The Zoologist/4th series, vol 2 (1898)/Issue 685/Moths and their Classification

by Edward Meyrick

 

THE ZOOLOGIST

 

---

 

No. 685.—July, 1898.

 

---

 

MOTHS AND THEIR CLASSIFICATION.

By E. Meyrick, B.A., F.Z.S., F.E.S.

 

Probably no group of animals has suffered so much at the hands of unscientific systematists as the Lepidoptera, and the reason of this is not far to seek. It is directly attributable to the great range of colour and variety of marking which are so striking a characteristic of the group, instantly attracting the attention of the superficial student, and causing him to neglect structural details as unnecessary, or even to intentionally reject them as contradicting the testimony of colour, and therefore untrustworthy. Allied species, he argues, are usually similar in colouring; therefore similar species are allied; which, however, does not follow.

It is true that this theory is now seldom put forward in a simple and unadulterated form. Dr. von Gumppenberg, indeed, has lately issued an elaborate monograph of Geometers, in which all structural investigations are cast to the winds, and the genera defined solely by colour and marking, but it does not seem to have been warmly received. But in a modified form the theory is still so extremely popular that it exercises considerable influence over the views of almost all lepidopterists. This modification consists in classifying species by superficial appearance in the first instance, and then trying to find structural excuses for it afterwards. No better example of the consistent working of this method can be found than in the works of Guenée. In the six volumes which he contributed to the 'Species Général des Lepidoptères,' I cannot recall a single instance in which he has suffered structural characters to override superficial; and yet each genus and family is prefaced by a careful statement of structural points which may readily, nay, is almost certain to be mistaken for a diagnosis, until one tries to work with it. For example, in characterizing his Geometrid family Acidalidæ, he correctly states "indépendante bien marquée aux quatre ailes; costale des inférieures isolée de la sous-costale ou simplement rapprochée," of which the equivalent in my terminology would be "vein 5 well-developed, 8 in the hind wings not connected with cell or near base only," which are really essential characters of his genus Acidalia and its allies, and no exceptions are mentioned. Moreover, if the characters given for the Acidalidæ and the Larentidæ, for instance, are compared throughout, it will be seen that there is no absolute distinction between them except in the latter of these two characters, the corresponding point for the Larentidæ being "costale des secondes ailes presque toujours bifide," that is, "8 in the hind wings anastomosing with cell to beyond middle." Now if we proceed to examine the genera attributed to the Acidalidæ, we find the first, Synegia, really has 5 in the hind wings obsolete, and should therefore be placed (not only correctly, but on Guenée's own definition) in his family Boarmidæ; the second, Drapetodes, is not a Geometer at all, but belongs to the Drepanidæ, having 5 approximated to 4 in both wings; the fifth, Pomasia, and the seventh to tenth, Cambogia, Asthena, Eupisteria, Venusia, all have 8 in the hind wings anastomosing strongly with the cell, and belong in truth to the Larentidæ. The family definition has in fact been disregarded altogether. On the other hand, the characters given for the Apamidæ and Hadenidæ, for instance, have little reference to structure, and are practically identical; hence it is not surprising that species rightly referable to the same genus are placed by Guenée, some in one and some in the other of these two families, which are separated from one another by four other considerable families of Noctuæ. Now these are not isolated instances, but fairly typical examples of the whole work; and yet this classification, having been adopted by both Doubleday and Stainton, has held the field in Britain for forty years, and has become so familiar that to most British entomologists all deviations from it seem artificial and unnatural.

In a sound system of classification the various groups must be capable of actual diagnostic definition in words, or the system is unworkable; they must be defined by structure, or it is misleading; and the system must be based on the study of the phylogeny (or scheme of ancestral descent), or it is artificial. Of late years some progress has been made in classifying the Lepidoptera in accordance with these principles, and as the general lines of such a classification are now fairly well established, it may be of interest to give a sketch of the results already reached, as illustrating principles of universal application.

Four-winged insects usually have some contrivance for holding together the two wings on one side, and ensuring their common action. Thus in the Hymenoptera (bees, &c.) there is a row of hooks and eyes; in the Trichoptera (caddis-flies) there is a membranous process (jugum) from the dorsum of the fore wings near the base, which projects beneath the edge of the hind wings, whilst the following part of the dorsum extends above it; again in the Lepidoptera there is normally a stout bristle or group of bristles (frenulum) rising from the edge of the hind wings near the base, and passing under a catch on the under side of the fore wings. Some Lepidoptera do not possess this frenulum, and in such cases the basal angle of the hind wings is made more prominent so as to project beneath the base of the fore wings and prevent dislocation. But some five years ago Prof. Comstock made the discovery that in two families, the Hepialidæ and the Micropterygidæ, instead of the usual lepidopterous structure, there is a jugum, quite as in the Trichoptera. There is no difficulty in seeing this structure, at any rate in the Hepialidæ, some of which are very large insects of five or six inches expanse of wing, and it remained undiscovered so long simply because no one had thought of looking for it; a striking instance of the ease with which characters of the highest importance can be over* looked by competent observers, unless their attention is specially directed towards them. Now it was known previously that these same two families agree together in possessing several additional veins in the hind wings, which are not found in any other Lepidoptera; these veins could not have been evolved from nonexistent structures, and are therefore ancestral; hence these two families constitute not only a separate group, but the oldest group of the Lepidoptera, and may be termed Micropterygina. Their close affinity to the Trichoptera is shown by the possession of the jugum, and by the fact that the complex neuration of the Micropterygidæ is practically quite identical with that of certain Trichoptera (as Rhyacophila). As the Trichoptera usually possess a much larger number of veins, especially in the hind wings, they must be the older group, and the Lepidoptera must have originated from them. The most ancient Micropterygidæ known are found in New Zealand, though the majority are European; but these little insects are so readily overlooked by collectors that their distribution is insufficiently ascertained. The Hepialidæ, standing considerably isolated from these, are presumably the highest development of a once extensive group, intermediate forms being apparently all extinct; they are now very widely distributed, probably as a result of their very powerful flight, but would seem to be Indo-Malayan in origin.

The next point is to ascertain the connection of the typical Lepidoptera with the Micropterygina; this cannot be at more than one point, since it is highly improbable that the frenulum and fixed type of lepidopterous neuration could have been evolved twice. This transition is undoubtedly indicated by the New Zealand genus Mnesarchæa, which in the character of the palpi and the neuration of fore wings approximates closely to some forms of Plutellidæ and Tineidæ, whilst remaining by strict definition a true Micropterygid. The origin of the Tineina is thus established; and in the two above-mentioned earliest families (divergent branches from the same stem) the excessively long antennae of the Adela group, and the occurrence of long six-jointed maxillary palpi in many genera of the Tineidæ, as also the porrected habit of the antennae in some genera of Plutellidæ, are distinct reminiscences of their Trichopterous origin, and may be quoted as examples of reversion. Further, there can be no question that the other families of the Tineina constitute a line of development originating in the Plutellidæ, all these being typically smooth-headed; whilst the Tortricina form a parallel branch taking its rise from the Tineidæ, all these being typically rough-headed. Whether the Tortricina are maintained as a separate group, as has usually been done, or, in accordance with the views of Lord Walsingham and Mr. Durrant, merged in the Tineina, is only a question of name and convenience, and therefore of no marked importance. It should be noted that in all the families of these groups there are normally three free veins (1a, 1b, 1c) between the cell and dorsum in the hind wings, though in cases where the area of wing is very small, as in some of the minute Tineina, these and most of the other veins are liable to disappear.

Here may be considered two particular cases, those of the Aegeriadæ and Trypanidæ. The Aegeriadæ, popularly known as "clear-wings" from the hyaline spaces on their wings, used formerly to be oddly placed near the Sphingidæ, but are in all essential characters undoubted Tineina, parallel in development with the Gelechiadæ and Oecophoridæ. The Trypanidæ (Cossidæ of some) must be regarded as unspecialized Tortricina, marking the transition from the Tineidæ; the comparatively gigantic size of the single British species is at first sight somewhat startling, but this is not always maintained in exotic forms, and there is really no other distinction at all. The wood-feeding habit of the larva is very characteristic of that group of the Tineidæ from which it is derived.

Having now established in the Tineina a base of origin, with which the connection of the general body of Lepidoptera has to be traced, we may consider what characters can be held to indicate nearness to or remoteness from this base. The best indication for this purpose will be furnished by the presence or loss of some ancestral character which when once lost is incapable of redevelopment. Prof. Comstock has employed the frenulum for this, but the choice appears to be unfortunate, for three reasons, viz. (1) the proportion separable as having lost the frenulum is comparatively small; (2) the frenulum may have been lost in different groups quite independently, and has in fact obviously been so lost in several families; (3) as the frenulum is apparently only the modification of hairs which are always present, there seems no reason why it might not exceptionally be redeveloped by reversion. A better character is furnished by the presence or absence of vein 1c in the hind wings, which is found to be usually constant not only in families, but in main groups, and a vein once lost can never be regained; but of course the loss may have, and apparently has, taken place independently in more than one line of descent. The whole of the remaining Lepidoptera may then be classed in two groups (Psychina and Pyralidina) which normally retain this vein, and four others which have entirely lost it.

The Psychina are a group of families of unspecialized type, which in fact approach the Tineina so closely as not to be separable as a whole by any single character, though each family considered by itself is so separable; at the same time the four families composing it are nearly related together in their lowest forms, and may therefore be regarded as parallel developments. Even the markings of the wings show this want of specialization, as they are, when present, irregular and without definite type of arrangement, and all the families show a marked tendency (in the Psychidæ becoming a fixed character) to produce thinly-scaled or semi-hyaline unicolorous forms. It is probable that this indicates approximate relationship to the Fumea group of the Tineidæ, to which the Psychidæ also display their affinity by their apterous females and the case-bearing habit of the larva; on this ground they have even been included together in the same family, but the true Psychidæ are always distinguished by the anastomosing subdorsal veins of the fore wings. The Zygænidæ include many large butterfly-like forms, brilliantly coloured with metallic blue, crimson, and white; the short, stout, often tuberculated and rather hairy larvae are of an early type, and though apparently very different in form and habits to the Psychidæ, both are probably the modified descendants of internal wood-feeders. The Heterogeneidæ (Limacodidæ of some) are remarkable for their larvæ, which are an exaggeration of the Zygænid type, the legs being often very short and retractile, so that the larva appears to be appressed to the leaf like a slug, whilst the dorsal tubercles are often developed into clusters of stinging spines; and for the small hard oval cocoons, which open by a lid. The tibial spurs of the imago are long, as in the Tineina generally, whilst in the allied families they are very short or absent, but there is here probably some connection with bulk or weight. Finally, the Zeuzeridæ, whilst structurally related to the Psychidæ, show by the wood-feeding habit of their larvæ and other characters collateral affinity with the Trypanidæ, traceable to their common descent. The rather confusing cross-relationships between these several families are characteristic of little-specialized forms; it is as if one had to disentangle a network of small divaricating twigs close to the stem, whilst the course of the larger branches is comparatively easy to trace.

The Pyralidina are commonly distinguishable from the other groups already mentioned by the structure of vein 8 of the hind wings, which is brought down so as to closely approach or anastomose with vein 7 beyond the cell. Another character almost constant throughout the group is the stalking of veins 8 and 9 of the fore wings; it is quite constant in most of the families, but in the mainly tropical family Thyrididæ these veins are more usually separate; we may therefore with considerable probability regard the Thyrididæ as ancestral. Finding further that they nearly approach the Heterogeneidæ, both structurally and superficially, whilst the other families are of a peculiar type which is remote from anything else, we shall be justified in looking to the Heterogeneidæ as the origin of the group. The mutual relations of the nine families composing this extensive division need not be discussed here in the main; but the case of the Pterophoridæ may be mentioned. These curious insects, the well-known "plume-moths," usually have the wings very narrow, and split into two or three feather-like lobes; hence the neuration tends to be much degraded for want of room, but in the earliest forms (and recognizing the transition afforded by the small Australian family Tineodidæ) it approaches the Pyralid type; with which also the unusually long and slender legs, the structure of the head, and the larval appearance and habits are also in accordance. The Orneodidæ (in which each wing is split into six plumes) can be traced to the same source.

Coming now to the groups which have vein 1 c of the hind wings constantly absent, it will be convenient to study first the Papilionina, generally termed "butterflies." Notwithstanding the amount of attention bestowed on this attractive group, little has been written as to its origin. It is characterized by the clubbed antennae, and absence of the frenulum, both these features being found in other cases but not in combination. As it falls into two sections, of which one (Hesperiadæ) has all the veins of the fore wings separate, and the median spurs of the posterior tibiae usually present, whilst the other has always two or more veins stalked, and the said spurs always absent, there can be no doubt that the Hesperiadæ are the most ancestral family. Their very simple neuration closely resembles that of the Thyrididæ, but is not found elsewhere in the higher groups, and there is no discordance in other structural characters; moreover, the tendency to show pale semi-hyaline spots in the fore wings, and the development of specific colour-characters on the lower surface of the hind wings, are marked points of superficial resemblance. Similarly knobbed antennae occurring in the higher Caradrinina (Agaristidæ) have been thought to indicate relationship to the Papilionina; but there the frenulum is always strong and persistent, and the required simple type of neuration is never found. Hence we must suppose that the Thyrididæ are the true starting-point of the group.

The Lasiocampina comprise five families of no great extent altogether. In these the frenulum either is or tends to be absent, and vein 8 of the hind wings is frequently approximated to 7 beyond the cell, the group being always separable from the Caradrinina by one or other of these characters. In the Pterothysanidæ, Lasiocampidæ, and Endromididæ the frenulum is constantly absent; in the Drepanidæ and Callidulidæ it is sometimes present, though tending towards obsolescence, and these two families, which are apparently collateral developments, must be the more primitive. They approach the Thyrididæ, and the Callidulidæ also appear to show near collateral relationship to the Papilionina, for which, except that the antennae are not knobbed, they might sometimes be mistaken even by an expert.

The Notodontina include all those families of the higher Lepidoptera in which vein 5 of the fore wings, instead of being approximated at its origin to 4, is parallel with it, or even sometimes more approximated to 6, thus appearing to form an independent vein from the cell, instead of a branch of the vein which forms the lower margin of the cell. There is no reason to suspect that this modification has arisen more than once, the whole of these families agreeing well together in all other respects. The Eupterotidæ, mostly large insects which, both in the imago and larva states, have considerable relationship to the Lasiocampidæ, are probably the least specialized; as they possess a frenulum, they cannot justly be derived from the Lasiocampidæ themselves, but have probably a common ancestor not far removed. From this original family are derived four branches, viz. (1) the Bombycidæ (this name has often been wrongly applied to the Lasiocampidæ, but here denotes Bombyx mori, the "silkworm" moth, and its allies) and Saturniadæ, which have lost the frenulum entirely; (2) the Notodontidæ, Polyplocidæ (Cymatophoridæ of some), and Sphingidæ, stout-bodied forms, whose larvæ are commonly furnished with various prominences; (3) the Uraniadæ and Epiplemidæ, in which veins 6 and 7 of the fore wings are normally stalked; (4) the several families formerly called Geometrina, whose larvæ have usually lost two or three pairs of prolegs.

Lastly, the Caradrinina contain seven families, of which the Ocneriadæ (Liparidæ of some) is doubtless the most ancestral, making in fact a close approximation in many points to the Psychidcs, and showing a tendency to exhibit similar apterous females. In this, and the allied family Hypsidæ, vein 8 of the hind wings is connected by a bar with the middle of the upper margin of the cell. In the Agaristidæ, Caradrinidæ, and Plusiadæ (these two latter forming the old group Noctuæ, whose name is untenable, belonging by right of priority to an owl), this is modified so that 8 anastomoses with the cell-margin very shortly near base, the Agaristidæ being characterized by the apically swollen or sometimes clubbed antennae, the Caradrinidæ by the obsolescence of vein 8 in the hind wings, which in the Plusiadæ is well-developed. In the Arctiadæ a further modification takes place, 8 anastomosing with the cell-margin for a considerable distance from base. In the Syntomididæ is reached the extreme of change in this direction, 8 becoming wholly absent by coincidence with the cell-margin and 7.

In this scheme the Caradrinina, Notodontina, Papilionina, and Tortricina are all terminal developments, i.e. growths which lead to nothing beyond themselves, and in translating this scheme into a linear form it would be possible to take any one of these as top, and the other branches in any convenient succession. But, considered as a whole, the Caradrinina, from the difficulty of sharply defining the families (which implies comparatively little extinction), and their dominant character, as shown by their wide distribution and the prodigious number of similar species and individuals, must be thought to be the most recent. The following order correctly expresses the phylogeny as indicated above, whilst paying some regard to collateral relationship also, viz. *Caradrinina, *Notodontina, Lasiocampina, *Papilionina, Pyralidina, Psychina, *Tortricina, Tineina, Micropterygina; where the asterisk marks terminal developments.

It is not uncommon to see futile discussions as to which of two groups, reached by different lines of descent, is the higher, i.e. the more highly organized. The question is not only always unanswerable, but the answer would be quite valueless if found; all that can be done is to find the more recent.

In conclusion, a word as to the practical value of structural characters in classification. Characters of colour and general form are bad only because they are particularly liable to be modified by changes of environment. Now some structural characters are quite as liable, and are therefore equally bad. For example, in birds the shape of the beak is obviously likely to be modified in accordance with a change of food, and is therefore (as between allied forms) probably little better than a colour character. Yet the teeth of mammals, used for the same purpose, afford an excellent character, because the element of number comes in, which gives definition and admits of greater variation. It may be doubted whether any group of animals exhibits a better character than the neuration of insects, which displays sufficient complexity and variation in the number and interconnection of the different veins, whilst at the same time it is practically unaffected by external forces, except occasionally the easily calculable influence of a change in form of wing; moreover, the modifications effected are often irrevocable, and therefore less puzzling to follow.