Edward Newman

It may be thought a strange propensity to grapple with difficulties, that leads me to select Lepidoptera as a class, by which to exemplify, in detail, the septenary and circular arrangement. There is no class so puzzling to systematists, or for which science has done so little—no class is at present so badly arranged, and in none are barbarous combinations so much in vogue. LinnÆus founded divisions at the outset, on characters, "loose, vague, and insufficient:"[32] modern genera have a little improved minor details, and but little, for their places appear to have been assigned them by lot, and without the slightest regard to similarity or approach: in a word, the arrangement of Lepidoptera appears to have been conducted by collectors, who aimed rather at a pretty picture than a related series; and all our writers have rushed headlong by the same path, without staying an instant to consider whether they were right or wrong, like boys playing at follow-the-leader,[33] each occasionally leaping some wider gap, or descending some more dangerous precipice than his predecessor, as though for the very love of frolic and bravado. One, a talented writer, an assiduous collector, a most accurate observer, hesitated awhile, it is true, and it was thought he would have broken the line, but no,—he kept precisely to the track of LinnÆus and the rest of them, through Papilio, Sphinx, Bombyx, Noctua, Geometra, Pyralis, Tortrix, and Tinea; but, as he stood pledged to traverse no more than five fields, he hit on the ingenious expedient of asserting roundly, that the four last named were but one. In fact, the whole of this immense class presents, at this hour, nothing but a vast chaos, which seems to await the operation of some predicted spell to call it into order.

SUB-CLASSES OF LEPIDOPTERA.

These were apparent difficulties only; for, as no system existed to direct, so none existed to encumber or perplex. Too much is known now of LinnÆan combinations, to assert, that he always thought correctly; and since his day no one has thought at all. Now, if you cannot obtain a nicely drawn plan, you prefer having a blank sheet of paper to one covered all over with scorings and markings, and then you may set to work and make your plan yourself. So, in natural history, contrariety of opinion perplexes, while the absence of opinion leaves the systematist perfectly unbiassed in the formation of his own. Again, copious and well-named[34] collections of this favourite class are by no means uncommon;[35] and through, the liberality of my friends, I had often been permitted to inspect them, and had gained a sufficient superficial knowledge of their contents, to be enabled, with the assistance of my own specimens, to cluster them pretty well into seven great families or sub-classes; and although, as I have noticed, nothing available existed on the subject of arrangement of Lepidoptera, either in essay, or treatise, or catalogue, or cabinet, yet there was to be found, up and down, much valuable matter, in the shape of what might be termed natural history of Lepidoptera. Finally, I knew, that could I master this class, I could stand my ground, because I had previously tried the experiment on the classes Hymenoptera and Coleoptera, and had found that, at the word seven, they fell into instant order, as at the touch of a magician's wand; and, as for the other classes, we are in such innocent and blissful ignorance of their contents, that were a scheme ever so futile, a century at least must elapse before its futility could be proved by Orthoptera, Hemiptera, or Diptera.

Whoever will give himself the trouble to examine thoroughly a collection of British Lepidoptera, will find a very great majority of them evincing very evident symptoms of relation to one or other of the following species:—Papilio Machaon, Sphinx Ligustri, Pyralis verticalis, Tinea pellionella, Noctua pronuba, and Geometra roboraria; and should any form widely different from either of these occur, it may, if the larva be known, be placed in the centre of a ring formed by the groups, which we will suppose surrounding their six respective types; or, if its larva be unknown, it must await the discovery of that most unerring stage of its existence. I am persuaded, did entomologists know how much depends on the form, habit, food, and clothing of larvÆ, they would not be so neglected as they are at present. I have much to regret my own remissness in this respect, for it has seldom happened that I have found the larva of any insect which had not been previously well known, but it has tended to point out some approaches that had never before been thought of,—approaches, even when thus pointed out, totally irreconcilable with existing ideas of arrangement and combination of groups, but which now open to my view the most beautiful chains of affinities; and wonderfully but indubitably prove, that a single individual may be related to three, four, or even more apparently disconnected groups.

Perhaps no better genus was ever formed than Papilio of LinnÆus; its diurnal flight, its erect wings, and its clavated antennÆ, at first bid defiance to the systematist who attempts to bring any other group into contact with it; indeed, in Britain we have nothing at all that will avail us in this respect, which compels me to have recourse to exotics, an assistance which I shall only avail myself of when I find it quite impracticable to furnish the approaches from indigenous species, the reference to which is attainable by every entomologist. Among foreign Papiliones, especially among those groups which approach our genera, Hesperia, LycÆna, Polyommatus, and Thecla, there appears to be an almost infinite variety of form. Now it is but reasonable to seize on any variations observable in genera or species from the prominent or typical genus or order from which they may be supposed to derive their more conspicuous character, and to employ such variations in arrangement as connecting links between the group to which they more decidedly belong, and the group to which, by such variation, they evince an approach: a precisely intermediate species or genus between two classes or sub-classes, or even orders, I have never met with, notwithstanding the renowned LinnÆan maxim, that Natura saltus non facit; for did nature make no leaps, surely the question were immediately at rest as to the existence of any other division than species among created beings, a conclusion which even the most strenuous supporters of the LinnÆan dogma decidedly resist. Among the Papiliones, this departure from the type may be looked for either in the form of the antennÆ, the position of the wings, or the time of flight. The first is obviously the most tangible should it occur, and it does occur. In Urania, the antennÆ have become setaceous; the club has entirely disappeared, yet the other peculiarities remain much as in Papilio. This single deviation may be assumed as pointing out a relation to Geometra, which the reader will perceive is supposed to meet the sub-class Papilio at this point. A second peculiarity is to be found in an insect figured by Godart, a Polyommatus in shape, but with pertinated antennÆ;[36] the genus he has very suitably named Barbicornis. This deviation, it must be observed, is in favour of the Bombyces, which we therefore suppose touching the sub-class at this point. A third deviation, of a very different kind, is observable in an insect which Latreille has figured in the RÈgne Animal, and placed among the Sphinges: he calls it Coronis D'Urvillii. The antennÆ in this genus, as in Castnia, are gradually incrassated, and they may probably be eventually both considered as Papiliones: of Coronis D'Urvillii, I cannot entertain a doubt, as the wings are too expansive, the antennÆ too long, the abdomen too short for it ever to retain its station among the Sphinges; the inferior wings are also very decidedly caudate, a common formation among Papiliones, but unknown among Sphinges; but, let this question be eventually decided pro or con, the approach between HesperiÆ and Sphinges is not likely to be disputed, nor the fact that it takes place somewhere in the neighbourhood of the genus Castnia.

The next type is Sphinx Ligustri; and here again our British collections are obliged to plead poverty; few, however, as they are at present, I am compelled, if I purpose consulting nature, to reduce them about half: the ÆgeriÆ and ZygÆnÆ must be moved elsewhere; they look like Sphinges, but are none. I will begin then with Castnia, of which no more need be said. The next striking departure from the type occurs in having the abdomen furnished with tufts or brushes, which the insect spreads as it hovers over flowers, somewhat in the manner of a bird's tail. The long porrected antlia also has a resemblance, perhaps rather fancied than real, to the slender bill of a humming-bird, whence the tribe has received with us the name of English humming-birds. The genus Sesia I will place on the circumference of the circle, not doubting but nearer approaches to the Cossi may be discovered, or are even now known, but no better exists among our own Sphinges. The next point of contact will be with Pyralis; and here the genus Œgocera, figured in the RÈgne Animal, seems to claim its station: it is a decided Sphinx, with the palpi of Hypena proboscidalis, and Latreille has placed it between Sesia and ZygÆna, from which it will be seen that I differ only in making ZygÆna pass over the boundary line and into the next section.

We enter the third sub-class then at Pyralis, and find ourselves among some of the most beautiful little creatures in existence—sylph-like beings, which spend their lives in the brightest sunshine and among the sweetest flowers. LinnÆus considered them Sphinges, from what character is not very apparent: the sub-character, applicable only to this section, is certainly correct; they are truly "larva diversÆ." As for the antennÆ being "medio crassiores," it is not the case, unless the increase and decrease of pectination can be considered as making them so. Of this particular tribe Latreille observes, "Les autres lepidoptÈres de cette division ont dans les deux sexes, des antennes garnies d'un double rang de dents alongÉes ou bipectinÉes. Ceux qui out une trompe distincte forment le genre Glaucopis; ceux oÙ cette organe manque ou n'est pas distinct celui d'Aglaope—ces crepusculaires semblent se lier avec les Callimorphes." The approach of the genus Aglaope to Aglossa, rather than to Callimorpha, seems to be presumable from the circumstance of its not possessing a tongue, the genus Glaucopis having more similarity to our genus Pyrausta, while some of its species, which appear to call for further generic division of the order, are closely allied to our Botys literalis, &c. The only British genera of this order are ZygÆna and Ino; the latter, however, appears to be merely a species of some extra-European genus, as I have remarked several exotics of precisely similar form. The insects of this order have a stout and rather hairy larva, much like those of the generality of the sub-class, and in no respect allied to that of the Sphinges. Early in the summer they spin a glossy silken cocoon, generally attached to blades of grass, and remain but a few days in the pupa state. A great proportion of the perfect insects have hyaline spots and patches in their wings, and nearly all of them are brilliantly coloured. It is known that LinnÆus occasionally, as in Tenebrio and the present instance, made his genera recipients of species, which he found a difficulty in locating properly; but it is really astonishing to find a naturalist like Latreille abiding by so absurd a combination as the contents of the LinnÆan genus Sphinx, and, in servile imitation, calling creatures which nothing but an unclouded sun ever tempts abroad—Crepuscularia.[37] It is no part of my present plan to assign names to orders, or to describe their contents, except in those particular instances in which the more immediate object of this Essay may render it imperative. I will, however, just observe, that I by no means consider ZygÆna the type of the order, but merely the nearest point of contact with Sphinx, and an evident departure from its true type, which perhaps may be found in that ill-divided genus Glaucopis, the form and appearance of which is altogether more Pyralis-like than ZygÆna. I am well aware that Œgocera and ZygÆna do not harmonize so beautifully as many other approaches, and fully expect to see the connexion between these sub-classes much improved; but I have seized on these genera as demonstrating a tendency in each individual towards the sub-class to which it does not belong. The circumstance of ZygÆna having been so long considered a Sphinx will warrant its situation on the very circumference of the circle which contains its order, until a more appropriate occupant of that situation can be found. At the central point of contact, the genus Aglossa presents a very Bombyx-like appearance; its shape, its want of the antlia, &c. indicate approach; and from the sub-class Tinea, the division of Pyralis is at present an imaginary one: at this point, after making what little comparison I am able, I am induced to place Galleria, Melia, and Ilithya, in Pyralis; and Chilo, and Crambus in Tinea.

The fourth sub-class, Tinea, far exceeds in numbers either of the others, and probably all of them together; and where such a multitude of species exists, great diversity in form and habit may be expected: the Pterophori are a most singular tribe, and greatly resemble the TipulÆ in many respects. I feel by no means certain that their situation would not be better between the lepidopterous sub-class, Tinea, and the dipterous sub-class, TipulÆ, thus throwing them completely out of the lepidopterous circle; but this I leave. I am now only sketching a rough and hasty outline from nature. If I attempt to finish my drawing as I proceed, I shall find occupation sufficient for a lifetime. I have observed that I considered the chain of relation entering from the last sub-class at Chilo, or about that genus; the same order must of course include Crambus, and its congeners; the next order will contain Yponomeuta, which I will place at the point of contact; and the next point being among the true Tortrices will drive Halias fagana as a decided departure from their typical form to the very circumference of the circle where it touches Noctua.

The fifth sub-class, Noctua, seems to be but one mighty genus: we will enter it from Halias fagana, an insect so nearly allied to Noctua in its larva, its pupa, and its imago, that for a long time I hesitated to which sub-class it belonged; again, in Cymatophora,[38] subtusa and retusa, I was fearful that by considering them NoctuÆ, I might deprive the order Tortrices of a genus on which perhaps many curious combinations might depend, and I now only place them in NoctuÆ until I may have an opportunity of examining their larvÆ, which I have not yet been fortunate enough to meet with. Towards the central sub-class there appear to be many genera which approach the line of contact; Agrotis and Chareas for instance:[39] I prefer taking the latter, and must mention the species Graminis, as I am fearful of encumbering my system with species to which I not only never intended to refer, but should probably place in some distant order, or perhaps sub-class. At the approach to Geometra, the genus Catocala, from its looping larva, seems to have a right to be placed: this I, however, look on with suspicion, as the larva appears to me any thing but a guide in the connexion of sub-classes; but I here succumb to customary usage in making this genus the approach to the real loopers, objecting, however, to the intervention of Phytometra, Euclidia, and Brepha.

The sixth and last of the exterior sub-classes is Geometra, and we shall find one insect which is completely a Geometra, and yet in the larva has two additional feet, and the abdominal fringe of Catocala: this is Metrocampus margaritaria,[40] an insect, without which the connexion of these sub-classes would have been difficult to establish. The next species I am acquainted with seems to be Rumia cratÆgaria, and after it the Thorn moths, as they are termed (Crocallis?): these lead to Geometra[41] in the centre, which may be considered the farthest removed from any of the surrounding sub-classes; from the genus Geometra a line may be drawn through Biston, Nyssia, and Hybernia, to the point of contact with PhalÆna in the centre, and another through Boarmia, Abraxas, and Ourapteryx to Urania, from which genus of Papiliones perhaps the reader will recollect we set out.

The seventh and central sub-class, PhalÆna, now claims our attention. The mere circumstance of having taken a little tour round it gives but a very poor idea of its contents, and although my reader may assure me he knows them sufficiently well already, that assurance will by no means satisfy me that he and I are at all agreed either as to what those contents may be, or as to their relative situations. Before, however, I again set in earnest to the task of pointing out relations and approaches, I feel that some apology is due for attempting the restoration of a beautiful and euphonious name to that grand group of Lepidoptera, to which it was originally assigned by the eminent naturalist who was the first to define and name such groups.[42] I am fully aware this is an attempt at innovation for which I can never be forgiven by the scientific; for the merit of the present day seems to consist in the total neglect of grouping and classifying, and in making a host of imaginary genera and species, for the mere pleasure of overwhelming us with a "farrago" of barbarous and unutterable names,—a practice which my unsophisticated and old-fashioned notions will never dwell on with that deferential awe which such profound science has an undoubted right to expect.

Again, on the subjects of orders, a term I have already been induced to use now and then, I am quite aware that I here am guilty of another misdemeanour, and more especially as I call them natural orders, meaning thereby orders among the contents of which nature has established the similarity; and to the formation of which "the cunningly devised fables" of man have contributed but very little; and meaning also that nature has implanted in us all, more or less, the power of distinguishing such orders by a mere glance, and without any reference to our books.

Furthermore, the naming of orders which I have been obliged to mention by name, in the unscientific way which I have adopted, merely making them plurals of established names, of large and overgrown genera, I acknowledge to be a confession of ignorance not usual in this our day, especially as these old genera have almost in every instance the disadvantage of being euphonious, easily pronounced, expressive, and universally understood; and an opportunity once missed of coining names for three hundred new orders, (and each might have been a combination of consonants which no one could spell, or speak, or read, or understand,) alas! alas! may never occur again.

To return; I suppose the sub-class PhalÆna to contain seven natural orders, a number precisely similar to that discovered from observations made on the larva by that most accurate and indefatigable naturalist, Dr. Horsfield;[43] and I may add, my own divisions are derived from the same source, together with the pupa and whole habit: the perfect insect has no characters, hitherto discovered, by which we can ascertain either sub-class or order, and from this circumstance I am compelled to omit those genera of whose larvÆ I am ignorant,[44] and even to leave those as doubtful, of which I possess but a partial knowledge of that state.

Natural Order—Bombyces. Has an elongate cylindrical downy larva, which rolls itself into a ring when touched; the pupa changes in a close gummy oval cocoon, remarkably small for the size of the imago. Among the exotic species of Lasciocampa, we find in the males particularly slender bodies, expansive wings, the inferior grooved to receive the abdomen, and diurnal flight, all of them characters so indicative of an approach to Papilio, that we scarcely hesitate a moment in assigning it the approaching station, not but I expect fully that time will eventually furnish us with a connexion on each side yet more conclusive.[45] The second genus of Bombyces is probably Odenestis, and the third Gastropacha, whose prominent and elongated palpi appear to point out an approach toward a tribe of insects with the same peculiarity, of which there are several to be found in the following order:

Natural Order—Cossi. The larva is depressed; naked, except a very few scattered hairs; has sixteen feet; lives through one or more winters; never rolls itself in a ring when touched; feeds on the solid interior woody parts of vegetables. The pupa generally changes in a tough oval cocoon, interwoven with particles of its food. It has a double ring of raised denticulations of each segment of the abdomen, by means of which it is endowed with a considerable power of locomotion. The genus Zeuzera is very near the point of contact with the Bombyces. In Zeuzera there is much resemblance to the antennÆ of Gastropacha. One genus, or group of genera, I expect will prove to be Stygia, a native of New-Holland. A second, at the point of contact with Sesia in Sphinx, must be Ægeria; thus retaining its place among British insects, immediately between Sesia and Cossus.[46] This is the first of a series of the most beautiful instances of approach, or rather, of what ought to be termed relations of analogy, that any system has ever previously disclosed. As a few words will again be necessary on this subject, I refrain from any further observation here, than merely requesting the reader to examine how minutely the Sphinx characters are appropriated by a true lignivorous PhalÆna, which cannot be said, in any of its prior and principal states, to have the most distant approach to Sphinx. A third genus is, probably, the strange and paradoxical exotic Oiketicos, which has been minutely described in the LinnÆan Transactions; and a fourth is Hepialus.[47] This genus has some slight points in which it differs from the others of the order already known, the larva being radicivorous only, seldom or never ascending internally the stems of plants: it changes in the earth.

Natural Order—NotodontÆ. The larva is naked, has sixteen feet, and is, in different genera, furnished with excrescences, and apparent distortions in various parts of the body. The eighth or last pair of feet, and three last segments of the abdomen, are elevated; when the insect is at rest, the head and first segment are raised in a similar manner. In one genus (containing Camelina) the head and extremity of the abdomen nearly meet over the back, when raised in this singular manner. The posterior feet are frequently useless in walking; in some genera, entirely obsolete. The pupa is smooth, in a cocoon, mostly among dead leaves on the surface of the ground: sometimes it is glutinous, and interspersed with fragments of wood, like the last. I confess I am exceedingly puzzled both with the contents and extent of this order; but this arises from my having seen so few of the species in the larva state. Ptilophora plumigera, figured by Mr. Curtis,[48] I had always considered a Notodonta; but the larva evidently excludes it from the order, and, I should imagine, places it among the NoctuÆ; where among them I know not, for I have not the slightest idea of any congeners, either of the larva or imago. The larva from which a collector of Lepidoptera could expect to obtain such an imago would be unicolourous, stouter in the middle, elevated in the penultimate segment, and more attenuated towards the head.[49] PygÆra appears doubtful at first, but when observed quite at rest, and in a perfectly natural position, elevates the head and tail, though in a much less degree than the typical genus. Mr. Curtis's genus Notodonta contains several good species, which may be considered as typing the order, as Ziczac, Tremula, and DictÆoides. The first species, Trepida (the Peridea serrata of Mr. Stephens,) seems more nearly related to Endromis. Both these may, however, probably belong to the order NotodontÆ, and be situate near the approach to the central order PhalÆnÆ. Petasia cassinea and Episema cÆrleocephala appear to be genuine NoctuÆ, and very near Chareas graminis, and Rusina ferruginea, as far as my very imperfect knowledge of these four species will allow me to judge. Clostera is another departure from the type; but this may be accounted for, in some degree, by its close proximity to Hepialus, from which genus it borrows its remarkably short antennÆ, and other peculiarities. It seems a strange perversion of judgment to place Platypteryx at the end of or among the GeometrÆ; but LinnÆus did so, and that is enough. Hubner, Haworth, and a few others, positively ventured, in this glaring instance, to refer this genus to the Bombyces; but their ideas were thought to be wrong, and their judgment was, nem. con. reversed. I have elsewhere expressed a wish that my readers should convince themselves, and the frequent occurrence of the larvÆ of Platypteryx and Cerura would afford any naturalist abundant opportunity of ascertaining, that they can be referred to but one order. The approach of Platypteryx and Cilix to the Pyralides, in assuming so much of their characters, is very interesting, and is a most striking departure from the typical form. These genera also approach the LithosiÆ, but not so nearly as some exotics.

Natural Order—LithosiÆ. The difference between LithosiÆ and ArctiÆ is rather difficult to point out; yet a difference exists, which it is perfectly impossible not to detect. The larva of LithosiÆ has sixteen feet, is very active, is moderately hairy, does not readily roll itself in ring, but occasionally assumes that attitude. The pupa is smooth, changes in a slight web, in which the hairs are intermixed. The approach of Lithosia[50] to the genus Yponomeuta, in Tinea, scarcely need be pointed out. It will be observed, that Mr. Samouelle[51] was aware of this approach, and placed the genera Lithosia and Yponomeuta following each other. The splendid Callimorpha dominula, although, to all appearance, a real Arctia, must be included in this order, and placed in contact with the following one.

Natural Order—ArctiÆ. Larva, with sixteen feet, generally very hairy, bear-like; rolls itself in a ring when touched; pupa smooth, in a slight web. Whether the whole of Mr. Curtis's genus Acronycta must be included in this order, I am not able positively to say: the genus Apatela of Mr. Stephens certainly must, and until I have obtained sufficient information to decide on Acronycta, we must bring Mr. Stephens's genus only into the order, leaving the remainder of the species undisposed of. The development of the antlia in Acronycta discovers as near an approach to Noctua, as Lithosia does to Tinea; but the bear-like, cocoon-spinning larva place these insects in close alliance with the true ArctiÆ.

Natural Order—LariÆ. Larva, with sixteen feet, and furnished with various brushes, or fascicles of hair, on different parts of the body, but mostly on the anterior dorsal segments; it rolls itself in a ring when touched. The genus Porthesia of Mr. Stephens may be considered a near approach to Eriogaster, in the following order, Bombyces, in many of its peculiarities, as the abdominal hair with which it covers its eggs, in its antennÆ, &c. Orgyia antiqua, on the other hand, is in habit, expansion of wing, slender body, and apterous female, a close approach to the GeometrÆ, near the genera Hybernia and Nyssia; in fact, were it not for the larva, that is, were the larva unknown, we should have no hesitation about placing this insect with the GeometrÆ.

Natural Order—PhalÆnÆ. Larva, with sixteen feet; it has a circle of wart-like protuberances on each segment, from each of which spring a few strong bristles; pupa, smooth, with a few bristles at the tail; changes in a cocoon, which is singularly left partly open at one end. We have but one species of this order in Britain, Saturnia carpini;[52] but among exotics there is a great variety, some remarkable for the immense expanse of their wings.[53] Probably PhalÆna Atlas of LinnÆus is the centre of the group, and, if so, the centre and type of the class Lepidoptera.

I have previously given, in a note, a list of the genera which are usually considered as Bombyces, and ought, therefore, if properly placed, to be included in the sub-class PhalÆna; five of those genera yet remain, and at present must be excluded from the sub-class: Penthophera, Heterogena, and Limacodes, because I know nothing of their history; Nudaria and Psyche, because, in the larva, pupa, and imago states, they have the habit and appearance of another class (Neuroptera). The time of their dwelling with Lepidoptera is over and gone; they have already occupied too long a position to which they were not entitled. The difficulty of assigning a situation to Limacodes I hope to see removed, as the larva is occasionally to be met with. I must also remark, that although I have proposed a situation for Endromis, I feel very doubtful as to its being the correct one. These doubts and difficulties will probably gain me much censure; but I must endeavour to shelter myself in some degree, by observing, that I am the first who has ever deviated from the original LinnÆan arrangement of Lepidoptera, the first who has ever thought of appealing to nature in support of theory, or rather has waited for nature to supply him with theory; and surely some allowance is to be made for a first attempt of any kind. I would also plead the poverty of our British Fauna in the sub-class, and my almost entire ignorance of exotic PhalÆnÆ. Even supposing myself acquainted with all our indigenous species, they will barely furnish a systematist with a clew to the truth: you may pick up a single link of a chain, yet fail to discover the length of that chain, or the situation in that chain which the link originally possessed.

Having, then, pointed out, as clearly as my limited knowledge of the subject will permit, not only the principal contents of the class Lepidoptera, but endeavoured to establish them in appointed and fixed stations, and to show their mutual approaches, at least those of the most striking kind and essential to my purpose, I must now proceed to make a few remarks on the nature of these approaches. It will be observed, that they are, almost without an exception, what Mr. MacLeay considers relations of affinity, that is, the relation is between species which, in their imago state, have a real and positive similarity to each other; so much so, that entomologists, unacquainted with the prior states, and frequently even in direct defiance of their own knowledge of those states, place them in orders, and even sub-classes to which they do not belong; to which fact all our systems and catalogues bear most ample testimony. This similarity is by no means confined to a cursory glance at the insects, but bears the test of a minute anatomical investigation, the antlia, palpi and antennÆ demonstrating the approach quite as forcibly as the form and appearance of the whole insect. Where a tribe has short biarticulate palpi, a genus departing from the type will assume elongated and triarticulate palpi, should another tribe with those characters approach it: again, should a tribe with long antlia approach a tribe whose character it is to have none, we shall be sure to find a genus without antlia at the point of approach. On the other hand, the very egg, the larva, the pupa, the mode of feeding and description of food, the mode of metamorphosis, and, in fact, every prior quality, or state, from which distinctions could be obtained, differ so decidedly, that the characters of these often bear as near an approach to those of Hymenoptera, Neuroptera, and even Coleoptera, as to those of their own kindred, into immediate contact with which these approaches will be found inevitably to bring them. What term can then be applied to designate the real value of this species of approach? Supposing the terms analogy and affinity to have had good, sound, and distinct meanings, as originally employed and explained by great naturalists, they have now been so confused, confounded, and utterly misunderstood by ignorant persons, that either of these terms is entirely out of the question:[54] in fact, a suitable term by which to designate this peculiar species of relation or approach, I neither know where to find, or how to invent; and, therefore, I shall purpose simply to call it relation of larva, relation of pupa, or relation of imago, as the case may be.

In one instance, the relation of imago is, from several combining causes, which it will be unavailing to recapitulate, uncertain enough—that of Barbicornis and Lasiocampa; but I would ask the impartial reader, is it half so far-fetched and untenable as those in common use? Can human sagacity, in sheer wantonness, invent combinations more unnatural than ——, twenty or thirty of which we could all point to in our own cabinets? For the value of the other relations (eleven others) I appeal to the judgment of the assiduous collector, the experienced observer, the real nature-loving naturalist,—to him who has spent days in the woods, and not only captured but observed these delightful beings,—to him who never invented or supported a theory,—to him who is pledged to no system, to no party,—I ask him, nothing doubting of his concurrence, whether these relations do not too plainly bear the impression of nature's seal, to allow him to doubt one instant of their reality.

In the next place a question occurs, how is the relative position of the sub-classes proved to be correct, seeing it is so totally at variance with what we have from our childhood been perfectly satisfied with?[55] It is proved correct, simply and solely by the harmony with which each flows into each,—with which neighbour meets neighbour,—comparable somewhat to that exquisite feeling which induces a man to bend to the peculiarities, and perhaps even little failings of a friend, until he makes them almost his own. It can hardly be supposed that the sub-classes naturally fell into the positions which I have assigned to them, without some little endeavour, on my part, to produce this harmony. This was far from the case. The discovery, if it be one, was the result of serious and deliberate study. Even after arriving at their present state, I have twice endeavoured to alter these positions, once in hopes of making some of the Tortrices meet the PapilionidÆ, as I had an idea that that very assiduous and ingenious naturalist, Dr. Horsfield, had mentioned the discovery of such an approach.[56] In vain, however, did I strive to discover such an approach, in either larva, pupa, or imago, while these points of resemblance were most abundant between the GeometrÆ and Papiliones; the pupa, as though in sport, being now suspended by the tail, now girted round the waist, now enveloped in a silken web; sometimes round-headed, sometimes pointed, sometimes eared; now smooth, anon angulated, black, brown, yellow, pure green, clouded, or spotted: of these, and a thousand other peculiarities, which tended to corroborate my ideas of arrangement, I refused invariably to avail myself, trusting to one guide only, which seems as steadfast as a rock: that relation of imago constitutes approach of divisions; relation of larva is the tie which holds divisions together. The second alteration I endeavoured to make, was to place the Papiliones in the centre, a situation to which their splendour and magnitude would really appear to give them a title. This idea seems every way so plausible, and so likely to be proposed by entomologists, should any such see merit enough in this system to give their attention to its minutiÆ, that I am compelled to consider it more at large.

To a sub-class selected for a centre, two qualities are indispensably requisite. They have been previously given from Mr. MacLeay, who, it will be remembered, discovered that one of each of his five groups contained types of the other four, besides a type peculiar to itself. This quality must hold good in any group thus selected for a centre; it must contain types of the six surrounding groups in the first place. Now, is this applicable to Papilio? Have we not already experienced the greatest difficulty in finding three good approaches, the smallest number which a sub-class can possess? How then can we hope, by any good fortune in discovery, to make ourselves masters of three other entirely new ones, and these to sub-classes to which it is confessedly the most unlike? PhalÆna, on the contrary, presents us with Lasiocampa, Ægeria, Cilix, Lithosia, Apatela and Orgyia, five of which genera beautifully typify the approximating sub-classes. The preference on this score then is decidedly with PhalÆna.

The second position, that it should contain a type peculiar to itself, is almost a matter of course; but my own idea is, that the very centre should not only be a type of the genus, or order, or sub-class, but of the class itself of which it is the centre. From this position, then, a further and still more important question arises,—What is the type of Lepidoptera? The parts which afford the generic characters of Lepidoptera, and, I believe, generic characters in the perfect state are the only ones of any value, are these—the mouth, palpi, antennÆ and wings; and, as no medium can constitute a type, the excess of these characters, whether superlatively or diminutively considered, must be resorted to as the most probable means we possess of discovering what this type may really be. First, then, the mouth. In Lepidoptera, we find two distinct characters in this;—first, its entire absence; secondly, its being furnished with prodigiously long antlia. The first character is that of PhalÆna, the second that of Sphinx. Next, the palpi are either entirely obsolete or exceedingly prominent, the first in PhalÆna, the second in Pyralis. Thirdly, the antennÆ are remarkably pectinated, or clavated, or setaceous: the first character is that of PhalÆna, the second that of Papilio, the third that of Noctua. Fourthly, the wings are enormously expansive in proportion to the body, or remarkably small,—the first is the character of PhalÆna, the second that of Sphinx. It need scarcely be added, that all these characters are to be met with in every intermediate degree of intensity. Now, it appears, that PhalÆna possesses an extreme of each of the four principal characters, Sphinx of two, Noctua of one, and Papilio of one; therefore PhalÆna is the typical genus, PhalÆnÆ the typical order, and PhalÆna the typical sub-class of Lepidoptera: and a necessary conclusion from this fact is, the type of Lepidoptera is an insect without antlia or palpi, with very pectinated antennÆ and enormously expansive wings, and we may add nocturnal flight: so that such peculiar characters as the thick full body and prodigiously long antlia of Sphinx, the clavate antennÆ, erect wings, and diurnal flight of Papilio, argue a departure from, and not an approach to, the type.

By a reference to the Diagrams exhibiting the classes of Insecta, and the sub-classes of Lepidoptera, it will at once be observed, that the central group in each case contains types of the surrounding groups. Now after a central group has thrown off a set of six forms, each representing, in general appearance, some group equally extensive with such central group, the faculty or power of throwing off such forms becomes, in a good degree, extinct, or, at any rate, very much debilitated. This can be no unforeseen, but a perfectly natural, and absolutely necessary consequence; for taking either of the two classes which are at present sought after, Lepidoptera and Coleoptera, we must observe, that did either of them possess as varied forms and characters as are to be found in Neuroptera, the essential and distinguishing character of that class, viz. variety, and the harmonious arrangement of the whole sub-kingdom, would both be entirely lost; and it would remain for human ingenuity to locate either of the classes centrally or externally, as caprice, or the love of differing from others, might dictate. I wish it to be observed, that Neuroptera, in the genera Psyche, CloËon, Termes, Psocus and Mantispa, does not merely assume the form of the genera, Tinea, Chironomus, Formica, Aphis and Mantis, but actually possesses the characters and appearance of the classes Lepidoptera, Diptera, Hymenoptera, Hemiptera and Orthoptera. The obviously homogeneous character of Lepidoptera and Coleoptera, although, probably, containing in every sub-class more species than the whole of Neuroptera, clearly disproves the existence of variety amongst their contents, equal to that amongst the contents of Neuroptera. Yet the power, although weakened, is by no means extinct; for, amongst the central group, PhalÆna, we find the sub-classes, Papilio, Sphinx, Pyralis, Tinea, Noctua and Geometra, most faithfully pourtrayed in Lasiocampa, Ægeria, Cilix, Lithosia, Apatela and Orgyia, and not merely the individual genera which may happen to approach. As far as I can discover, after this second series of types the faculty becomes much weaker, and, after a third, ceases entirely. A decided difference existing between the first and second series of types, must on no account be lost sight of, because it so decidedly proclaims the superiority of the first: in the first instance, the whole character of the central type, Libellula, is completely lost in each of the varying types; whereas, in the second instance, the characters of PhalÆna are preserved most decidedly to the remotest ramifications of the class, subject, however, to the variations already pointed out.

The natural order, Cossi, of which the larva and pupa have been already described, contains but ten genera, even including those whose claim to a place in the order is somewhat doubtful; and these ten are readily referable to six families. The genus, Stygia, of New Holland, seems from Latreille's description, decidedly to belong to this order. Speaking of Stygia Australis, he says, "M. Villiers la considÈre comme intermediare entre les Sesies et les ZygÈnes; mais elle n'a point de trompe; ses palpes sont ceux de Cossus; ses antennes sont courtes, et nullement en fuse, et plus analogues a celles de certains Bombyx qu'a celles des Sesies et des ZygÈnes."[57] Now the fact, as M. Latreille supposes, of having no antlia, argues most forcibly the impossibility of uniting this genus either with Sphinx or ZygÆna; for the sub-class Sphinx not only possesses the most elongate and conspicuous antlia of any sub-class, but retains this character to its very circumference, and imparts it to approaching groups, whose types will be found entirely aglossate: its similarity therefore in shape to the SesiÆ, which tribe is generally understood to include the ÆgeriÆ, is merely that relation of imago which I have before so repeatedly pointed out. The situation, which without this genus must have been vacant, thus filled, gives us a most perfect chain of families throughout the order, except at the point of connexion with PhalÆna, a point of no consequence, because it too much favours old theories to be contested.

It is rather remarkable, that in this order no instance should occur of more than three genera belonging to any one family, a number which I should hardly suppose complete, because a difficulty must always occur in placing, as in discovering the typical genus or species, where the number is confined to three.

The introduction of a new generic name, after what has been said on that subject, may appear rather an inconsistency, but I found it indispensable, as the species in question would not bend to either of the established genera, Trochilium or Ægeria; it will, moreover, afford those whose labours in this way I have somewhat deprecated, a fair opportunity for retaliation. The families and their relative situations, as far as my immature and hastily-formed judgment will allow me to decide, I have shown in the annexed diagram: but it is now time for me to describe the species whose situation I am endeavouring to point out.

Sub-kingdom, Insecta.

Characters from the imago.

The body is divided into three parts, head, thorax, and abdomen; the head has two fixed compound eyes, and two moveable antennÆ. Insects have six jointed legs in pairs; they breathe by lateral spiracles.

Class, Lepidoptera.

Characters from larva, pupa, and imago.

Larva polypod, bears no resemblance to the imago; pupa quiescent, bears no resemblance to the imago. Imago has four scaly wings, and the mouth aglossate or antliate.

Sub-class, PhalÆnÆ (central).

Characters from larva, pupa, and imago.

All varying (the universal character of such central groups).

Natural order, Cossi.

Characters from the larva and pupa.

Larva depressed, kaned; has sixteen feet, lives through one or more winters, never rolls itself in a ring when touched, feeds on the solid interior woody parts of vegetables; pupa changes in a tough cocoon, in which are interwoven particles of the larva's food; it has a double row of small raised denticulations on each segment of the abdomen, which give it partially the power of locomotion.

Family ÆgeriidÆ, Stephens.

Characters from the Imago.

Palpi triarticulate, incrassated at the base, acuminate at the apex, prominent, enclosing the antlia; antennÆ, sub-cylindric, gradually incrassated from the base nearly to the apex, the apex itself acuminate and terminated with a fascicle of hairs; ocelli, two. Flight diurnal in the hottest sunshine.

Genus Memythrus.—Sphinx, Linn.; Sesia, Laspeyres; Ægeria, Fab.

Characters from the imago.

Palpi very prominent, and densely clothed with scales at the base, in appearance angulated; antlia fine, not so long as the antennÆ; antennÆ the length of the thorax, in the male much pectinated, in the female simple; superior wings clothed with scales, inferior hyaline.

Sp. 1. Memythrus Vespiformis.—Sphinx Vespiformis, Linn. Syst. Nat. II. p. 804, n. 31. Ægeria Asiliformis of Fabricius, and other authors.

Characters from the imago.

Palpi black, yellow at the apex; antennÆ black, beneath testaceous; fulvous at the base; head black, excepting a white mark before each eye; a yellow ring round the neck; thorax black, with a yellow spot at the base of each superior wing; abdomen black, slightly barbate, with three equidistant yellow belts; superior wings deep fuscous, inferior hyaline; femora and anterior tibiÆ black, posterior tibiÆ and all the tarsi yellow.

Inhabits England, but is very rare.

Sp. 2. Memythrus crabroniformis.—Sesia crabroniformis, Lasp.

Inhabits Italy.

Sp. 3. Memythrus crassipes.—Sphinx crassipes, Drury.

Inhabits Africa.

Sp. 4. Memythrus tibialis.—Ægeria tibialis, Fab.

Sp. 5. Memythrus ——?—Unnamed in the LinnÆan cabinet.

Several other species probably exist, with which I have not happened to meet.

The principal distinctions between Memythrus and Ægeria are, that the antennÆ in the former are not longer than the thorax; in the latter they are much longer; in the males of the former genus they are decidedly pectinated, in those of the latter but obscurely ciliated; in the former the anterior wings are always opaque, in the latter always hyaline.

Natural Divisions to which the Sphinx Vespiformis
of LinnÆus is referable.

FINIS.

R. CLAY, PRINTER, BREAD-STREET-HILL.

DIRECTIONS FOR PLACING THE DIAGRAMS.

1. The Classes of Insecta to face page 21.

2. The Sub-classes of Lepidoptera to face page 31.

3. The Natural Order of Cossi to face page 52.