under the thick waxy test in its adult state; and the variation is emphasized, so to speak, by the absence of the feet. The characters, however, of the larva and of the second stage being so clearly those of the genus Cœlostoma, and the rostrum and mentum of the adult being entirely absent, I have no hesitation in retaining the insect in that genus despite the form of the antenna; and, indeed, this may be looked on as merely atrophied.

The deep cavity made by this species in the wood of its tree, and its usual position in the axils of the twigs, make it resemble Xylococcus filiferus, Löw (Verh. der zool.-bot. Gesellschaft, Wien, 1882), which insect also has a small covering of wax, and a very long slender filament protruding from the abdominal extremity. But the presence in Xylococcus of a very distinct rostrum (which, indeed, seems to be the only member preserved by the adult female) clearly distinguishes that genus from Cœlostoma; and, indeed, the forms of the larva and of the second stage are also quite different.

The late Dr. Löw, whose death a few months ago deprived entomology of one of its brightest ornaments, devoted some remarks in his description of Xylococcus filiferus to the question of the manner in which that Coccid managed to produce a deep cavity in the wood of its food-plant, Tilia europæa. He was, however, unable to give any other explanation than that it proceeded “by a most peculiar kind of suction, and besides by a certain influence which these insects exercise on