of beetles of different families, stridulating areas occur on various segments of the abdomen, and are scraped by the elytra. It is remarkable that these organs are found in similar positions in genera belonging to widely divergent families, while two genera of the same family may have them in different positions. It follows, therefore, that they have been independently acquired in the course of the evolution of the Coleoptera.
Stridulating organs among beetle-larvae have been noted, especially in the wood-feeding grub of the stag-beetles (Lucanidae) and their allies the Passalidae, and in the dung-eating grubs of the dor-beetles (Geotrupes), which belong to the chafer family (Scarabaeidae). These organs are described by J. C. Schiödte and D. Sharp; in the stag-beetle larva a series of short tubercles on the hind-leg is drawn across the serrate edge of a plate on the haunch of the intermediate legs, while in the Passalid grub the modified tip of the hind-leg acts as a scraper, being so shortened that it is useless for locomotion, but highly specialized for producing sound. Whatever may be the true explanation of stridulating organs in adult beetles, sexual selection can have had nothing to do with the presence of these highly-developed larval structures. It has been suggested that the power of stridulation would be advantageous to wood-boring grubs, the sound warning each of the position of its neighbour, so that adjacent burrowers may not get in each other’s way. The root-feeding larvae of the cockchafer and allied members of the Scarabaeidae have a ridged area on the mandible, which is scraped by teeth on the maxillae, apparently forming a stridulating organ.
Luminous Organs.—The function of the stridulating organs just described is presumably to afford means of recognition by sound. Some beetles emit a bright light from a portion of their bodies, which leads to the recognition of mate or comrade by sight. In the wingless female glow-worm (Lampyris, fig. 15 f) the luminous region is at the hinder end, the organ emitting the light consisting, according to H. von Wielowiejski (1882), of cells similar to those of the fat-body, containing a substance that undergoes oxidation. The illumination is intermittent, and appears to be under the control of the insect’s nervous system. The well-known “fire-flies” of the tropics are large click-beetles (Elateridae), that emit light from paired spots on the prothorax and from the base of the ventral abdominal region. The luminous organs of these beetles consist of a specialized part of the fat-body, with an inner opaque and an outer transparent layer. Its structure has been described by C. Heinemann, and its physiology by R. Dubois (1886), who considers that the luminosity is due to the influence of an enzyme in the cells of the organ upon a special substance in the blood. The eggs and larvae of the fire-flies are luminous as well as the perfect beetles.
Fossil History.—The Coleoptera can be traced back farther in time than any other order of insects with complete transformations, if the structures that have been described from the Carboniferous rocks of Germany are really elytra. In the Triassic rocks of Switzerland remains of weevils (Curculionidae) occur, a family which is considered by many students the most specialized of the order. And when we know that the Chrysomelidae and Buprestidae also lived in Triassic, and the Carabidae, Elateridae, Cerambycidae and Scarabaeidae, in Liassic times, we cannot doubt that the great majority of our existing families had already been differentiated at the beginning of the Mesozoic epoch. Coming to the Tertiary we find the Oligocene beds of Aix, of east Prussia (amber) and of Colorado, and the Miocene of Bavaria, especially rich in remains of beetles, most of which can be referred to existing genera.
Classification.—The Coleoptera have been probably more assiduously studied by systematic naturalists than any other order of insects. The number of described species can now hardly be less than 100,000, but there is little agreement as to the main principles of a natural classification. About eighty-five families are generally recognized; the difficulty that confronts the zoologists is the arrangement of these families in “superfamilies” or “sub-orders.” Such obvious features as the number of segments in the foot and the shape of the feeler were used by the early entomologists for distinguishing the great groups of beetles. The arrangement dependent on the number of tarsal segments—the order being divided into tribes Pentamera, Tetramera, Heteromera and Trimera—was suggested by E. L. Geoffroy in 1762, adopted by P. A. Latreille, and used largely through the 19th century. W. S. Macleay’s classification (1825), which rested principally on the characters of the larvae, is almost forgotten nowadays, but it is certain that in any systematic arrangement which claims to be natural the early stages in the life-history must receive due attention. In recent years classifications in part agreeing with the older schemes but largely original, in accord with researches on the comparative anatomy of the insects, have been put forward. Among the more conservative of these may be mentioned that of D. Sharp (1899), who divides the order into six great series of families: Lamellicornia (including the chafers and stag-beetles and their allies with five-segmented feet and plate-like terminal segments to the feelers); Adephaga (carnivorous, terrestrial and aquatic beetles, all with five foot-segments); Polymorpha (including a heterogeneous assembly of families that cannot be fitted into any of the other groups); Heteromera (beetles with the fore and intermediate feet five-segmented, and the hind-feet four-segmented); Phytophaga (including the leaf-beetles, and longhorns, distinguished by the apparently four-segmented feet), and Rhynchophora (the weevils and their allies, with head prolonged into a snout, and feet with four segments). L. Ganglbauer (1892) divides the whole order into two sub-orders only, the Caraboidea (the Adephaga of Sharp and the older writers) and the Cantharidoidea (including all other beetles), since the larvae of Caraboidea have five-segmented, two-clawed legs, while those of all other beetles have legs with four segments and a single claw. A. Lameere (1900) has suggested three sub-orders, the Cantharidiformia (including the Phytophaga, the Heteromera, the Rhynchophora and most of the Polymorpha of Sharp’s classification), the Staphyliniformia (including the rove-beetles, carrion-beetles and a few allied families of Sharp’s Polymorpha), and the Carabidiformia (Adephaga). Lameere’s classification is founded on the number of abdominal sterna, the nervuration of the wings, the number of malpighian tubules (whether four or six) and other structural characters. Preferable to Lameere’s system, because founded on a wider range of adult characters and taking the larval stages into account, is that of H. J. Kolbe (1901), who recognizes three sub-orders: (i.) the Adephaga; (ii.) the Heterophaga, including the Staphylinoidea, the Actinorhabda (Lamellicornia), the Heterorhabda (most of Sharp’s Polymorpha), and the Anchistopoda (the Phytophaga, with the ladybirds and some allied families which Sharp places among the Polymorpha); (iii.) the Rhynchophora.
Students of the Coleoptera have failed to agree not only on a system of classification, but on the relative specialization of some of the groups which they all recognize as natural. Lameere, for example, considers some of his Cantharidiformia as the most primitive Coleoptera. J. L. Leconte and G. H. Horn placed the Rhynchophora (weevils) in a group distinct from all other beetles, on account of their supposed primitive nature. Kolbe, on the other hand, insists that the weevils are the most modified of all beetles, being highly specialized as regards their adult structure, and developing from legless maggots exceedingly different from the adult; he regards the Adephaga, with their active armoured larvae with two foot-claws, as the most primitive group of beetles, and there can be little doubt that the likeness between larvae and adult may safely be accepted as a primitive character among insects. In the Coleoptera we have to do with an ancient yet dominant order, in which there is hardly a family that does not show specialization in some point of structure or life-history. Hence it is impossible to form a satisfactory linear series.
In the classification adopted in this article, the attempt has been made to combine the best points in old and recent schemes, and to avoid the inconvenience of a large heterogeneous group including the vast majority of the families.