15. Siphonaptera (= Aphaniptera, a division of Diptera with us).
16. Coleoptera (= Coleoptera).
17. Hymenoptera (as with us).
The chief characters on which Brauer bases his system are: (1) The existence or absence of wings. (2) The condition of the mouth, and whether it undergoes radical changes in the ontogeny, arriving thus at the categories Menognatha, Metagnatha, and Menorhyncha, as we have mentioned on p. [161]. (3) The metamorphosis; the grouping adopted being Ametabola, Hemimetabola, Metabola. (4) The number of the Malpighian tubules; Oligonephria, Polynephria. (5) The nature of the wings, the relative proportions of the thoracic segments, and some other characters.
Brauer's treatise is accompanied by a valuable and in many respects very sagacious consideration of the generalised characters of the Insecta; as a classification based partly on generalisations and partly on structures, it is, so far as the present condition of our knowledge goes, a good one. But it is of little value as a practical guide, and as a basis for theoretical speculation it cannot be treated as of importance, because the generalisations it makes use of are premature, owing to the small proportion of the forms that have been examined. And even now the groups adopted are known to be subject to many exceptions.
Thus it begins by a division of Insecta into winged and wingless; but the winged division is made to comprehend an enormous number of wingless Insects, whole subdivisions of Orders such as the Mallophaga being placed in the winged series, although all are without wings. This first division is indeed entirely theoretical; and if a classification on generalisations were adopted, it would be more natural to begin with the old division into Homomorpha and Heteromorpha, and treat the Order Aptera as the first division of the Homomorpha, while the Heteromorpha would commence with the Ephemeridae and Odonata, in which, though the individual in the early part of the ontogeny is very different from the perfect Insect, there is no marked division of the later larval and the pupal stages. Brauer's system is also defective inasmuch as it takes no account of the embryological or oogenetic processes, though these are of equal importance with the later phases of the Ontogeny. Even as regards the division into Orders, it is far from being free from reproach; for instance, the separation of the Dermaptera from the Orthoptera, while Rhynchota remains intact, although including a more extensive series of heterogeneous forms; the division of the Neuroptera into widely separated groups, each of which is treated as equivalent to the great Orders, such as Coleoptera (in which Strepsiptera are included), Hymenoptera, and Diptera, is not reasonable. The association of Mallophaga and Termitidae, while Dermaptera are separated from Orthoptera, is also undeniably arbitrary, and other similar disparities are to be seen on scrutinising the details of the system.
On comparing the three arrangements we have outlined, it will be seen that the chief discrepancies they present come under two heads: (1) The treatment of the Neuroptera, opinions differing as to whether these Insects shall be grouped as a single Order, or shall be divided into numerous Orders; and as to what, if this latter course be adopted, the divisions shall be. (2) The treatment of the parasitic groups Mallophaga, Aphaniptera, etc. It must be admitted that whichever of the three systems we have sketched be adopted, the result is, as regards both these points, open to criticism. The Order Neuroptera, if we take it in the broad sense, differs from the other Orders in the greater variety of metamorphosis exhibited by its members; while if, on the contrary, it be dismembered, we get a number of groups of very unequal extent and not distinguished from one another by the same decisive and important characters as are the other Orders of which they are considered equivalent. The discrepancy exists in nature, and can scarcely be evaded by any system. A similar observation may be made as to the parasitic groups, viz. Mallophaga, Anoplura, Aphaniptera, and Strepsiptera. If these be treated as separate Orders the result is not satisfactory; while, if they be associated with the larger groups to which they are respectively nearest allied, it is almost equally unsatisfactory.
We may mention that Packard and Brauer have in their treatises discussed the question of super-orders, and have gone so far as to propose names for them. These two authorities do not however agree in their conclusions; and as the names proposed are of little practical value, and are but rarely met with, we need not explain them or discuss the comparative merits of the two systems.
The divisions of inferior value to the Order are, after repeated scrutiny by many naturalists, becoming of a more satisfactory character, and notwithstanding various anomalies, may be, many of them, considered fairly natural.[[111]] Unfortunately entomologists have not been able to agree on a system of terminology, so that for these subdivisions terms such as sub-order, series, legion, section, tribe, etc., are used by different authorities in ways so various as to cause much confusion. In the following pages the terms sub-order and series will be used in a somewhat vague manner, the term sub-order being preferred where the group appears to be an important one and of a fairly natural character, while the word series will be adopted when the groups are connected in a conventional manner. The designation "family" will be used for groups of subordinate importance; and as regards this term we may remark that systematic entomologists are making genuine efforts to define the "families" in an accurate and comprehensible manner. The endeavour to make these systematic families dependent throughout the Class Insecta on characters of similar morphological value has, however, scarcely been entered on, and it is perhaps not desirable, seeing how very small a portion of the Insects of the world have been critically examined, that much effort should be yet expended on an attempt of the kind. It must be admitted that the species of Insects should be obtained before they can be satisfactorily classified, and it is estimated[[112]] that at least nine-tenths of the Insects of the world are still unknown to entomologists.
Geological Record.—Although Insects have a very long pedigree, it is as yet a very imperfect one. The remains of creatures that can be referred to the Class Insecta have been found, it is said, in Silurian strata; only one or two of these very early forms are at present known, and the information about them is by no means satisfactory; if Insects at all—as to which some doubt exists—they apparently belong to very different forms, though, like all the earliest fossil Insects, they are winged. In the strata of the Carboniferous epoch numerous Insects have been detected, in both Europe and North America. These earlier Insects are by Scudder called Palaeodictyoptera, and separated from the Insects around us on the ground that he considers there existed amongst these palaeozoic Insects no ordinal distinctions such as obtain in the existing forms, but that the primeval creatures formed a single group of generalised Hexapods. Brauer does not accept this view, considering that the earlier Insects can be referred to families existing at the present time and forming parts of the Orthoptera, Neuroptera, and Hemiptera. The discrepancy between these two authorities depends to a great extent on the different classifications of existing Insects that they start from; Scudder treating the wings as of primary importance, while Brauer assigns to them only a subordinate value. From the point of view taken in the present work Scudder's view appears to be in the main correct, though his expression as to the primary fossil Insects forming a single homogeneous group is erroneous. The Neuroptera, still in existence, certainly form a heterogeneous group, and it is clear that the Palaeozoic fossils represent a more diverse assemblage than the present Neuroptera do.[[113]]