Art V.—Further Notes on Coccidæ in New Zealand, with Descriptions of new Species.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 16, 1883, Page 120

Art V.—Further Notes on Coccidæ in New Zealand, with Descriptions of new Species. By W. M. Maskell, F.R.M.S. [Read before the Philosophical Institute of Canterbury, 4th October, 1883.] Plates I. and II. 1st Group.—Diaspidæ. Genus, Aspidiotus, Bouché. 1. Aspidiotus aurantii, mihi. (Trans., vol. xi., p. 199: vol. xiv., p. 217.) This must be abandoned as a species. From a communication from M. Signoret, I find that it is identical with A. coccineus, Gennadius. A description of this last is to be found, I believe, in a Report to the Minister

of Agriculture in Greece. Mr. Comstock, who adopted my name for the insect, in his Report on the Coccidæ of the United States, seems to have been equally unaware with myself of the description of Gennadius. I have always thought that this insect must have been originally introduced from Europe. 2. Aspidiotus sophoræ, sp. nov. Fig. 1. The puparium is normal in shape, flat and nearly circular, of a bluish-grey tint; diameter about 1/24 inch. The female resembles generally A. nerii or any other of the genus, having a peg-top shape, the abdominal segments shrinking up into the thoracic portion after gestation. There are five groups of spinnerets, of which the upper group has four orifices, the remainder seven or eight. In some specimens only four groups are to be made out. The edge of the abdomen ends in two conspicuous median lobes, and at each side for a little distance are a number of scaly serrated hairs resembling those of A. nerii. From Sophora tetraptera. Only a few species of Aspidiotus are reported as having five groups of spinnerets, the normal number being four, and some having no groups. The present insect seems to resemble generally A. oxyacanthæ or A. tiliæ, Signoret; but differs in the arrangement of the hairs on the edge of the abdomen. Genus Poliaspis, mihi. (Trans., vol. xii., p. 293.) Mr. Comstock, in a monograph of the Diaspidæ (Second Rep. of the Dep. of Entomology of the Cornell University Experiment Station, 1883, p. 126), which he has kindly sent me, adopts this genus, and describes a new American species of it; but remarks that he is “far from feeling sure that the genus will prove to be a natural one.” I would urge that it has at least as good a claim as the kindred genus Leucaspis, which Mr. Comstock admits without remark. Dr. Signoret, in a letter to me, says of my genus Poliaspis—“I think this genus is distinct. Genus Mytilaspis, Targioni-Tozzetti. (Trans., vol. xi., p. 192.) 1. Mytilaspis pyriformis, mihi. (Trans., vol. xi., p. 192: vol. xiv., p. 215.) Mr. Comstock (loc. cit., p. 125) asks whether this insect is really a Mytilaspis, and seems inclined to refer it rather to Chionaspis. I find that in my former papers I have omitted to observe that the male puparium is not greatly different from that of the female, though from my placing it under Mytilaspis I inferred this. The male puparium of Chionaspis is quite different.

Genus, Diaspis, Costa. (Trans., vol. xi., p. 192.) 1. Diaspis santali, sp. nov. Puparium of female yellowish-grey in colour, sometimes with a greenish tinge, outline oval; very convex: pellicles black, very inconspicuous, placed at one end of the puparium. Adult female orange-red in colour, reaching 1/15 inch in length, peg-top shaped; the abdominal segment very small as compared with the rest of the body, and the two next segments slightly overlap it. Abdomen ending in two conspicuous, prominent, median lobes, and at each side of them two semi-circular depressions: several branched and serrated hairs in the region of these lobes. There are no groups of spinnerets, nor can I detect any single ones. There is no wide depression of the edge between the median lobes. Male puparium elongated, with black pellicle: but I have not seen the insect itself. On twigs of Maire (Santalum cunninghamii), sent me by C. Winkelmann, Esq., from Napier. It is in company with Rhizococcus fossor, described below: but the latter is always on the leaves of the plant. The forms of the male and female puparia show this insect to be a Diaspis, though it seems in many respects to resemble rather Aspidiotus. I find no described species of the genus possessing the same features in the abdominal segment, especially the absence of spinneret groups. Nor do I think it is a Chionaspis. Genus, Fiorinia, Targioni-Tozzetti. Mr. Comstock (loc. cit., p. 110), discards the above name and calls the genus “Uhleria,” for the reason, he says, that Targioni, establishing it to include the original species Diaspis fioriniæ, changed the specific name to pellucida. “According to the rules of nomenclature now generally adopted by zoologists,” he says “the original specific name must be restored and a new generic name given.” I think this is a great pity if correct, and tends to introduce endless confusion, and prefer to leave the original name as above. The characteristic feature of this genus is the great size of the female pellicle of the second stage, which almost, or quite, fills the puparium. 1. Fiorinia minima, sp. nov. Figs. 2, 3. The puparium is oval, being indeed almost altogether made up of the second pellicle, with a narrow edge of fibrous secretion, as shown in the

diagram (fig. 3). The first pellicle is comparatively large, and the exuviæ of antennæ are visible on it. These pellicles are, as usual in the Diaspidæ, yellow: the fibrous portion white and very thin. Length of puparium 1/36 inch nearly. Adult female of the general form of Mytilaspis. The abdominal segment (fig. 2) is somewhat long, the edge broken by a number of deepish curvilinear serrations, and ending in two inconspicuous median lobes, with three others much smaller on each side. From the serrations spring some hairs. There are five groups of spinnerets, but the three upper ones, almost or quite conjoined, form a nearly continuous arch containing 40 to 50 orifices: the two lower groups have 15 to 20. There are several single spinnerets. The adult insect before gestation nearly fills the space covered by the second pellicle: after gestation it shrinks up into very small compass at the cephalic end of the puparium: colour pink. I have not seen the male insect, but the male puparium is rather longer than that of the female, much narrower, and carinated above. From Brachyglottis repanda and Panax arboreum, but the insect seems to be uncommon. This insect differs from the European F. pellucida in its extremely minute size, in the serrations of the abdomen and the number of its hairs, and in the absence of two tubercles seen in the European species between the antennæ of the young insect. 2. Fiorinia grossulariæ, sp. nov.? Fig. 4. Puparium irregularly oval, being formed chiefly by the second pellicle, with a narrow edge of fibrous secretion. Length about 1/20 inch. Pellicles dark yellow. Adult female (fig. 4) of general form of Mytilaspis, but the cephalic end is slightly prolonged into a compressed cylinder. Lateral corrugations five, each bearing at the edge three sharp spines. The edge of the abdominal segment is much broken by serrations and ends in two broadish median lobes with two smaller lobes on each side. Several sharp, long spines are set in pairs along the serrated edge. Groups of spinnerets normal, the three upper forming a continuous arch. Colour of insect dark grey. I have not seen the male. This insect was sent to me on some gooseberries from a garden at Amberley by Dr. Morris. I scarcely know what to make of it. From the locality it is hardly likely to be indigenous; but no other insect of the genus is at all like it.

3. Fiorinia stricta, sp. nov. Figs. 5–9. Puparium very narrow, the length being about five times the width. The second pellicle fills almost the whole of it, half of the first pellicle appearing at the cephalic end, and a very narrow edge of secretion running down the sides, which are almost straight and parallel. The diagram (fig. 6) shows the arrangement. Length of puparium about 1/11 inch: colour, for the female, almost black; for the male, white. The first pellicle shows the compressed cylindrical form of the head, spoken of under the last insect, F. grossulariæ. Indeed, this appears to be a character common to all the genus Fiorinia in this country, with the exception of F. minima. The second pellicle, which forms the puparium, appears to be entire throughout almost all its length. But at the posterior end, as shown in fig. 6, it is cut across by several transverse corrugations dividing it into narrow segments; and the edge, generally semicircular, is sharply serrated, but the serrations are not so deep as in F. minima. This pellicle is very strong and hard, and by its dark colour makes the puparium altogether look black. The adult female is very small in comparison with the second pellicle (fig. 6). After gestation, indeed, it shrinks up at the cephalic end so as to become difficult of detection. It has the general form of Mytilaspis, with compressed cylindrical head. The lateral corrugations bear no spines. The abdominal segment (fig. 5) is somewhat elongated, and the edge is broken into a number of protruding, sharp-pointed lobes, giving it the appearance of a comb. There are no median lobes like those of almost all other Diaspidæ. Spinnerets normal of the genus, the three upper groups joined in an arch. Length of adult female before gestation about 1/30 inch: colour dark brown or purple. The male paparium is, in shape, similar to that of the female: but of course, as there can be here no second pellicle, it is composed almost altogether of fibrous secretion. Its colour, therefore, is white, with the first pellicle, which is black, at the cephalic end. Length, about 1/11 inch. The adult male has the normal form, generally, of the Diaspidæ. The antennæ (fig. 7) have ten joints, of which the first two are very short, the rest long, thin, hairy, and about equal to each other, except the last, which is a little shorter and broader, being irregularly fusiform. Amongst the hairs on this last joint is one a little longer than the rest and bearing a terminal knob. Foot normal (fig. 8); the claw is very slender, and the four knobbed digitules are fine hairs. There is a strong spine at the extremity

of the tibia (see fig. 8). The abdominal spike (fig. 9) springs from a tubercle at the end of the abdomen, having a small setiferous tubercle on each side. The haltere is normal. I have this insect on Dendrobium (sent by the Rev. Mr. Colenso), and on Hedycarya (sent by C. P. Winkelmann, Esq.), both from Hawke's Bay. I do not think it is common. There is no mistaking F. stricta: the very long and narrow second pellicle, with its segmented extremity and black colour, and the comb-like serrations of the abdominal segment in the adult female, clearly distinguish it from all others. The male, alone, would not suffice for identification: it might easily be taken for a male of Mytilaspis cordylinidis or Chionaspis dubia, though it may be known from the former by the black colour of its first pellicle, and from the latter by the absence of keels on the puparium. 2nd Group.—Lecanidæ. In my first paper (Trans., vol. xi., p. 205) I divided this group into three subsections—Lecanieæ: Lecanio-diaspidæ: Pulvinarieæ. In the second of these I included the New Zealand genera Ctenochiton and Inglisia: and I afterwards added Lecanochiton. I am inclined now to propose a new general arrangement of the whole group. M. Signoret, in his Monograph (Annales de la Soc. Entom. de la France, 25 Mars, 1868, p. 268 et seq.), admitted the subsection Lecanio-diaspidæ, first proposed by Professor Targioni, but confined it to the four genera Pollinia, Planchonia, Asterolecanium, and Lecaniodiaspis. Of these, the first three cover themselves with a hard test, the fourth forming a sac of felted matter. He excluded from the subsection the following genera of covered Lecanidæ:—Signoretia, Eriopeltis, Philippia, Vinsonia, Ceroplastes, Fairmairia, and placed the two genera Ericerus and Carteria in a separate position, being somewhat abnormal. Mr. Comstock follows a similar course, describing several species of Ceroplastes, but remarking that he has not found any Lecanio-diaspidæ in America (Report of the U.S. Commissioner of Agriculture, 1881: Report of the Entomologist, p. 278, note). When the subsection was established by Professor Targioni one of its characters was that the females become “apodous in the adult stage.” I imagine that it is this which induced M. Signoret and Mr. Comstock to narrow its limits. But I see no sufficient reason for this. The Diaspidæ are not separated from the Lecanidæ because the females become apodous, but because their whole habit differs, by the formation of tests or puparia composed partly of discarded pellicles, by their life-history in these tests, and by the peculiar nature of the abdominal segment in the female. Moreover, there would seem to be no greater reason for selecting the loss of the feet as a character than the loss of the antennæ. Now, in the

Lecanio-diaspidæ of Targioni, Planchonia loses its antennæ, Lecaniodiaspis keeps them. But it appears to me that there is an organic difference between the secretion of a test, whether of wax, or felt, or cotton, and the absence of any test at all. The difference between the naked genus Lecanium and the covered genus Ceroplastes is, I take it, much greater than that between Ceroplastes and Vinsonia. In my paper of 1878 (Trans., vol. xi., p. 207), I introduced the subsection Lecanio-diaspidæ with an extension of its limits; and I was led to this in great measure because the term itself seemed so apt for expressing the main difference between certain genera and the rest of the Coccid family, that is, the possession of certain characters common to all Lecanidæ together with the formation of tests as in the Diaspidæ (though not necessarily including any pellicles). I then added to the subsection the genera Ctenochiton and Inglisia, and in my paper of 1881 (Trans., vol. xiv., p. 221) the genus Lecanochiton. There appears to be only one character common to all the Lecanidæ which can be easily and clearly made out. It is the presence, at the abdominal extremity of the female, of a more or less deep cleft, above which, on the dorsal side, are two small protruding (usually more or less triangular) lobes. Other characters, of course, exist, such as the generally monomerous mentum, the usually stationary position of the adult female, the antennæ with almost always (in the adult) six or seven joints, and so on. But these distinctions cannot always be exactly observed. For instance, it is often exceedingly difficult, if not impossible, to tell whether the mentum is monomerous or dimerous. But, as far as my own observation goes, and from the description of M. Signoret and others, I know of no species of Lecanidæ which does not exhibit quite clearly enough the abdominal cleft and its two lobes. In the group Diaspidæ the abdomen has no such appearance: there is sometimes, as in Chionaspis dubia, mihi, or Diaspis rosæ, Sandberg, a slight median depression, but the whole form of the part is quite distinct. In the group Coccidæ the abdomen either has no appearance of division, as in the Dactylopii (see fig. 19 g., Trans., vol. xi., pl. viii.), or else ends with prominent processes, which I have called “anal tubercles,” as in Eriococcus hoheriæ, mihi (Trans., vol. xii., pl. vii., figs. 14, 20). I append to this paper (figs. 22, 23, 24, and 25), diagrams to give a comparative view of the three groups. It follows then that the group to which an insect belongs can be at once discerned by reference to the abdominal extremity. There is one exception to this in the insect producing stick-lac, Carteria lacca, where the cleft and lobes are not easily made out; and there are the genera Kermes, Pollina, etc., where the adult insect shows the cleft and lobes of Lecanidæ while the young has the anal tubercles of Coccidæ. But exceptions prove the rule.

The number of joints of the antennæ, the number and character of the digitules, and the sheath of the penis in the male, are points on which specific and generic distinctions may well be founded; but the joints of the antennæ are often difficult to determine, and various observers may imagine seven joints where others see only six and others eight. An instance of this difficulty is given below, under Ctenochiton perforatus. With regard to the mentum, it seldom happens that this can be thoroughly well examined in an adult or old female. It is usually stated also that a special character of the Lecanidæ is the immobility of the adult female. But, first, this is certainly not a distinction from the Diaspidæ, for in that group the insects become apodous, and therefore fixed, from an early stage; and, secondly, no Lecanid is more stationary than Icerya purchasi, an undoubted member of the Coccid group; for this insect, as soon as it begins to form its peculiar ovisac, ceases altogether to move about and simply becomes gradually raised up a tergo. In point of fact, with the exception of the abdominal cleft and lobes, there is probably not any character of the Lecanidæ which cannot be found in some one or other of the remaining groups. This being so, I would confine the distinguishing character of this group to its abdominal features, and would employ for its subsections the nature of, or the absence of, the secretion produced. In this way the following key would show a gradually progressive series, linked with the Diaspidæ by Lecanochiton and with the Coccidæ through Kermes and Planchonia. 2nd. Group. Insects presenting at all stages a cleft at the abdominal extremity, and, above it on the dorsal side, two more or less conspicuous and triangular lobes Lecanidæ. Subsection I. Insects covering themselves with a secretion, composed chiefly of waxy, horny or glassy matter Lecanio-diaspidæ. Test horny, partly formed of the second pellicle Lecanochiton. Test waxy, with single fringe of broad segments Ctenochiton. Test waxy, produced into radiating branches Vinsonia. Test waxy, without fringe or branches Ceroplastes. Test waxy, elevated, as if double Fairmairia. Test glassy, conical or elongated, elevated, striated with air-cells Inglisia. Test agglomerated in a waxy mass containing colonies of insects, male and female Carteria. Test absent for the female, present and aggregated in a waxy mass for the male Ericerus.

Subsection II. Insects naked Lecanieæ. Insects propagating without ovisac, arboreal Lecanium. Insects propagating without ovisac, subterranean, retaining feet and antennæ Lecanopsis. Insects propagating without ovisac, subterranean, losing feet and antennæ Aclerda. Insects forming ovisac Pulvinaria. Subsection III. Insects covering themselves with a secretion of cottony or felted matter Lecanio-coccidæ. Secretion felted, appearing only in the last stage, after gestation Signoretia. Secretion felted, forming a nearly complete sac, before gestation Lecaniodiaspis. Secretion felted, forming complete sac Philippia. Secretion cottony, covering the insect only in its last stage, after gestation Lichtensia. Secretion cottony, forming complete sac, before gestation Eriopeltis. The series thus presented would contain all the Lecanidæ proper. Between it and the Coccidæ proper would come a small group, combining the characters of the two, and linking them together, thus:— 3rd Group. Insects presenting at one stage the anal tubercles of Coccidæ, at another the cleft and lobes of Lecanidæ Hemi-coccidæ. Subsection I. Insects naked Kermitidæ. Insects globular Kermes. Subsection II. Insects covering themselves with a secretion of glassy or waxy matter Crypto-kermitidæ. Test hard, waxy, with single fringe Pollinia. Test hard, waxy, with double fringe Asterolecanium. In the foregoing series I have omitted the genus Physokermes, Targioni, not being able to make out exactly its distinction from Lecanium; also the genus Planchonia, because, as stated in my paper of 1881 (Trans., vol. xiv., p. 223), it has the anal tubercles of Coccidæ in all stages. In the case of many of the genera given above, as for instance Pollinia and Asterolecanium, or Lecanopsis and Aclerda, I do not know whether the distinctions given by authors are sufficient: but I have not seen all these insects.

Of the genera given above, the following have been reported by me as indigenous to New Zealand—Lecanochiton, Ctenochiton, Inglisia; and several species of Lecanium have been introduced. Strictly speaking, perhaps Ctenochiton, Vinsonia, Fairmairia, and Inglisia might properly be united under the same genus Ceroplastes. Still there is, I think, a sufficiently marked distinction between the tests of all to allow of their separation; at least, the distinction is quite as marked as that of the genera Aspidiotus and Diaspis in another group. Subsection I.—Lecanio-diaspidæ. 1st Genus, Lecanochiton, mihi. 1. Lecanochiton metrosideri, mihi. (Trans., vol. xiv., p. 222.) Figs. 26, 27. I have found the male of this insect, last year. It has the generally normal form of Lecanidæ, the abdominal spike or sheath of the penis being moderately long, straight, and stoutish. Antennæ (fig. 26) of ten joints, of which the first two are very short, the third much longer and expanded at the tip, the fourth more than twice the length of the third; the remaining six almost equal, about the length of the third, but stouter and rounder, being almost moniliform. All the joints but the first two have hairs. Foot generally normal; the tibia slightly expanded at its tip, with a strong spine; claw slender, with four digitules, fine hairs (fig. 27). Colour dark red; thoracic band inconspicuous. Length, exclusive of spike, about 1/40 inch. The test is white and glassy. When describing the female of this species in 1881 I had specimens only from rata trees in Milford Sound, at the extreme south-west of New Zealand. The male above described I found, with a great number of females, on the pohutukawa, near Auckland, almost at the extreme north. This last tree, Metrosideros tomentosa, of the same genus as the rata, does not grow, I believe, far south of Auckland. The occurrence of Lecanochiton on it is therefore another instance of the persistence of certain Coccids on certain trees. Milford Sound and Auckland are, I suppose, 700 miles apart, but in both the insect attacks the same genus of plant. In the same way Inglisia leptospermi may be looked for with almost certainty on the manuka (Leptospermum) from one end of the islands to the other, and probably both it and Lecanochiton confine themselves to one kind of tree. Some Coccids, as the Dactylopii, or Fiorinia asteliæ, or Lecanium hesperidum, are not so particular. The antennæ of the male Lecanochiton are peculiar, and distinguish it from all Coccids known to me. As a rule, the four or five terminal joints of the male antennæ are elongated, in this insect they are nearly globular. This character, and the employment of the second pellicle in the test of the female, clearly differentiate the species.

2nd Genus, Ctenochiton, mihi. 1. Ctenochiton perforatus, mihi. (Trans., vol. xi., p. 208.) Fig. 28. I have come to the conclusion that the female antennæ of this insect have only six joints. The point is by no means easy to settle. After the second, which is very short, comes a long joint, which I have hitherto taken for two. At the extremity of this there are three or four hairs, and half-way along its length a hair springing from a very small depression, which seems to run in a ring round the joint, and this ring I have considered as a true division. I think now that it is not so, and that the third joint is really a very long one, almost equal indeed to all the rest of the antenna. Fig. 28 shows the features referred to. The point is of importance in this respect, that one of the characters distinguishing the genus Ctenochiton from Ceroplastes I have taken to be the seven-jointed antennæ, Ceroplastes having six. I have not made a thorough examination lately of the antennæ of Ctenochiton viridis. C. piperis and C. spinosus have, I think, undoubtedly seven joints. In any case, the fringed test is a sufficient distinction of the genus. 2. Ctenochiton flavus, sp. nov. Figs. 20–21. Female test (fig. 10) golden, waxy, flat beneath, convex above; outline circular or slightly elliptical, with a fringe of broadly triangular segments round the edge. Apex of the test an irregular elongated mass of wax, the remainder divided into two concentric series of plates, the inner series pentagonal with sharp angles, the outer pentagonal with rounded angles and with the outer side forming the base of the segments of the fringe. The inner series forms often irregular lumps of wax. The adult female (fig. 11) fills the test, shrivelling up after gestation: it is consequently flat beneath, convex above, with general outline of Lecanidæ. The spiracular spines, as shown in figs. 11 and 13, are very long and conspicuous: from their base a double row of minute circular spinneret orifices runs as far as the spiracle, with two or three outlying ones at the base of the spine: and I think a single row of the same kind of orifices runs across the body to the spiracle on the other side. Along the edge of the body there is a series of conical sharp spines (fig. 13): and scattered all over are many tubular projecting spinnerets as shown in the same figure. The abdominal cleft is deep, and the two lobes are conspicuous on the dorsal side: these lobes are not, as usual in the Lecanidæ generally, smooth, but irregular, and each bears at the end three or four strong spines. The antennæ have six joints (fig. 14): but the third joint often looks like two, as

Coccidæ.

here is a sort of false division near its end, with a hair or two, like that noticed above under C. perforatus. The last joint has several very long hairs. Feet (figs. 15, 16) normal of the group: the upper digitules fine long hairs, the lower pair very broad. The anal ring (fig. 17) bears a number of long hairs of which eight seem to be conspicuous. The colour of the insect is a golden brown; diameter about 1/12 inch. The second stage of the female (fig. 12) is normal of the genus, showing the wavy outline, somewhat strongly marked in many specimens but not conspicuous in others. The spiracular spines are prominent, and a row of conical spines runs round the edge of the body, as in the adult. The test is at first very thin and brittle, and with a fringe of broad, shallow, segments: but afterwards becomes thicker, and in the end, before the change to the final stage, it approaches almost the form of the waxy test of an adult Ceroplastes. The young insect is normal. The male test (fig. 18) is much narrower than that of the female, having an irregularly rectangular edge with deep curvilinear depressions. It is glassy, white and shining, flat beneath and elevated above, and marked with numerous horizontal striæ. The upper central portion is sometimes flat, sometimes an irregular mass of the glassy secretion. On the lower side there is often a plate of secretion, so that the pupa is almost entirely enclosed. The adult male (fig. 19) is normal of the genus. The legs are very long and slender; the four digitules are fine hairs. At the extremity of the tibia there is a strong spine. Abdominal spike, or sheath of the penis, slightly curved, with a seta on each side of its basal tubercle. Antennæ of ten joints (fig. 20): the first two very short, the rest longer and equal. On the last joint (fig. 21) are several long hairs, of which three are knobbed. From Brachyglottis repanda and Panax arboreum, only in the North Island, as yet. Mr. Buchanan, of the Geological Survey, has kindly sent me specimens. The shape and colour of the test, and the arrangement and form of the spinnerets, differentiate this species from others of the genus. In the prominence of the spiracular spines it resembles C. elongatus, mihi: in the row of conical spines round the edge it resembles C. fuscus, described below; but both of these are otherwise different. 3. Ctenochiton fuscus, sp. nov. Figs. 29, 30. Test of the adult female elliptical in outline, flat below, convex above, the elevation being greater than usual; almost black in colour, composed of a thin dark waxy secretion. The fringe is conspicuous, and has the

appearance of teeth, the segments being triangular and set somewhat closely together. It attains sometimes a length of nearly ¼ inch, a breadth of 1/7 inch, and a height of 1/10 inch, being thus rather a large species. The inside of the test is whitish. The adult female fills the test, and shrivels after gestation into a conical mass, requiring maceration or boiling in potash in order to make out the organs. The antennæ (fig. 29) are not long. I cannot say exactly whether there are six or seven joints (I have figured seven), as they are much confused: on the last joint are several longish hairs. The foot (fig. 30) shows the tibia broadening to its extremity, with two hairs at the tip: the upper digitules are stronger and thicker than usual, and the lower pair end in conspicuously broad plates. On the edge of the body is a row of small conical spines, as in the last species. Colour of the insect almost black. In the second stage the female is less wavy in outline than in other species of the genus, and in its later period is somewhat thick, with the edges turned inwards. Feet normal: digitules fine. Antennæ short and thick, with six joints, of which the third and fourth are the longest: on the last joint some long hairs. The abdominal lobes are irregularly triangular. Young insect normal. I have not seen the male; but its test appears to be narrow, white, and glassy. From Brachyglottis repanda, in the “Dry Bush,” near Christchurch. The black colour both of the test and the female, its large size, and the digitules of the foot, sufficiently distinguish this species. Much of the blackness of Coccids is due to the presence of fungoid growths which always accompany them (I suppose, Fumago); but in this case the colour is that of the insect. 4. Ctenochiton depressus, sp. nov. Figs. 31, 32. Test of female flat, nearly circular, thin, waxy, greyish-coloured: the fringe is inconspicuous or absent in the latest stage, but normal at earlier periods. Diameter about 1/7 inch. There are no rows of air-cells, or perforations, in the test. Adult female filling the test and as usual shrivelling after gestation: colour brownish or grey. Antennæ (fig. 31) of six joints, the third being the longest and, as is commonly the case, often appearing like two. On the last joint a few long hairs. Foot normal: the upper digitules are fine hairs, the lower pair only a little broader. Anal ring and lobes normal. In the second stage the usual wavy edge is conspicuous: the test is thin, glassy, with normal fringe. Antennæ and feet normal. The insect is somewhat thick, with yellowish colour. Young insect normal.

Test of the male elongated, narrow, flat beneath, slightly convex above, white, glassy, thin and brittle, with a conspicuous fringe of which the segments are truncato-triangular (fig. 32). The test is divided into tessellations, the median row of which is quadrangular, with two series of pentagonal divisions between it and the fringe. Near the abdominal extremity a transverse narrow slit cuts the test in two, leaving a small segment at the extreme end apparently separate. Length of the test about 1/14 inch. The adult male is yellowish-red in colour, about 1/25 inch in length, exclusive of the wings. General form normal. Antennæ long, with ten joints, all long and equal, except the two first which as usual are very short: all the joints have several hairs. Legs normal, but the tibiæ are very long and slender and only a little thickened at the tip; tarsi somewhat thick; digitules fine hairs. Abdominal spike longer, I think, than usual, and very slightly curved. Four pairs of eyes. On Plagianthus, Cyathea, and a few other plants sent to me from Hawke's Bay by the Rev. Mr. Colenso. This insect resembles, to the naked eye, somewhat nearly Ctenochiton perforatus, mihi, but the female differs in the absence of the curious perforations in the test of that species and in the shorter and thicker antennæ with also more long hairs on the last joint. The test of the male is also different. Subsection II.—Lecanieæ. Genus Lecanium, Illiger. 1. Lecanium sp., parasitized. It is by no means uncommon, especially in the North Island, to find on many trees in the forests a number of circular brown spots on the leaves, varying from 1/20 to 1/10 inch in diameter, slightly convex and with a velvety appearance. On examination these spots are found to be chiefly fungoid: they cannot always be easily detached from the leaf, and often several of them are connected together by a thin sheet of fungoid growth so that a large patch comes off at once. On turning them over, very often nothing more is to be seen than from the upper side: but many specimens may be found showing in the centre of the under-surface a small oval object embedded in the brown mass. Closer examination shows this to be, in most cases, a Lecanid insect; but identification is very difficult, and it is almost impossible to make out the organs, even after prolonged maceration and boiling. It is impossible to render the insect transparent enough for complete study. This is the effect of parasitism, and the insect in question is a Lecanium which has become covered with fungoid growth. I do not know exactly to what species to refer it. In general form it resembles L. hesperidum, but it

is much smaller, averaging only 1/30 inch in length. I do not quite see, also, why L. hesperidum, which is free from fungus in our gardens, should be so subject to it in the forests; indeed, almost exclusively so, for few other insects suffer in the same way. Fiorinia asteliæ is sometimes found in the same state, and I have specimens of an Aleurodes from Pelorus Sound also attacked by fungus; but this little Lecanium seems to be the chief victim. Subsection III.—Lecanio-coccidæ. The species forming this subsection are all European, and I have not seen any in New Zealand. 3rd Group.—Hemi-Coccidæ. I have not seen a true Kermes in New Zealand, but have received from South Australia an insect of this genus which deserves full examination. The genus Asterolecanium, which I have placed under this group in my foregoing list of genera, is to the naked eye similar to Planchonia, and it requires an examination of the abdominal region in all stages to detect the difference. But M. Signoret distinctly states that the adult Asterolecanium has the abdomen of the Lecanidæ. 4th Group.—Coccidæ. Characterized by the absence in all stages of the deep abdominal cleft, and by the prolongation of the abdomen into lateral processes, more or less conspicuous, which I have called in these papers “anal tubercles.” In some genera, as Dactylopius, these tubercles are scarcely noticeable, but may usually be made out on close examination (see Trans., vol. xi., pl. viii., figs. 19 d and 19 g). In others, as Icerya, there is so much secretion and hair on the abdomen of the adult that the tubercles can scarcely be detected. But in no case, I think, can any of the species be mistaken for Lecanidæ. As for the mentum, it is usually tri-merous; but this character is variaable, and also most difficult to make out. Genus Planchonia, Signoret. 1. Planchonia epacridis, mihi. (Trans., vol. xiv., p. 224.) I think this species may stand. As remarked above, it outwardly resembles Asterolecanium. Genus Eriococcus, Targioni. Eriococcus araucariæ, mihi. (Trans., vol. xi., p. 218.) This appears to be certainly distinct. M. Signoret informs me that he has lately received some specimens from Spain. Mr. Comstock reports it also from the United States, but refers it to the next genus Rhizococcus,

partly because it has, he says, seven-jointed antennæ. I am unable to agree with him: at least, after examination of many specimens I can only detect six, though certainly one of them sometimes appears double. In Eriococcus hoheriæ, mihi, there is no room for doubt. But this character of the antennæ is, as Mr. Comstock admits, most uncertain. It will be seen below that the two insects which I propose to place under Rhizococcus have less than seven joints in the antennæ. Genus Rhizococcus, Signoret. M. Signoret and Mr. Comstock agree in attributing to this genus the distinctions of—1st, seven-jointed antennæ; and 2nd, absence of cottony sac, at least until gestation. I am doubtful how far either of these may be really sufficient: but probably the second may be admitted: the first is of no value. I think, with Mr. Comstock himself (Ann. Report of the Dep. of Agric., 1830, p. 339, note), that it would be best to include all under Eriococcus. 1. Rhizococcus celmisiæ, sp. nov. Figs. 33–35. Female (fig. 33) deep red in colour, of oval outline, convex above and flattened below; length about 1/11 inch. The segments of the body are not very distinct. The abdomen ends in two large and conspicuous anal tubercles, each of which bears one strong and fairly long terminal seta and three other, spines (fig. 34). The anal ring has eight hairs. Antennæ (fig. 35) of six joints, sometimes looking like seven. Mentum doubtfully di-merous. The four digitules of the foot are long fine hairs. The tibia is a little shorter than the tarsus (a character exceptional in an adult insect, occurring only in this genus and Acanthococcus). The trochanter bears one long hair and two short ones. A few large conical spines (spinnerets) are scattered over the body, and a row of smaller ones, like hairs with tubercular bases, runs transversely on each segment: also some circular spinnerets. At the edge of the body, all round, is a row of the large conical spines, which are set in groups of three on the posterior segments, of four or five on the median segments, and almost continuous on the head. When the insect is alive these spines are often agglutinated with cottony secretion so as to give the appearance of a short fringe. The four spiracles are somewhat large and circular. On Celmisia, sp., from the Southern Alps. I have only one stage of this insect, the adult female before gestation, and cannot say precisely what sort of sac, if any, it forms. Rut this would be probably normal. The insect differs from the European R. gnidii in size, colour, and habitat, in the digitules of the foot, and in the grouping of the conical spines of the edge, which are only in pairs in R. gnidii.

2. Rhizococcus fossor, sp. nov. Figs. 36–38. Adult female (fig. 36) greenish yellow in colour, sometimes brown, almost circular in outline, flat beneath and slightly convex above: length about 1/15 inch. In the last stage, after gestation it becomes dark brown. The cephalic part is smooth; the remainder segmented. The abdomen ends in two very small anal tubercles, which are nevertheless somewhat conspicuous on account of their brown colour. Between them there protudes a long thick pencil of white cotton, which is resolvable into six. Antennæ (fig. 37) short, with six joints, the last joint bearing several long hairs. Feet very small; the femur rather thick: the tibia is shorter than the tarsus by about one-third: the four digitules are long fine hairs. The anal tubercles have not terminal setæ; and I can only make out four hairs on the anal ring. A row of a few conical spines, set far apart, runs round the edge of the body, but I can see none elsewhere, nor any circular spinnerets. There is no sign of a sac in any stage. In the second stage the insect is oval, flatter than the adult, and of a rich golden colour: length about 1/40 inch. The segments of the body are somewhat more distinct than in the adult. The anal tubercles are proportionately larger, and bear terminal setæ. Antennæ longer than in the adult, with six joints. Feet also longer. All round the edge runs a row of conical spines, set more closely than in the adult; and from each of these springs a long curly tube of white cotton, making a kind of fringe to the body; each tube is a little dilated at the end, and then tapers to a narrow point (fig. 38). The base of each conical spine is a somewhat large tubercle. The young insect has the general form of the young Eriococcus hoheriæ: colour yellow; length about 1/60 inch. Antennæ as in adult, with six joints. The feet are somewhat large. Anal tubercles thick, bearing a terminal seta and one shorter hair. Along the edge of the body is the usual row of conical spines, set somewhat far apart, and four other longitudinal rows are seen on the surface. The mentum is large and, I think, di-merous. This insect is viviparous. The females are often full of young larvæ, and, as these are born, they are sheltered in a cavity beneath the mother, as in some of the Lecanidæ. They do not remain there long, but soon begin their travels, and move rapidly. The male insect is red in colour, about 1/30 inch in length, undergoing its last transformation in a minute, white, cottony, oval sac. Antennæ of nine joints: the first two short and thick, the third very long and slender, the fourth, fifth, sixth, and seventh about half as long as the third, thicker and rounder, the eighth rather shorter, and the ninth very short and nearly globular. All the joints have hairs. Legs slender: the tarsus

Coccidæ.

rather less than half as long as the tibia: the four digitules are fine hairs. The usual hairs and spine on the tibia, and two spines on the lower edge of the tarsus, not far from the claw. There are three pairs of eyes. The abdominal spike is short and thick, and exhibits a curved appendage similar to, but rather longer than, that of Acanthococcus multispinus, mihi (Trans., vol. xi., pl. viii., fig. 18 f). This appendage is, indeed, common to the three genera Eriococcus, Acanthococcus, and Rhizococcus. At the base of the spike are two rather strong setæ, one on each side. Rhizococcus fossor does not construct a cottony sac (for the female), but, instead, buries itself usually in a circular hole or pit, in the leaf it lives on. Many adult insects may be found simply resting on the leaf, accompanied by a number of young larvæ and females in the second stage. But in most cases they pass their last stage in a pit. The young insect is very active: the female of the second stage moves about sluggishly; in the last stage it is fixed and stationary. At first it appears simply to lie on the leaf (on the under-side), but in a short while, whether from some chemical action produced by it or by mechanically compressing the cells of the plant, a circular elevation or wall on the leaf is raised up round the insect. At the same time the portion of the leaf beneath the body is pressed downwards, and a corresponding elevation appears on the other side: this elevation assumes a brown tint. As the depression continues and the wall grows the insect sinks deeper in the leaf, becoming more and more buried, and the wall curls over it a little, so that in fact the orifice of the pit is somewhat smaller than the cavity below. The insect lies in the pit (which may average about 1/18 inch in diameter at the opening, and is circular) with the head downward, and the anal tubercles and pencil of white cotton appearing over the wall, I presume to attract the male. Afterwards, as gestation proceeds, the whole body disappears in the pit, where the young larvæ are born. These cavities in the leaf look like small volcanic craters, and the corresponding brown elevation on the other side of the leaf is quite conspicuous. Sometimes two insects may be found in the same pit, one lying over the other. I suppose the upper one simply took advantage of a ready-made domicile. The insects which are not in pits are generally darker in colour than the others. On leaves of maire, Santalum cunninghamii, from the North Island, sent to me by C. P. Winkelmann, Esq., of Te Aute. This is an interesting and peculiar insect, certainly differing from any that I know of in the genus. The curious mode adopted by it of burrowing into the leaf is, I think, unique. Ctenochiton viridis, mihi, produces a certain depression in the leaf it lives on, but by no means so complete a shelter for itself as does this little Rhizococcus.

Genus, Dactylopius, Signoret. (Trans., vol. xi., p. 218.) 1. Dactylopius alpinus, sp. nov. Figs. 39, 40. Adult female dark purple in colour, the body thick and fat, inactive, enclosed in a mass of white closely felted cotton. When immersed in alcohol it produces a rich red tint. Length sometimes as much as 1/7 inch. The internal substance is very oily. The body is normally segmented; anal tubercles inconspicuous. Anal ring large, with six hairs. Antennæ normal of the genus, with eight joints (fig. 39). The feet are normal; the two upper digitules are long fine hairs, the lower pair are somewhat broader. Mentum di-merous. All over the body there are a number of tubular projecting spinnerets, and many others circular; and on the three posterior segments are three rows of large conical spines, similar to those of Rhizococcus. Although the anal tubercles are not as prominent as in some of the Dactylopii, they are somewhat thick, with broad bases. Young insect dark brown in colour, of generally normal form, about 1/40 inch in length. Antennæ (fig. 40) of six joints. Feet normal. Anal tubercles thick and broad, and rather more prominent than in the adult. There are a few small spines on the body. In the second stage the insect generally resembles the young larva, but is larger, and the body is covered with a great number of small circular spinnerets intermixed with others which have a bulbous base and fine hair-like tubes. Anal tubercles inconspicuous, bearing the usual spines. A few conical spines appear on the posterior segments. Antennæ of six joints. I have not seen the male. On a species of Veronica, sent me by Mr. J. D. Enys, from the upper valley of the Waimakariri, in the Southern Alps, near the glaciers. The large conical spines, or spinnerets, on the posterior segments, distinguish this species, besides its size and very deep purple colour producing a rich tint in alcohol. The thick sac of white cotton in which it envelopes itself may probably be also an important character: in no stage, I think, is it like the ordinary “mealy bug,” the type of Dactylopius. 2. Dactylopius calceolaria, mihi. (Trans., vol. xi., p. 218.) Since first describing this insect I have received specimens from the forests of Stewart Island, in the extreme South of New Zealand, on a native grass, Danthonia, and on a leaf of Phormium. As Stewart Island is almost destitute of cultivation, and is certainly the last part of the colony where one would expect to find in the forests a new importation from other countries, I take it that D. calceolariæ is undoubtedly indigenous.

Some of the specimens on Danthonia were more than ¼ inch long, so that the insect is quite a large one. 3. Dactylopius glaucus, mihi. (Trans., vol. xi., p. 219.) I consider this also a good species, distinct from the last by its green colour and small size, which never exceeds 1/12 inch. I have obtained one mutilated specimen of a male. As far as I can make it out it offers no very distinctive character, except that the abdominal spike or sheath of the penis seems to be accompanied by a curved appendage as in Acanthococcus multispinus, mihi (Trans., vol. xi., pl. viii., fig. 18 f). Genus Pseudococcus, Westwood. A genus separated from Dactylopius by having nine joints in the antennæ of the adult female, and only two digitules on the foot. Westwood's original genus is described as “having females not fixed and clothed with a woolly secretion” (Int. to Mod. Class of Insects, vol. ii., appendix, p. 119), characters which, in point of fact, would include the whole group Coccidæ. M. Signoret has confined its limits as above. 1. Pseudococcus asteliæ, sp. nov. Figs. 41–44. Adult female about 1/10 inch long, yellowish-brown, covered with a not very abundant white cotton. General form resembling Dactylopius: the body segmented, anal tubercles inconspicuous, anal ring with six hairs. Antennæ (fig. 41) with nine joints, of which the third, fourth, and fifth are the longest; the second, sixth, and ninth a little shorter; the first, seventh, and eighth the shortest. The fourth, fifth, and sixth are the narrowest, the two ends of the antenna being thicker than the middle. The eighth joint is a little expanded at the tip; and the ninth is fusiform, with a shallow depression at the extremity. All the joints have a few long hairs, and on the eighth is one a good deal stronger than the others. The legs (fig. 42) have the tibia twice as long as the tarsus: the claw is slender, and has no tooth on the inner edge. There are only two digitules (the lower pair) which are long and fine. The trochanter bears one short bristle. The whole leg is slender and long. The eyes (fig. 43) are tubercular and smooth, showing after maceration in potash a small dark terminal spot. The body is covered with a number of spinnerets of two kinds, as shown in fig. 44; those with simple concentric circles are the largest, and are found all over the integument: the others are placed in groups at the edges of the segments and also in great numbers at the cephalic and abdominal extremities. Interspersed with these spinnerets are several hairs, mostly very short, but on the head are some pretty long. From the

anal tubercles spring two strong setæ with tubercular bases, not very long. The mentum is di-merous and bears a few hairs on the tip. In the groups of spinnerets at the edges of the segments are found a few small conical spines. The four spiracles are small and simple. I have not seen the male. On a species of Astelia sent to me by the Rev. Mr. Colenso, from the forests of Hawke's Bay. This insect appears to be most nearly allied to Pseudococcus mespili, Geoffroy, found in France on medlar and plane trees. The arrangement of the spinnerets in groups on the segments and in great numbers at the two extremities, the long hairs on the head and the length of the tibia, are similar in both. I find, however, in M. Signoret's description of the European insect no mention of the two kinds of spinnerets. The main differences between the two seem to be in the antennæ and the foot. In P. mespili the second joint of the antenna is the longest and the rest diminish gradually to the ninth: and the claw of the foot is strong and broad and shows a small tooth on the inner edge near the point. In P. aceris, another European insect, the mentum bears at its tip a large number of hairs. Both of these insects, too, are red in colour. Genus Icerya, Signoret. 1. Icerya purchasi, mihi. (Trans., vol. xi., p. 221.) Through the kindness of M. Signoret I have had an opportunity of comparing this insect with I. sacchari, the Mauritian species, and I find that it is undoubtedly and markedly distinct. I. sacchari does not seem to form an ovisac with longitudinal grooves, nor does the body of the insect, although somewhat hairy, show the great tufts of black hairs and the curious projecting glassy tubes springing from large brown “coronetted” bases which are marked features of I. purchasi. The number of the circular spinneret orifices is also much smaller in the Mauritian insect. Icerya purchasi has spread greatly in the last two years. It had just reached Napier at the date of my last paper; it has now established itself in that district, not only in gardens but in the native forests. In Auckland it is attacking all sorts of plants, from apple and rose trees to pines, cypresses, and gorse, and it is spreading over a large district. It has reached Nelson, and I have had many communications from that place complaining of its ravages. From the “Nelson Colonist” I learn that it is devouring wattles, cypresses, gorse, and many other plants. At the same time nobody seems to try to destroy it. With the example of California and the Cape of Good Hope before us we may be sure that ere long this pest will become a dreadful nuisance. I have tried to warn the people of

Nelson and Auckland, but no remedy has been attempted. The best authorities in America have come to the conclusion that the only cure is the destruction of infected plants. Some day the people of New Zealand will have to find this out also, but the longer it is delayed the worse the work will be. Whether this pest will spread in our colder southern climate as it has in the warmer north remains to be seen. Our gardeners here are not in much dread of outdoor insects, they confine their attentions to those in greenhouses. They may be right, still the winter even in Canterbury is not severe enough to kill these insects, and I know that in the Christchurch public gardens many trees have had to be burnt simply on account of the ravages of Coccidæ. If Icerya purchasi, as seems likely, makes it way down here, I very much doubt whether gardeners will find it an easy matter to keep it in subjection. Genus Cælostoma, mihi. (Trans., vol. xii., p. 294.) 1. Cælostoma zælandicum, mihi. (Trans., vol. xii., p. 294: vol. xiv., p. 226.) The haltere of the male in this insect, which I had at first taken to resemble that of Porphyrophora, has, as I find, four terminal setæ, and generally resembles that described and figured below, under my next species. 2. Cælostoma wairoense, sp. nov. Figs. 45–51. Male insect so nearly resembling C. zælandicum in outward appearance that it may be very easily mistaken for it, having a red or purplish body about 1/6 inch long, bluish wings with strong red nervures, prominent eyes with numerous large facets, and antennæ of ten joints. The antennæ (fig. 45) are, however, more slender than in that species, and have fewer and finer hairs. The legs also have fewer hairs, especially on the tibia. The claw differs entirely, not only from C. zælandicum, but also from every other species known to me. Instead of having two, or four, or no digitules, it shows a great number of them, forming indeed a sort of brush round the claw (fig. 47). I have counted as many as twenty-four of these digitules, each of which springs from the claw itself, the tarsus not exhibiting any. All the digitules are knobbed. On the tarsus there are several strong spines on the inner edge. The sheath of the penis and the penis itself resemble those of C. zælandicum. The haltere (fig. 48) has an inflated sac-like body diminishing to a cylindrical tube, from the extremity of which spring four strong curved setæ of no great length; very frequently only three are seen. The abdomen has several segments, and each segment bears a number of short fine hairs springing from a minute tubercular base; the hairs are intermixed with numerous circular spots, as in C. zælandicum,

but in that species these are simply two concentric circles, whereas in C. wairoense they exhibit (fig. 49) an outer circle with six smaller ones within it. The male pupa is bright red, and is enclosed in a small cylindrical sac of white cotton slightly tinged with red. Apparently these sacs are found in great numbers. A mass of several hundreds together was sent me on one occasion, and, after about six weeks (in the depth of winter) two or three hundred males emerged. Amongst this mass of male puparia were a few (perhaps half a dozen) objects which I took for females, at least before the hatching of the males. They certainly were not male pupæ, but I cannot exactly make out what they are. They were much shrivelled, but on maceration in potash regained their original form. Length about 1/7 inch: the body elliptical, segmented, generally resembling C. zælandicum; colour dark purple. Antennæ (fig. 50) of nine joints, all nearly equal in length, the first two somewhat broad, the rest narrowing to the fifth and then widening again to the last, which is almost globular. All the joints are short, and on each there is one ring of hairs with tubercular bases, forming a sort of crown. The eyes are small, tubercular and smooth, conical and a little projecting, with a small terminal spot. Legs (fig. 51) strong and thick. The trochanter bears one very long hair. Tibiæ somewhat dilated at the extremity, with a few spines on the inner edge. Tarsi thick, tapering towards the claw, which is normal in shape, and has two lower digitules, which are long, fine, knobbed hairs. Anal ring oval, without hairs. There is no sign of anal tubercles. The segments of the body bear a number of longish fine hairs with tubercular bases, interspersed with circular spinneret orifices. These last show two concentric circles and a central spot. The hairs and spinnerets are most numerous at the two extremities. There are twenty-four spiracles, i.e., one at each side of each segment; the spiracles are simple, and the tracheæ small. There is not the least sign of a mouth, neither rostrum, mentum, or rostral setæ. I should have mentioned that the tibiæ are twice as long as the tarsi. I have not seen any specimens which I can take to be adult females, nor any young larvæ. I presume that the adult female will have antennæ of eleven joints. My specimens were sent to me by T. Cheeseman, Esq., of Auckland, who informs me that the insect is found in the District of Wairoa, attacking Phormium and Leptospermum. There is no doubt that this is a distinct species. The antennæ and legs of the male differ from those of C. zælandicum, but the claw of the foot is quite sufficient to distinguish it from that and all other Coccidæ. I have

therefore no hesitation in considering it as new, even without seeing the adult female. As regards the stage last described, which is certainly not the male pupa, I have been in some doubt. The specimens were found amongst several hundred male puparia; but there were only five or six of them to be seen, though I made diligent search. A very similar form is found amongst male puparia of C. zælandicum—so similar indeed that, except in colour, the two are almost identical. In my paper of 1879 (Trans., vol. xii., p. 296), I described, or rather alluded to, this form as the second stage of the female C. zælandicum. I afterwards discovered the error and corrected it in vol. xiv. Unless this form be the second stage of the male insect I do not know what it can be; but if it is, then the male of Cælostoma must pass through more transformations than any other Coccid. The normal stages are—1, the egg; 2, the young larva (identical, or nearly so, for male and female); 3, the pupa; 4, the perfect insect. Now, the form under consideration is certainly not the pupa, and equally certainly not the young larva just hatched; it is neither the egg nor the perfect male. Similarly (in C. zælandicum) it is not the larva, nor the adult, nor the second stage. It would seem therefore that the insects of this genus pass through a transformation more than other Coccids. There is one point to be noted. I have observed above that in this form the antennæ have nine joints, and the tibiæ are twice as long as the tarsi. The first character is conclusive against the insect being adult; the second character is one which, in most Coccids, is considered to indicate an advanced stage. M. Signoret (Ann. de la Soc. Entom. de France, 1874, p. 548, note) says: “In studying the Coccidæ it should be noted that almost always, when a specimen is found with the tibia shorter than the tarsus, it is a larva.” The rule is not without exceptions, e.g. Kermes, Acanthococcus, etc. Still, it holds good generally, and in this case shows, I think, clearly that the form in question is by no means the first after the egg. I take it to be an intermediate state preceding the pupa of the male. I am still puzzled by the absence of any sort of mouth. In the form just mentioned there is an orifice between the second pair of legs, as in the adult C. zælandicum, but nothing more. Description of Plates I. and II. Plate I. Fig. 1. Aspidiotus sophoræ, abdomen of female × 90 2. Fiorinia minima " × 100 3. " diagram of female in puparium: a. 1st pellicle: b. 2nd pellicle: c. adult female: d. secretion 4. Fiorinia grossulariæ, abdomen of female × 50 5. " stricta, abdomen of female × 100 6. " diagram of female in puparium: a. 1st pellicle: b. 2nd. pellicle: c. adult female: d. secretion

Fig. 7. Fiorinia stricta, antenna of male × 60 8. " foot of male × 100 9. " spike of male × 100 10. Ctenochiton flavus, dorsal view of test × 10 11. " female × 10 12. " female, 2nd stage × 10 13. " spiracular spine and spinnerets × 200 14. " antenna of female × 150 15. " foot of female × 150 16. " " to show digitules × 300 17. " anal ring of female × 100 18. " test of male, dorsal view × 10 19. " male × 30 20. " antenna of male × 60 21. " last joint of antenna of male × 250 Plate II. Fig. 22. Diagram of abdomen of Diaspidæ. 23. " Lecanidiæ. 24. " Coccidæ (Eriococcus). 25. " Coccidæ (Dactylopius). 26. Lecanochiton metrosideri antenna of male × 90 27. " foot of male × 90 28. Ctenochiton perforatus, antenna of female × 70 29. " fuscus, antenna of female × 90 30. " " foot of female (two digitules shown) × 90 31. " depressus, antenna of female × 90 32. " " fringe of test of male × 20 33. Rhizococcus celmisiæ, adult female × 10 34. " abdomen of female × 30 35. " antenna of female × 90 36. " fossor, adult female × 12 37. " " antenna of female × 90 38. " " spines and spinnerets, 2nd stage × 200 39. Dactylopius alpinus, antenna of female × 90 40. " antenna of young insect × 90 41. Pseudococcus asteliæ, antenna of female × 90 42. " foot of female × 90 43. " eye of female × 200 44. " spinnerets of female × 500 45. Cælostoma wairoense, antenna of male × 40 46. " foot of male × 40 47. " claw and digitules of male × 180 48. " haltere of male × 40 49. " spots and hairs of male × 350 50. " antenna, 2nd stage of male × 90 51. " foot, 2nd stage of male × 90