From a further consideration of the foregoing principles it will be seen that all existing species are held to be descended by true generation from pre-existing species, and that, consequently, all the relationships we observe between species are explained by community of origin. The most natural system of classification is, therefore, that which best reveals the scheme of descent, or, as it is termed, the phylogeny, of the group of organisms classified. To construct a perfect system of classification on these principles a knowledge of not only all the existing species of Lepidoptera would be essential, but also of all the extinct species, and it is needless to say that such knowledge is quite unattainable. Nevertheless large numbers of species are now known from many parts of the world, and a very extensive collection has recently been employed by Mr. Meyrick in framing a classification of the Lepidoptera, which is, to the best of my belief, the first constructed on strictly Darwinian principles. Although adopting Mr. Meyrick's system in the present work I do not agree unreservedly with all his conclusions; but I have not attempted to alter his system in accordance with my own views, as I conceive that the conclusions of a naturalist, who has only had the opportunity of studying a restricted fauna, would necessarily be liable to considerable error.

The general principles on which Mr. Meyrick has founded his system are practically those laid down by Darwin in his 'Origin of Species,' and may be thus summarised:—

A. Resemblances between all organisms are explained by community of origin, the amount of difference representing the amount of modification and expressible in the classification as varieties, species, genera, families, groups, orders, &c. The amount of difference does not necessarily bear any direct relation to time, many forms remaining almost stationary whilst others are undergoing development.

B. By a consideration of the following laws the age of a division can be approximately arrived at; that is to say, its position in the great genealogical tree of the Lepidoptera can be, to some extent, determined:—

"(1) No new organ can be produced except as a modification of some previously existing structure.

"(2) A lost organ cannot be regained.

"(3) A rudimentary organ is rarely redeveloped."—(Meyrick.)

C. The greatest care is necessary to avoid being misled by adaptive characters, i.e., characters which are very important to the welfare of the species, and hence much modified through the agency of natural selection. A familiar instance of superficial resemblance, due to the presence of similar adaptive characters, may be observed in fishes and whales, where two groups of animals with but little real relationship have, through living under similar conditions, become extremely like each other in external appearance. Other examples might be given amongst exotic Lepidoptera. Thus, many noxious species are closely mimicked by harmless forms which are often far removed from them in real affinity. These cases of adaptive resemblances abound amongst all organisms, and have often deceived experienced naturalists. It is in consequence of the illusive nature of these external resemblances amongst different members of the Lepidoptera, that the structure of the neuration of the wings is now considered of such great importance as a character for purposes of classification. The numerous modifications in the position of the veins and their presence or absence in certain groups can, so far as we are able to see, have had very little effect on the well-being of those insects possessing such modifications. Hence it may fairly be assumed, that these structures have been free from the influence of natural selection for a very lengthened period. It is thus contended that the neuration of a Lepidopterous insect probably reveals more plainly than any other character its true relationship with other species.

The descent of all the Lepidoptera from some ancient member of the Trichoptera (or caddis-flies) is thus proved, according to Mr. Meyrick:—

"From a consideration of the laws enunciated above, there can be no doubt that the Micropterygina are the ancestral group of the Lepidoptera, from which all others have descended; this is sufficiently proved by the existence of the four or more additional veins in the hind-wings of that group, for these veins, if not originally present, could not have been afterwards produced. Of the two families of that group, the MicropterygidÆ, which possess an additional vein (or veins) in the fore-wings, and fully developed six-jointed maxillary palpi, must be more primitive than the HepialidÆ. Now if the neuration of the whole of the Lepidoptera is compared with that of all other insects, it will be found that in no instance is there any close resemblance, except in the case of the MicropterygidÆ; but the neuration of these so closely approaches that of certain Trichoptera (caddis-flies) as to be practically identical. The conclusion is clear, that the Lepidoptera are descended from the Trichoptera, and that the MicropterygidÆ are the true connecting link. If the other marked structural characters of the MicropterygidÆ are taken into consideration, viz., the possession of the jugum, the large development of the maxillary palpi as compared with the labial, and the sometimes functionally active mandibles, they will be all found commonly in the Trichoptera, affording additional confirmation. It may be added that in one New Zealand species of MicropterygidÆ (PalÆomicra chalcophanes) vein 1b is basally trifurcate, a character frequent in the Trichoptera, but not yet discovered in any other Lepidopteron. In most Trichoptera the veins of the hindwings are much more numerous than those of the fore-wings, in the Micropterygina they are usually equal in number, in other Lepidoptera they are less numerous; in the course of descent there has therefore been a greater progressive diminution in the number of veins of the hind-wings as compared with those of the fore-wings, though these also have diminished.

"It is unnecessary to trace back the descent of the Lepidoptera further; but it may be worth while to point out that we may assume as the primitive type of Trichopterous neuration, a system of numerous longitudinal veins gradually diverging from the base, mostly furcate terminally, and connected by a series of irregularly placed cross-bars near base, and another series beyond middle."

The following is Mr. Meyrick's method of arrangement, which has been adopted in this book:—

"The natural order of arrangement, which is that of a much-branched tree, cannot be adequately expressed by a simple linear succession, such as is alone practicable in a book. It is, however, possible to devise a linear succession which shall be consistent with the natural genealogical order, if some additional explanation can be given. The method here adopted is as follows:—

"Suppose the accompanying diagram represents a portion of the genealogical tree; then the order will begin at M and descend to K, recommence at L and descend to K, and thence to G, recommence at H and descend to G, and thence to B, recommence at F and descend to D, recommence at E and descend to D and thence to B, recommence at C and descend to B and thence to A, and so on. Thus the order begins with the most recently developed forms and descends gradually to the earliest or most ancestral, which are the last in the book. To understand the order in practice, it may be assumed that each genus is descended from that which immediately follows it in the book, unless its actual descent is expressly stated otherwise; such statement will, of course, require to be made before every recommencement of a fresh branch. This system has been adhered to throughout, and after a little use will not be found unintelligible. If adopted in the arrangement of a collection in the cabinet, it would be a good plan to indicate the recommencement of a fresh branch by a special mark, such as a red bar drawn above the first (or highest) species."

Phylogeny of Lepidoptera. (After Meyrick.)