THE THORAX AND ITS APPENDAGES

a. The thorax; its external anatomy

The middle region of the body is called the thorax, and in general consists of three segments, which are respectively named the prothorax, mesothorax, and metathorax (Figs. 88, 89, 98).

Fig. 88.—External anatomy of Melanoplus spretus, the head and thorax disjointed.

The thorax contains the muscles of flight and those of the legs, besides the fore intestine (œsophagus and proventriculus), as well as, in the winged insects, the salivary glands.

In the more generalized orders, notably the Orthoptera, the three segments are distinct and readily identified.

Fig. 89.—Locust, Melanoplus, side view, with the thorax separated from the head and abdomen, and divided into its three segments.

Each segment consists of the tergum, pleurum, and sternum. In the prothorax these pieces are not subdivided, except the pleural; in such case the tergum is called the pronotum. The prothorax is very large in the Orthoptera and other generalized forms, as also in the Coleoptera, but small and reduced in the Diptera and Hymenoptera. In the winged forms the tergum of the mesothorax is differentiated into four pieces or plates (sclerites). These pieces were named by Audouin, passing from before backwards, the præscutum, scutum, scutellum, and postscutellum. In the nymph stage and in the wingless adults of insects such as the Mallophaga, the true lice, the wingless Diptera, ants, etc., these parts by disuse and loss of the wings are not differentiated. It is therefore apparent that their development depends on that of the muscles of flight, of which they form the base of attachment. The scutum is invariably present, as is the scutellum. The former in nearly all insects constitutes the larger part of the tergum, while the latter is, as its name implies, the small shield-shaped piece directly behind the scutum.

Fig. 90.—Thorax of Telea polyphemus, side view, pronotum not represented: em, epimerum of prothorax, the narrow piece above being the prothoracic episternum; ms, mesoscutum; scm, mesoscutellum; ms″, metascutum; scm‴, metascutellum; pt, a supplementary piece near the insertion of tegulæ; w, pieces situated at the insertion of the wings, and surrounded by membrane; epm″, episternum of the mesothorax; em″, epimerum of the same; epm‴, episternum of the metathorax; em‴ epimerum of the same, divided into two pieces; c′, c″, c‴, coxæ; te′, te″, te‴, trochantines; tr, tr, tr, trochanters. A, tergal view of the mesothorax of the same; prm, præscutum; ms, scutum; scm, scutellum; ptm, postscutellum; t, tegula.

The præscutum and postscutellum are usually minute and crowded down out of sight between the opposing segments. As seen in Fig. 90, the præscutum of most moths (Telea) is a small rounded piece, bent vertically down so as not to be seen from above. In Polystœchotes and also in Hepialus the præscutum is large, well-developed, triangular, and wedged in between the two halves of the scutum. The postscutellum is still smaller, usually forming a transverse ridge, and is rarely used in taxonomy.

Fig. 91.—Thorax of the house-fly: prn, pronotum; prsc, præscutum; sc′, mesoscutum; sct′, mesoscutellum; psct′, postscutellum; al, insertion of squama, extending to the insertion of the wings, which have been removed; msphr, mesophragma; h, balancer (halter); pt, tegula; mtn, metanotum; epis, epis′, epis″, episternum of pro-, meso-, and metathorax; epm′, epm″, meso- and meta-epimerum; st′, st″, meso- and metasternum; cx′, cx″, cx‴, coxæ; tr′, tr″, tr‴, trochanters of the three pairs of legs; sp′, sp″, sp‴, sp‴′, sp‴″, first to fifth spiracles; tg′, tg″, tergites of first and second abdominal segments; u′, u″, urites.

The metathorax is usually smaller and shorter than the mesothorax, being proportioned to the size of the wings. In certain Neuroptera and in Hepialidæ and some tineoid moths, where the hind wings are nearly as large as those of the anterior pair, the metathorax is more than half or nearly two-thirds as large as the mesothorax. In Hepialidæ the præscutum is large and distinct, while the scutum is divided into two widely separated pieces. The postscutellum is nearly or quite obsolete.

The pleurum in each of the three thoracic segments is divided into two pieces; the one in front is called the episternum, since it rests upon the sternum; the other is the epimerum. To these pieces, with the sternum in part, the legs are articulated (Fig. 89).

Between the episterna is situated the breastplate or sternum, which is very large in the more primitive forms, as the Orthoptera, and is small in the Diptera and Hymenoptera.

Fig. 92.—Prothorax of Geometra papilionaria: n, notum; p, pleura; st, sternum; pt, patagia; m, membrane; f, femur; h, a hook bent backwards and beneath, and connecting the pro- with the mesothorax.—After Cholodkowsky.

The episterna and epimera are in certain groups, Neuroptera, etc., further subdivided each into two pieces (Fig. 102). The smaller pieces, hinging upon each other and forming the attachments of the muscles of flight, differ much in shape and size in insects of different orders. The difference in shape and degree of differentiation of these parts of the thorax is mentioned and illustrated under each order, and reference to the figures will obviate pages of tedious description. A glance, however, at the thorax of a moth, fly, or bee, where these numerous pieces are agglutinated into a globular mass, will show that the spherical shape of the thorax in these insects is due to the enlargement of one part at the expense of another; the prothoracic and metathoracic segments being more or less atrophied, while the mesothorax is greatly enlarged to support the powerful muscles of flight, the fore wings being much larger than those appended to the metathorax. In the Diptera, whose hinder pair of wings are reduced to the condition of halteres, the reduction of the metathorax as well as prothorax is especially marked (Fig. 91).

The patagia.—On each side of the pronotum of Lepidoptera are two transversely oval, movable, concavo-convex, erectile plates, called patagia (Fig. 92). On cutting those of a dry Catocala in two, they will be seen to be hollow. Cholodkowsky[[19]] states that they are filled with blood and tracheal branches; and he went so far as to regard them as rudimentary prothoracic wings, in which view he was corrected by Haase,[[20]] who compares them with the tegulæ, regarding them also as secondary or accessory structures.

The tegulæ.—On the mesothorax are the tegulæ of Kirby (pterygodes of Latreille, paraptera of McLeay, hypoptère or squamule), which cover the base of the fore wings, and are especially developed in the Lepidoptera (Fig. 90, A, t) and in certain Hymenoptera (Fig. 95, c).

The external opening of the spiracles just under the fore wings, is situated in a little plate called by Audouin the peritreme.

Fig. 93.—Transformation of the bumble bee, Bombus, showing the transfer of the 1st abdominal larval segment (c) to the thorax, forming the propodeum of the pupa (D) and imago; n, spiracle of the propodeum. A, larva; a, head; b, 1st thoracic; c, 1st abdominal segment. B, semipupa; g, antenna; h, maxillæ; i, 1st; j, 2d leg; k, mesoscutum; l, mesoscutellum; m, metathorax; d, urite (sternite of abdomen); e, pleurite; f, tergite; o, ovipositor; r, lingua; q, maxilla.

In the higher or aculeate Hymenoptera, besides the three segments normally composing the thorax, the basal abdominal segment is during the change from the larva to the pupa transferred to this region, making four segments. This first abdominal is called “the median segment” (Figs. 93–95). In such a case the term alitrunk has been applied to this region, i.e. the thorax, as thus constituted. Latreille wrongly stated that in the Diptera the first abdominal segment also entered into the composition of the thorax; but Brauer has fully disproved that view, as may be seen by an examination of his sketches which we have copied (Fig. 94).

Fig. 94.—7, 8, thorax of Tipula gigantea; 9, of Leptis; 10, thorax of Tabanus bromius after the removal of the abdomen, in order to bring into view the inner mesophragma (f), and to show the extension of the metathorax g and g′; tr, trochanter; 11, hind end of the mesothorax, the entire metathorax, and the 1st and 2d abdominal segments of Volucella zonaria, seen from the side. The internal mesophragma (f), and the position of the muscle inserted in it, are indicated by the two lines M. p, Callus postalaris; pr (pz in 8), callus præalaris Osten Sacken (= “patagium” of some authors); g, metanotum; g′, metepimerum, “segment médiaire” of Latreille (wrongly considered by him to be the 1st abdominal segment); 4, metasternum (hypopleura of Osten Sacken); 5 (? “episternum of metathorax” (Brauer) = metapleura of Osten Sacken); 6, and also H, halter; st1, mesothoracic stigma; st2, metathoracic stigma; st3, first abdominal stigma; γ, dorsopleural; δ, sternopleural; ε, mesopleural sutures; h, 1st, i, 2d, abdominal segment; al, wing; alul, alula. 12, the head and the three thoracic rings, and the 1st abdominal segment of Ephemera vulgata, the connecting membranes are in white: a, prothorax; b, præscutum; c, scutum; d, scutellum; e, postscutellum; ps, postscutellum of mesothorax.—After Brauer.

Fig. 95.—Alitrunk of Sphex chrysis: A, dorsal aspect; a, pronotum; b, mesonotum; c, tegula; d, base of fore,—e, of hind, wing; f, g, divisions of metanotum; h, median (true first abdominal) segment; i, its spiracle; k, second abdominal segment, usually called the petiole or first abdominal segment. B, posterior aspect of the median segment; a, upper part; b, superior,—c, inferior, abdominal foramen; d, ventral plate of median segment; e, coxa.—After Sharp.

The sternum is in rare cases subdivided into two halves, as in the meso- and metathorax of the cockroach; in Forficula the prosternum is divided into four pieces besides the sternum proper (Fig. 96); and in Embia, also, the sternites, according to Sharp, are complex.

Fig. 96.—Sternal view of pro-, meso-, and metathorax of Forficula tæniata: pst, præsternum, divided into 4 pieces; st, pro-, st′, meso-, st″, metasternum; cx, coxa; not, notum.

Fig. 97.—A, under surface of prothorax, or prosternum, of Dyticus circumflexis: 2.g, prosternum; 2.f, episternum; 2.h, epimerum; 2.s, antefurca or entothorax.

Fig. 98.—Meso- (G2) and metathoracic ganglia (G1), with the apodemes of Gryllotalpa.—After Graber.

Fig. 99.—Parts of the mesothorax of Dyticus: A, mesosternum; 3.a, præscutum; 3.b, scutum; 3.c, scutellum; 3.d, postscutellum; 3.e, parapteron; 3.g, mesosternum; 3.f, episternum; 3.h, epimerum; 3.s, medifurca or entothorax.

Fig. 100.—Parts of the metathorax of Dyticus: A, metasternum; 4.a, præscutum; 4.b, scutum; 4.c, scutellum; 4.d, postscutellum; 4.e, parapteron; 4.f, episternum; 4.g, metasternum; 4.h, epimerum; 4.s, postfurca.—This and Figs. 97 and 99 from Audouin, after Newport.

The apodemes.—The thorax is supported within by beam-like processes, or apodemes, which pass inward and also form attachments for the muscles. Those passing up from the sternum form the entothorax of Audouin, and the process of each thoracic segment is called respectively the antefurca, medifurca, and postfurca. In the Orthoptera (Caloptenus and Anabrus), the antefurca is large, thin, flattened, directed forward, and bounds each side of the prothoracic ganglion. In the Coleoptera two plates (Fig. 97, 2.s) arise from the inside of the sternum and “form a collar or leave a circular hole between them for the passage of the nervous cord” (Newport). The medifurca is a pair of flat processes which diverge and bridge the commissure, while the postfurca is situated under the commissure. In beetles (Dyticus) Newport states that it is expanded into two broad plates, to which the muscles of the posterior legs are attached. Graber also notices in the mole cricket between the apodemes of the meso- and metathorax, a flattened spine (Fig. 98, do) with two perforations through which pass the commissures connecting the ganglia. Besides these processes there are large, thin, longitudinal partitions passing down from the tergum (or dorsum), called phragmas; they are most developed in those insects which fly best, i.e. in Coleoptera (Figs. 97–101), Lepidoptera, Diptera, and Hymenoptera, none being developed in the prothorax. (The term phragma has also been applied to a partition formed by the inflexed hinder edge of this segment, and is present only in those insects in which the prothorax is movable.—Century Dictionary.) All these ingrowths may be in general termed apodemes. There are similar structures in Crustacea and also in Limulus; but Sharp restricts this term to minute projections in beetles (Goliathus) situated at the sides of the thorax near the wings. (Insecta, p. 103, Fig. 57.) The internal processes arising from the sternal region have been called endosternites.

Fig. 101.—Internal skeleton of Lucanus cervus, ♂, head: A, antenna; f, mandible; d, mentum; 2, 4, tendons of mandible; f, u, t, parts of the tentorium; 3 e, labial muscles. Thorax: 2, prothorax; 3, 4, meso- and metathorax fused solidly together; 3 r, acetabulum of prothorax, into which the coxa is inserted; 2 s, sternum; 3t, acetabulum of mesothorax, 4r, of metathorax; 3 s, mesothoracic sternum fused with that of the metathorax (4g); 4 s, apodeme.—After Newport.

The acetabula.—These are the cavities in which the legs are inserted. They are situated on each side of the posterior part of the sternum, in each of the thoracic segments. They are, in general, formed by an approximation of the sternum and epimerum, and sometimes, also, of the episternum, as in Dyticus (Fig. 97, A). This consolidation of parts, says Newport, gives an amazing increase of strength to the segments, and is one of the circumstances which enables the insect to exert an astonishing degree of muscular power.

Tabular View of the Segments, Pieces, and Appendages of the Thorax
Name of SegmentPieces (Sclerites)Appendages
1. ProthoraxPronotum, sometimes differentiated into
Scutum1st pair of legs
ScutellumPatagia
Episternum
Epimerum
Sternum
Antefurca
2. MesothoraxPræscutum
Scutum2d pair of legs
Scutellum1st pair of wings
ProscutellumTegulæ
EpisternumSquamæ (Alulæ)
EpimerumPeritreme
Sternum
Mesofurca
Mesophragma
Apodemes
3. MetathoraxPræscutum
Scutum3d pair of legs
Scutellum2d pair of wings
Postscutellum(Halteres of Diptera)
Episternum
Epimerum
Sternum
Postfurca
Metaphragma
Apodemes

Fig. 102.—External anatomy of the trunk of Hydröus piceus: A, sternal—B, tergal aspect; 2, pronotum; 2 a, prosternum; 2 f, episternum; 3 a, præscutum; 3 b, scutum; 3 c, scutellum; 3 d, postscutellum; 3 g, mesosternum; 3 h, episternum; 3 f, epimerum; 3 i, crest of the mesosternum; 3 a, parapteron; 3 k, coxa; 4 a, metapræscutum; 4 b, metascutum; 4 c, metascutellum; 4 d, postscutellum; 4 e, tegula; 4 f, episternum; 4 h, epimerum; 4 g, metasternum; 4 i, crest of metasternum; 4 k and l, coxa; 4 m, trochanter; n, femur; o, tibia; p, tarsus; q, unguis; 7–11, abdominal segments.—After Newport.

b. The legs: their structure and functions

The mode of insertion of the legs to the thorax is seen in Figs. 90, 97, 101, and 103. They are articulated to the episternum, epimerum, and sternum, taken together, and consist of five segments. The basal segment or joint is the coxa, situated between the episternum and trochanter. The coxa usually has a posterior subdivision or projection, the trochantine; sometimes, as in Mantispa (Fig. 103), the trochantine is obsolete. We had previously supposed that the trochantine was a separate joint, but now doubt whether it represents a distinct segment of the leg, and regard it as only a subdivision of the coxa. It is attached to the epimerum, and is best developed in Panorpidæ, Trichoptera, and Lepidoptera. In the Thysanura the trochantine is wanting, and in the cockroach it merely forms a subdivision of the coxa, its use being to support the latter. The second segment is the trochanter, a more or less short spherical joint on which the leg proper turns; in the parasitic groups (Ichneumonidæ, etc., Fig. 104) it is usually divided into two pieces, though there are some exceptions. The trochanter is succeeded by the femur, tibia, and tarsus, the latter consisting of from one to five segments, the normal number being five. Tuffen West believed that the pulvillus is the homologue of an additional tarsal joint, “a sixth tarsal joint.” The last tarsal segment ends in a pair of freely movable claws (ungues), which are modified setæ; between the claws is a cushion-like pad or adhesive lobe, called the empodium or pulvillus (Fig. 105, also variously called arolium, palmula, plantula, onychium, its appendage being called paronychium and also pseudonychium). It is cleft or bilobate in many flies, but in Sargus trilobate. All these parts vary greatly in shape and relative size in insects of different groups, especially Trichoptera, Lepidoptera, Diptera, and Hymenoptera. In certain flies (e.g. Leptogaster) the empodium is wanting (Kolbe). By some writers the middle lobe is called the empodium and the two others pulvilli.

Fig. 103.—Side view of meso- and metathorax of Mantispa brunnea, showing the upper and lower divisions of the epimerum (s. em′, s. em″, i. em′, i. em″); s. epis, i. epis″, the same of the episternum.

Fig. 104.—Divided (ditrochous) trochanter of an ichneumon: cx, coxa; tr, the two divisions of the trochanter; f, femur.—After Sharp.

The fore legs are usually directed forward to drag the body along, while the middle and hind legs are directed outward and backward to push the body onwards. While arachnids walk on the tip ends of their feet, myriopods, Thysanura, and all larval insects walk on the ends of the claws, but insects generally, especially the adults, are, so to speak, plantigrade, since they walk on all the tarsal joints. In the aquatic forms the middle and hind tarsi are more or less flattened, oar-like, and edged with setæ. In leaping insects, as the locusts and grasshoppers, and certain chrysomelids, the hind femora are greatly swollen owing to the development of the muscles within. The tibia, besides bearing large, lateral, external spines, occasionally bears at the end one or more spines or spurs called calcaria. The fore tibia also in ants, etc., bear tactile hairs, and chordotonal organs, as well as other isolated sense-organs (Janet), and, in grasshoppers, ears.

In the Carabidæ the legs are provided with combs for cleaning the antennæ (Fig. 107), and in the bees and ants these cleansing organs are more specialized, the pectinated spine (calcar) being opposed by a tarsal comb (Fig. 106, d; for the wax-pincers of bees, see g). In general the insects use their more or less spiny legs for cleansing the head, antennæ, palpi, wings, etc., and the adaptations for that end are the bristles or spinules on the legs, especially the tibiæ.

Fig. 105.—Foot of honey-bee, with the pulvillus in use: A, under view of foot; t, t, 3d–5th tarsal joints; a n, unguis; f h, tactile hairs; p v, pulvillus; cr, curved rod. B, side view of foot. C, central part of sole; pd, pad; cr, curved rod; pv, pulvillus unopened.—After Cheshire.

Fig. 106.—Modifications of the legs of different bees. A, Apis: a, wax-pincer and outer view of hind leg; b, inner aspect of wax-pincer and leg, with the nine pollen-brushes or rows of hairs; c, compound hairs holding grains of pollen; d, anterior leg, showing antenna-cleaner; e, spur on tibia of middle leg. B, Melipona: f, peculiar group of spines at apex of tibia of hind leg; g, inner aspect of wax-pincer and first tarsal joint. C, Bombus: h, wax-pincer; i, inner view of the same and first tarsal joint, all enlarged.—From Insect Life, U. S. Div. Ent.

Osten Sacken states that among Diptera the aerial forms (Bombylidæ, etc.) with their large eyes or holoptic heads, which carry with them the power of hovering or poising, have weak legs, principally fit for alighting. On the other hand, the pedestrian or walking Diptera (Asilidæ, etc.) “use the legs not for alighting only, but for running, and all kinds of other work, seizing their prey, carrying it, climbing, digging, etc.; their legs are provided not only with spines and bristles, but with still other appendages, which may be useful, or only ornamental, as secondary sexual characters.”

Fig. 107.—End of tibia and tarsal joints of Anophthalmus; c, comb.

Tenent hairs.—Projecting from the lower surface of the empodium are the numerous “tenent hairs,” or holding hairs, which are modified glandular setæ swollen at the end and which give out a minute quantity of a clear adhesive fluid (Figs. 108, 109, 130, 134). In larval insects, and the adults of certain beetles, Coccidæ, Aphidæ, and Collembola, which have no empodium, there are one or more of these tenent hairs present. They enable the insect to adhere to smooth surfaces.

Fig. 108.—Transverse section through a tarsal joint of Telephorus, a beetle: ch, cuticula of the upper side; m, its matrix; ch′, the sole; m′, its matrix; h, adhesive hair; h′, tactile hair, supplied with a nerve (n′), and arising from a main nerve (n); n″, ganglion of a tactile hair; t, section of main trachea, from which arises a branch (t′); dr, glands which open into the adhesive hairs, and form the sticky secretion; e, chitinous thickening; s, sinew; b, membrane dividing the hollow space of the tarsal joint into compartments. See p. 111.—After Dewitz.

Striking sexual secondary characters appear in the fore legs of the male Hydrophilus, the insect, as Tuffen West observes, walking on the end of the tibia alone and dragging the tarsus after it. The last tarsal joint is enlarged into the form of an irregular hollow shield. The most completely suctorial feet of insects are those of the anterior pair of Dyticus (Fig. 132). The under side of the three basal joints is fused together and enlarged into a single broad and nearly circular shield, which is convex above and fringed with fine branching hairs, and covered beneath with suckers, of which two are exceptionally large; by this apparatus of suckers the male is enabled to adhere to the back of its mate during copulation. The line branching hairs around the edge prevent the water from penetrating and thus destroying the vacuum, “while if the female struggle out of the water, by retaining the fluid for some time around the sucker, they will in like manner under these altered conditions equally tend to preserve the effectual contact.” (Tuffen West.)

Fig. 109.—Cross-section through tarsus of a locust: ch, cuticula of upper side,—ch′, ch″, ch‴, of sole; ch, tubulated layer; ch″, lamellate layer; ch‴, inner projections of ch″. Other lettering as in Fig. 101. See p. 113.—After Dewitz.

In the saw-flies (Uroceridæ and Tenthredinidæ) and other insects, there are small membranous oval cushions (arolia, Figs. 109 and 131) beneath each or nearly each tarsal joint.

The triunguline larvæ of the Meloidæ are so called from apparently having three ungues, but in reality there is only a single claw, with a claw-like bristle on each side.

Why do insects have but six legs?—Embryology shows that the ancestors of insects were polypodous, and the question arises to what cause is due the process of elimination of legs in the ancestors of existing insects, so that at present there are no functional legs on the abdomen, these being invariably restricted (except in caterpillars) to the thorax, and the number never being more than six. It is evident that the number of six legs was fixed by heredity in the Thysanura, before the appearance of winged insects. We had thought that this restriction of legs to the thorax was in part due to the fact that this is the centre of gravity, and also because abdominal legs are not necessary in locomotion, since the fore legs are used in dragging the insect forwards, while the two hinder pairs support and push the body on. Synchronously with this elimination by disuse of the abdominal legs, the body became shortened, and subdivided into three regions. On the other hand, as in caterpillars, with their long bodies, the abdominal legs of the embryo persist; or if it be granted that the prop-legs are secondary structures, then they were developed in larval life to prop up and move the abdominal region.

The constancy of the number of six legs is explained by Dahl as being in relation to their function as climbing organs. One leg, he says, will almost always be perpendicular to the plane when the animal is moving up a vertical surface; and, on the other hand, we know that three is the smallest number with which stable equilibrium is possible; an insect must therefore have twice this number, and the great numerical superiority of the class may be associated with this mechanical advantage. (This numerical superiority of insects, however, seems to us to be rather due to the acquisition of wings, as we have already stated on pages 2 and 120.)

Loss of limbs by disuse.—Not only are one or both claws of a single pair, or those of all the feet atrophied by disuse, but this process of reduction may extend to the entire limb.

In a few insects one of the claws of each foot is atrophied, as in the feet of the Pediculidæ, of many Mallophaga, all of the Coccidæ, in Bittacus, Hybusa (Orthoptera), several beetles of the family Pselaphidæ, and a weevil (Brachybamus). Hoplia, etc., bear but a single claw on the hind feet, while the allied Gymnoloma has only a single claw on all the feet. Cybister has in general a single immovable claw on the hind feet, but Cybister scutellaris has, according to Sharp, on the same feet an outer small and movable claw. In the water bugs, Belostoma, etc., the fore feet end in a single claw, while in others (Corisa) both claws are wanting on the fore feet. Corisa also has no claws on the hind feet; Notonecta has two claws on the anterior four feet, but none on the hind pair. In Diplonychus, however, there are two small claws present. (Kolbe.)

Fig. 110.—Last tarsal joint of Melolontha vulgaris, drawn as if transparent to show the inner mechanism: un, claws; str, extensor plate; s, tendon of the flexor muscle; vb, elastic membrane between the extensor plate and the sliding surface u; krh, process of the ungual joint; emp, extensor spine, and th, its two tactile hairs.—After Ockler, from Kolbe.

Among the Scarabæidæ, the individuals of both sexes of the fossorial genus Ateuchus (A. sacer) and eight other genera, among them Deltochilum gibbosum of the United States, have no tarsi on the anterior feet in either sex. The American genera Phanæus (Fig. 111), Gromphas, and Streblopus have no tarsal joints in the male, but they are present in the female, though much reduced in size, and also wanting, Kolbe states, in many species of Phanæus. The peculiar genus Stenosternus not only lacks the anterior feet, but also those of the second and third pair of legs are each reduced to a vestige in the shape of a simple, spur-like, clawless joint. The ungual joint is wanting in the weevil Anoplus, and becomes small and not easily seen in four other genera.

Ryder states that the evidence that the absence of fore tarsi in Ateuchus is due to the inheritance of their loss by mutilation is uncertain. Dr. Horn suggests that cases like Ateuchus and Deltochilum, etc., “might be used as an evidence of the persistence of a character gradually acquired through repeated mutilation, that is, a loss of the tarsus by the digging which these insects perform.” On the other hand, the numerous species of Phanæus do quite as much digging, and the anterior tarsi of the male only are wanting. “It is true,” he adds, “that many females are seen which have lost their anterior tarsi by digging; have, in fact, worn them off; but in recently developed specimens the front tarsi are always absent in the males and present in the females. If repeated mutilation has resulted in the entire disappearance of the tarsi in one fossorial insect, it is reasonable to infer that the same results should follow in a related insect in both sexes, if at all, and not in the male only. It is evident that some other cause than inherited mutilation must be sought for to explain the loss of the tarsi in these insects.” (Proc. Amer. Phil. Soc., Philadelphia, 1889, pp. 529, 542.)

Fig. 111.—Fore tibia of Phanæus carnifex, ♂, showing no trace of the tarsus.

Fig. 112.—Fore leg of the mole-cricket: A, outer, B, inner, aspect; e, ear-slit.—After Sharp.

The loss of tarsi may be due to disuse rather than to the inheritance of mutilations. Judging by the enlarged fore tibiæ, which seem admirably adapted for digging, it would appear as if tarsi, even more or less reduced, would be in the way, and thus would be useless to the beetles in digging. Careful observations on the habits of these beetles might throw light on this point. It may be added that the fore tarsi in the more fossorial Carabidæ, such as Clivina and Scarites, as well as those of the larva of Cicada and those of the mole crickets (Fig. 112), are more or less reduced; there is a hypertrophy of the tibiæ and their spines. The shape of the tibia in these insects, which are flattened with several broad triangular spines, bears a strong resemblance to the nails or claws of the fossorial limbs of those mammals which dig in hard soil, such as the armadillo, manis, aardvark, and Echidna. The principle of modification by disuse is well illustrated in the following cases.

In many butterflies the fore legs are small and shortened, and of little use, and held pressed against the breast. In the Lycænidæ the fore tarsi are without claws; in Erycinidæ and Libytheidæ the fore legs of the males are shortened, but completely developed in the females, while in the Nymphalidæ the fore legs in both sexes are shortened, consisting in the males of one or two joints, the claws being absent in the females. Among moths loss of the fore tarsi is less frequent. J. B. Smith[[21]] notices the lack of the fore tarsi in the male of a deltoid, Litognatha nubilifasciata (Fig. 113), while the hind feet of Hepialus hectus are shortened. In an aphid (Mastopoda pteridis, Esl.) all the tarsi are reduced to a single vestigial joint (Fig. 114).

Fig. 113.—Leg of Litognatha: cx, coxa; f, femur; t, tibia; ep, its epiphysis, and sh, its shield-like process. The tarsus entirely wanting.—After Smith.

Entirely legless adult insects are rare, and the loss is clearly seen to be an adaptation due to disuse; such are the females of the Psychidæ, the females of several genera of Coccidæ (Mytilaspis, etc.), and the females of the Stylopidæ.

Apodous larval insects are common, and the loss of legs is plainly seen to be a secondary adaptive feature, since there are annectant forms with one or two pairs of thoracic legs. All dipterous and siphonapterous larvæ, those of all the Hymenoptera except the saw-flies, a few lepidopterous larvæ, some coleopterous, as those of the Rhyncophora, Buprestidæ, Eucnemidæ, and other families, and many Cerambycidæ are without any legs. In Eupsalis minuta, belonging to the Brenthidæ, the thoracic legs are minute.

The legs of larvæ end in a single claw, upon the tips of which the insect stands in walking.

c. Locomotion (walking, climbing, and swimming)

Mechanics of walking.—To Graber we owe the best exposition of the mechanics of walking in insects.

“The first segment of the insect leg,” he says, “upon which the weight of the body rests first of all, is the coxa. Its method of articulation is very different from that of the other joints. The enarthrosis affords the most extensive play, particularly in the Hymenoptera and Diptera.”

In the former the development of their social conditions is very closely connected with the freest possible use of the legs, which serve as hands. In the beetles, however, which are very compactly built, there exists a solid articulation whereby the entire hip rests in a tent-like excavation of the thorax, and can only be turned round a single axis, as may be seen in Fig. 115, where c represents the imaginary revolving axis and d the coxa. In the case we are supposing, therefore, only a backward and forward movement of the coxa is possible, the extent of the play of which depends on the size of the coxal pan, as well as certain groin or bar-like structures which limit further rotation. In the very dissimilar arrangement which draws in the fore, middle, and hind legs toward the body it is self-evident that their extent of action is also different. This arrangement seems to be most yielding on the fore legs, where the hips, to confine ourselves to the stag-beetles, can be turned backward and forward 60° from the middle or normal position, and therefore describe on the whole a curve of 120°. The angle of turning on the middle leg hardly exceeds a legitimate limit, yet a forward as well as a backward rotation takes place. The former is entirely wanting in the hind hips; they can only be moved backward.

Fig. 114.—Leg of an Aphid, with the tarsus (t) much reduced: 1, 2, 3, legs of 1st, 2d, and 3d pairs.

The number and strength of the muscles on which the rotation of the hips depends, correspond with these varying movements of the individual legs. Thus, according to Straus Durckheim, the fore coxa of many beetles possesses five separate muscles and four forward and one backward roll; the middle coxa a like number of muscles but only two forward rolls, while the hind hips succeed in accomplishing each of the motions named with a single muscle.

One can best see how these muscles undertake their work, and above all how they are situated, if he lays bare the prothorax of the stag beetle (Fig. 116). Here may be seen first the thick muscle which turns to the front the rotating axis in its cylindrical pan, and thus helps to extend the leg, while two other tendons, which take the opposite direction, are fitted for reflex movements.

Fig. 115.—Mechanics of an insect’s leg: d, coxa,—c, axis of revolution; a and b, the coxal muscles; e, trochanter muscle (elevator of the femur); f, extensor,—g, flexor, of the tibia (pn); n, tibial spine; h, flexor.—i, extensor, of the foot; k, extensor,—l, flexor, of the claw; po, place of flexure of the tibia; p1q, leg after being turned back by the coxa.—p1r, by the simultaneous flexure of the tibia. The resulting motion of the end of the tibia, through the simultaneous movement (no) and revolution (nq), indicates the curve nr.—After Graber.

In Fig. 115 the muscles mentioned above, and their modes of working, may be distinguished by the arrows a and b.

In order to simplify matters, we will imagine the second component part of the normal insect leg, i.e. the trochanter (Figs. 116, 117, r), as grown together with the third lever, i.e. the femur, as the movement of both parts mostly takes place uniformly.

Fig. 116.—Section of the fore leg of a stag-beetle, showing the muscles: S, extensor,—B, flexor, of the leg; s, extensor,—b, flexor, of the femur; o, femur; u, tibia; f, tarsus; k, claw; 109, s, extensor,—b, flexor, of the femoro-tibial joint, both enlarged.—After Graber.

The pulling of the small trochanter muscle works against the weight of the body when this is carried over on to the trochanter by means of the coxa, as seen at the arrow e in Fig. 115. It may be designated as the femoral lever.

The plane of direction in which the femur, as seen by the rotation just mentioned, is moved, exactly coincides in insects with that of the tibia and the foot, while all can be simultaneously raised or dropped, or, as the case may be, stretched out or retracted. Therein, therefore, lies an essential difference from the fully developed extremities of vertebrates among which, even on the lever arms which are stationary at the end, an extensive turning is possible.

The muscles which move the tibia, and indirectly the femur, also consist of an extensor muscle which is situated in the upper side of the femur (Fig. 116, s, Fig. 115, f), and of a flexor (Fig. 116, b, Fig. 115, g), which lies under the former.

The stilt-like spines on the point (Figs. 115 and 118, L3n) on which this segment is directly supported are important parts of the tibia. (Graber.)

Fig. 117.—Left fore leg of a cerambycid beetle: h, coxa; r, trochanter; o, femur; u, tibia; f, tarsus; k, claw.—After Graber.

Considering the respective positions of the individual levers of the leg and the nature of the materials of which they are made, the legs of insects may be likened, as Graber states, to elastic bows, which, when pressed down together from above, their own indwelling elasticity is able to raise again and thus keep the body upright.

This is very plainly shown in certain stilt-legged bark-beetles, in which, as in a rubber doll, as soon as the body is pressed down on the ground, the organs of motion extend again without the intervention of muscles; indeed this experiment succeeds even with dead, but not yet wholly stiff, insects.

Graber then turns to the analysis of the movements of insect legs when in motion, and the mode of walking of these insects in general. This subject had been but slightly investigated until Graber made a series of observations and experiments, of which we can give only the most important results.

The locomotion of insects is an extremely complicated subject.

Let us consider, Graber says, first, a running or carabid beetle, when walking merely with the fore and hind legs. The former will be bent forward and the latter backward.

“Let us begin with the left fore leg (Fig. 118, L1). Let the same be extended and fixed on the ground by means of its sharp claws and its pointed heel. Now what happens when the tibial flexors draw together? As the foot, and therefore the tibia also, have a firm position, then the contraction of the muscles named must cause the femur to approach the tibia, whereby the whole body is drawn along with it. This individual act of motion may be well studied in grasshoppers when they are climbing on a twig by stretching out their long fore leg directly forward, and then drawing up the body through the shortening of the tibial flexors until the middle leg also reaches the branch.

“But while the fore legs advance the body by drawing the free lever to the fixed leg-segment, the hind legs do this in exactly the opposite way. The hind leg, namely, seeks to stretch out the tibia, and thus to increase the angle of the knee (R3), thereby giving a push on the ground, by means of which the body is shoved forward a bit.

“Though it might be supposed that the feet would remain stationary during the extension or retraction of the limbs, this never occurs in actual walking. Not merely the upper, but also the lower, thigh is either drawn in or stretched out, as the case may be. The latter then describes a straight line with its point during this scraping or scratching motion (Fig. 115, no), which is obviously the chord to that quadrant which would be drawn by the tibia or foot in a yielding medium, as water, for instance. But even this motion results extremely rarely, and never in actual walking. If we fix our eye anew upon the fore leg at the very moment when it is again retracted, after the resultant ‘fixing,’ we shall then observe that the hip also is simultaneously turned backward in a definite angle. The tibia would describe the arc nq (Fig. 115) by means of the latter alone.

“This plane, in conjunction with the rectilinear ‘movement’ (no) obtained by the retraction of the tibia, produces a path (nr), and this is what is actually described by a painted foot upon a properly prepared surface, as a sheet of paper;[[22]] supposing, however, that the body in the meantime is not moved forward by other forces. In the last case, and this indeed always takes place in running, the trunk is moved a bit forward, together with the leg which is just describing its curve with a rapidity corresponding to the momentum obtained; the result of this is that the curve of the foot from its beginning (n) to its end (a) bends round close to itself, just as a man who, when on board a ship in motion, walks across it diagonally, and yet on the whole moves forward, because his line of march, uniting with that of the ship, results in a change of position in space.

“The case is the same in the middle and hind legs, which must make a double course also, yet in such a way that the straight line is drawn, not during the retraction, but during the extension; during which, however, quite as in the fore leg, the members mentioned (R3) gradually approach the body.

“When the legs have reached the maximum of their retraction, or of their extension, as the case may be, and therefore the end of their active course for that time, then begins the opposite or backward movement; that is, the fore legs are again extended, while their levers draw the remaining legs together again.

Fig. 118.—A Carabus beetle in the act of walking or running: three legs (L1, R2, L3) are directed forward, while the others (R1, L2, R3), which are directed backward toward the tail, have ended their activity; ab, cd, and ef are curves described by the end of the tibiæ, and passing back to the end of the body; bh, di, and fg are curves described by the same legs during their passive change of position.—After Graber.

“At the same time, as we may see by the uniting leg, the limb is either a little raised, that there may be no unnecessary friction, or it remains during the passive step also, with its means of locomotion in slight contact with the ground.

“The curve of two steps, as inscribed by the end of the tibia of the left fore leg of a stag-beetle, affords an instructive summary of the conditions of which we have been speaking (Fig. 121, B). We see two curves. The thick one (ab), directed toward the axis of the body, corresponds to the effective act of a single walking function, which brings the body a bit forward; the thinner, on the other hand, or we might say the hair line (bc), which, however, is but rarely made quite clearly, is produced by the ineffectual backward movement, by which the insect again approaches its working posture (c). It is at first placed at some distance from the body, in order that (like c also) it may draw near to the body again; but in such a way, naturally, that it coincides with the starting-point of the following active curve (cd). It is evident that even the passive curve is not the imprint of the movement accomplished exclusively by the leg, for this latter, while struggling to reach its resting-place, is really involuntarily carried forward with the rest of the body.

“The scroll-like lines drawn by the swimming beetle (Dyticus), with the large, sharp points of its hind tibia, are also very instructive (Fig. 119, A).

Fig. 119.—A, trail curves described by the tibial spines of the right and left hind limb of Dyticus. B, the same made by the right hind leg (r3) alone. Natural size.—After Graber.

Fig. 120.—The same by the two hind legs of Melolontha: a, the active and thickened section of the curve. Natural size.

Fig. 121.—A, track curves of two of the tibial spines of the left, middle legs of a stag-beetle. Natural size. B, the same enlarged; fg, the longitudinal axis of the trunk; cd and ab, the active curve passing inward,—bc and de, the passive going outward. C, two curves described by the left hind legs; in this case, the curves are not inwards or backwards, but partly directly inward (b), and in part obliquely forwards (a).

“The diversions and modifications in the course of the active step, as furnished by the moving factor of the remaining legs, are already clearly illustrated by the curves shown by the joints of the hind tibia of a May-beetle (Fig. 120) and a stag-beetle (Fig. 121, c). The actual faint line in this case does not run from the front toward the back, as would correspond to the active leg-motion, but either directly inward (Fig. 121, cb), or even somewhat to the front. In the May-beetles, and even more in the running garden-beetle, the curves of the hind legs present themselves as screw-like lines (Fig. 122, l3), while the scrawling of the remaining members (l1, l2) is much simpler.

“Inasmuch as we now have a cursory knowledge of the movements made by each individual leg for itself,—movements, however, which plainly occur very differently according to the structure of these appendages,—the question now is of the combined play, the total effect of all the legs taken together, and therefore of the walk and measure of the united work of the foot.

“In opposition to the caterpillars and many other crawling animals which extend their legs in pairs and really swing them by the worm-like mode of contraction of the dermo-muscular tube, the legs of fully grown insects are moved in the contrary direction and in no sense in pairs, but alternately—or, more strictly speaking, in a diagonal direction.

“For an examination of the gait of insects, we choose, for obvious reasons, those which have very long legs and which at the same time are slow walkers.

“Insects may be called ‘double-three-footed,’ from the manner in which they alternately place their legs. There are always three legs set in motion at the same time, or nearly so, while in the meantime the remaining legs support the body, after which they change places.

Fig. 122.—The same by the left fore (l1), middle (l2), and hind, leg (l3) of a Carabus. Natural size.

Fig. 123.-Tracks of a Blaps mortisaga marked by the differently painted tibial points: ●, tracks of fore, —○, middle, —/, hind leg. Natural size.

Fig. 124.—Tracks of Necrophorus vespilio. Natural size.

“To be more exact, it is usually thus: At first (Fig. 118) the left fore leg (L1) steps out, then follows the right middle leg (R2), and the left hind leg (L3). Then while the left fore leg begins to retract and thus make the backward movement, the right fore leg is extended, whereupon the left middle leg and the right hind leg are raised in the same order as the first three feet.”

Graber[[23]] painted the feet of beetles and let them run over paper, and goes on to say:

“Let us first pursue the tracks of the Blaps, for example (Fig. 123). Let the insect begin its motion. The left fore leg stands at a, the right middle leg at β, and the left hind leg at c. The corresponding number of the other set of three feet at α, b, γ. At the first step the three feet first mentioned advance to a′β′c′, the second set on the other hand to α′b′γ′. Thereby the tracks made by the successive steps fall quite, or almost quite, on each other, as appear also in the tracks of a burying beetle (Fig. 124).

“As the fore legs are directed forward and the hind legs backward, while the middle legs are placed obliquely, the reason of the more marked impressions of the latter is evident.

“The highest testimony to the precise exactitude and accuracy of the walking mechanism of insects is furnished by the fact that in most insects, and particularly in those most fleet of foot, which, whether they are running away or chasing their prey, must be able to rely entirely upon their means of locomotion;—the fact, we say, that whether they desire to move slowly or more quickly, the distances of the steps, measured by the length as well as by the cross-direction, hardly differ a hair’s breadth from one another, and this is also the case when the tarsi are cut off and the insects are obliged to run on the points of their heels (tibiæ).

“Thence, inasmuch as the trunk of insects is carried by two legs and by one on each side alternately, it may surely be concluded a priori that when walking it is inclined now to the right and now to the left, and that the track, too, which is left behind by a precise point of the leg, can in no wise be a straight line; and in reality this is not the case.

“A plainly marked regular curve, which approaches a sinuous line, as seen in Fig. 125, is often obtained by painting many insects, for example Trichodes, Meloë, etc., which, when running, either bring the end of their hind body near to the ground or into contact with it.

Fig. 125.—Tracks of Trichodes; the middle sinuous line is made by the tip of the abdomen. Natural size.

Fig. 126.—Tracks of another insect which, in running, can only use three legs (r1, l4, r3) which become indicated differently from normal conditions. Natural size.

Fig. 127.—The same of an insect crossing over a surface inclined 30° from the horizon, whereby the placing of the feet becomes changed. Natural size.—This and Figs. 120–126 after Graber.

“The locomotive machine of insects may be called, to a certain extent, a double set of three feet each, as most insects, and particularly those provided with a broad trunk, are able to balance themselves with one of these two sets of feet, and indeed when walking, as well as when standing still, can move about even better with one set of these feet than with four legs. In the latter case, that is, if one cuts off a pair of legs from an insect, the trunk can balance itself only with extreme difficulty, and there is therefore little prospect that insects will ever become four-footed.

“But if one compels insects to run on three legs, he will thus make the interesting discovery that to make up the deficiency they place the remaining feet and bring them to the ground somewhat differently than when the second set of feet is active. Figs. 124 and 126 may be compared for this purpose. The former shows the footprints of a burying beetle running with all six legs, the latter the track of the same insect, which, however, has at its disposal only the right fore leg, the left middle leg, and the right hind leg. One may plainly see here that the track of the hind leg on the right side (r3) approaches the track of the middle leg on the left side, and then further, that the right fore leg (r1) steps out more to the right to make up for the deficiency of the middle leg.

“A similar adaptation of the position of the legs, which is entirely dependent on the choice of the insect, may also be observed there, if one compels insects which are not provided with corresponding adhesive lobes to run away over crooked surfaces. Fig. 123 shows the footprints of a Blaps when running upon a horizontal plane. Fig. 127, on the contrary, shows the tracks of the legs when going diagonally over a gradually inclined surface. Here, also, the insect holds on with his fore and middle legs (r1, r2) stretched upward, whereby also the impressions on both sides come to lie farther apart than in the normal mode of walking.

“It will not surprise the reader who is familiar with the gait of crabs, to hear that many insects also understand the laudable art of going backward, wherein the hind legs simply change places with the fore legs.

“The jumping motion of insects may be best studied in grasshoppers. When these insects are preparing for a jump, they stretch out the upper thigh horizontally, clap the tibiæ together, and also retract the foot-segment. After a slight pause for rest, during which they are getting ready for the jump, they then jerk the tibiæ suddenly backward and against the ground with all their strength by means of the extensor muscles.”

The correctness of Graber’s views has been confirmed by Marey by instantaneous photographs (Figs. 128, 129).

Locomotion on smooth surfaces.—How flies and other insects are able to walk up, or run with the body inverted, on hard surfaces has been lately discovered by Dewitz, Dahl, and others. All authors are agreed that this power is due to the presence of the specialized empodium of each tarsus.

Dewitz confirmed the opinion of Blackwell, that a glutinous liquid is exuded from the apices of the tenent hairs which fringe the empodium. By fastening insects feet uppermost on the under side of a covering glass which projects from a glass slide, the hairs which clothe the empodia of the foot of a fly (Musca erythrocephala) may be seen to be tipped with drops of transparent liquid. On the leg being drawn back from the glass, a transparent thread is drawn out, and drops are found to be left on the glass. In cases where these hairs are wanting, as in the Hemiptera, the adhesive fluid exudes directly from pores in the foot. In the beetles (Telephorus dispar) and other insects the tenent hairs on the foot end in sharp points, below which are placed the openings of the canals. The glands, Dewitz states, are chiefly flask-shaped and unicellular, situated in the hypodermis of the chitinous coat; each gland opening into one of the hairs (Fig. 108); they are each invested by a structureless tunica propria, and contain granular protoplasm, a nucleus placed at the inner side, and a vesicle, prolonged into a tube which, traversing the neck of the gland, is attached to the root of the hair; the vesicle receiving the secretion. Each gland is connected with a fine nerve-twig, and secretion is probably voluntary. Among the tenent hairs of the empodium are others which must be supplied with a nerve, forming tactile hairs, as they each proceed from a unicellular ganglion (Fig. 108, n″). The secretion is forced out of the gland by the contraction of the protoplasm, Dewitz having seen the secretion driven out from the internal vesicle into its neck.

Fig. 128.—The walk of an orthopterous insect: series to be followed from right to left.—After Marey.

Fig. 129.—Beetle walking: series to be followed from left to right.—After Marey.

Fig. 130.—A, end of an adhesive hair of a weevil (Eupolus): i′, canal: i‴, its external opening at the end of the hair. B, end of a similar hair of Telephorus with drops of the secretion.—After Dewitz.

In the spherical last tarsal joint of Orthoptera (Fig. 109), which is without these tenent hairs, nearly all the cells of the hypodermis are converted into unicellular glands, each of which sends out a long, fine, chitinous tubule, which is connected with its fellows by very fine hairs and is continuous with the chitinous coat of the foot and opens through it. The sole of the foot is elastic and adapts itself to minute inequalities of surfaces, while the anterior of each tarsal joint is almost entirely occupied by an enlargement of the trachea, which acts on the elastic sole like an air chamber, rendering it tense and at the same time pliant. Dewitz adds that the apparatus situated on the front legs of the male of Stenobothrus sibiricus (Fig. 131) must have the function of causing the legs to adhere closely to the female by the excretion of an adhesive material. The hairs of the anterior tarsi of male Carabi also appear to possess the power of adhesion. In the house-fly the empodia seem to be only called into action when the insect has to walk on vertical smooth surfaces, as at other times they hang loosely down.

Burmeister observed the use of a glutinous secretion for walking in dipterous larvæ, and Dewitz found that the larva of a Musca used for this purpose a liquid ejected from the mouth. The larvæ of another fly (Leucopis puncticornis) perform their loop-like walk by emitting a fluid from both mouth and anus. A Cecidomyia larva is able to leap by fixing its anterior end by means of an adhesive fluid. The larva of the leaf-beetle, Galeruca, moves by drawing up its hinder end, fixing it thus, and carrying the anterior part of the body forward with its feet until fully extended, when it breaks the glutinous adhesion. The abdominal legs of some saw-fly larvæ have the same power.

Dahl could not detect in the foot of the hornet (Vespa crabro) any space which could be considered as a vacuum.

Fig. 131.—Stenobothrus sibiricus pairing: A. the ♂, fore tarsus (t) greatly enlarged; ar, arolia; p, pulvillus.—After Pagenstecher.

Simmermacher states that in most cases of climbing beetles the tubular tenent hairs pour out a secretion (Figs. 133, 134), “and it is probable that we have here to do with the phenomena not of actual attachment by, as it were, gluing, but of adhesion; the orifice of the tubes is divided obliquely, and the tubes are, at this point, extremely delicate and flexible, so as to adhere by their lower surface; in this adhesion they are aided by the secreted fluid.” In the case of the Diptera he does not accept the theory by which the movement of the fly along smooth surfaces is ascribed to an alternate fixation and separation, but believes in a process of adhesion, aided by a secretion, as in many Coleoptera. (In the Cerambycidæ there is no secretion, and the tubules are merely sucking organs, like those observed in the male Silphidæ.) “The attaching lobes, closely beset with chitinous hairs, are enabled, in consequence of the pressure of the foot, to completely lie along any smooth surface; this expels the air beneath the lobes, which are then acted on by the pressure of the outer air.” (Journ. Roy. Micr. Soc., 1884, p. 736.) Another writer (Rombouts) thinks this power is due to capillary adhesion.

Fig. 132.—Fore leg of ♂ Dyticus, under side, with sucker, formed of 3 enlarged tarsal joints: with a small cupule highly magnified. × 120.—After Miall.

The action of the pulvillus and claws when at rest or in use by the honey-bee is well shown by Cheshire (Fig. 135, B). In ascending a rough surface, “the points of the claws catch (as at B) and the pulvillus is saved from any contact, but if the surface be smooth, so that the claws get no grip, they slide back and are drawn beneath the foot (as at A), which change of position applies the pulvillus, so that it immediately clings. It is the character of the surface, then, and not the will of the bee, that determines whether claw or pulvillus shall be used in sustaining it. But another contrivance, equally beautiful, remains to be noticed. The pulvillus is carried folded in the middle (as at C, Fig. 105), but opens out when applied to a surface; for it has at its upper part an elastic and curved rod (cr, Figs. 105 and 135), which straightens as the pulvillus is pressed down; C and D, Fig. 135, making this clear. The flattened-out pulvillus thus holds strongly while pulled, by the weight of the bee, along the surface, to which it adheres, but comes up at once if lifted and rolled off from its opposite sides, just as we should pull a wet postage stamp from an envelope. The bee, then, is held securely till it attempts to lift the leg, when it is freed at once; and, by this exquisite yet simple plan, it can fix and release each foot at least twenty times per second.” (Bees and Bee-keeping, p. 127.)

Fig. 133.—Cross-section through a tarsal joint of fore leg of Dyticus, ♂, showing the stalked chitinous suckers (s), with a marginal bristle on each side: t, trachea; a, an isolated tubule or sucker of Loricera,—b, of Chlænius,—c, of Cicindela; d, two views of one of Necrophorus germanicus, ♂.

Fig. 134.—Section through the tarsus of a Staphylinid beetle; the glandular or tenent hairs arising from chitinous processes. A, section through the tarsal joint of the pine weevil, Hylobius abietis, showing the crowded, bulbous, glandular, or tenent hairs arising from unicellular glands.—This and Fig. 133 after Simmermacher.

Ockler divides the normal two-clawed foot into three subtypes: (1) with an unpaired median empodium; (2) with two outer lateral adhesive lobes; (3) with two adhesive lobes below the claws; the latter is the chief type and forms either a climbing or a clasping foot. The amount of movement possessed by the claws is limited, and what there is, is effected by means of an elastic membrane and the extensor plate (Fig. 110). The “extensor sole” which is always present in insects with an unpaired median fixing or adhesive organ (empodium) is to be regarded as a modification of the extensor seta. The extensor plate is peculiar to an insect’s foot. Ockler states that the so-called “pressure plate” of Dahl is only a movably articulated, skeletal, supporting plate for the median fixing lobule.

Fig. 135.—Honey-bee’s foot in the act of climbing, showing the automatic action of the pulvillus, × 30: A, position of foot in climbing on a slippery surface, or glass; pv, pulvillus; fh, tactile hairs; un, unguis; t, last tarsal joint. B, position of foot in climbing rough surface. C, section of pulvillus just touching flat surface; cr, curved rod. D, the same applied to the surface.—After Cheshire.

Climbing.—In certain respects the power of climbing supplies the want of wings, and even exists often in house-flies among which there is shown a many-sided motion that is quite unheard of in other groups of insects.

The best climbers are obviously those insects which live on trees and bushes, as, for example, longicorn beetles and grasshoppers. These may be accurately called the monkeys of the insect kind, even if their movements take place less gracefully, and indeed rather stiffly and woodenly. We already know what are the proper climbing organs; that is, the sharp easily movable claws on the foot. With the help of these claws certain insects, May-beetles for example, can hang upon one another like a chain; indeed, bees and ants in this manner bind themselves together into living garlands and bridges. There are still added to the chitinous hooks flaps and balls of a sticky nature, by help of which likewise the insects glue themselves together. To facilitate the spanning of still thicker twigs, the climbing foot of insects has a greater movability even than when it only serves as a sole. (Graber.)

The mode of swimming of insects.—To study the swimming movements of insects, let us examine a Dyticus. It will appear, as Graber states, to be wonderfully adapted to its element.

“The body resembles a boat. There is nowhere a projecting point or a sharp corner which would offer unnecessary resistance to motion; bulging out in the middle and pointed at the end, it cuts through the resistance of the water like a wedge. The movable parts, the oars, seem to be as well fitted for their purpose as the burden to be moved by them. That the hind legs must bear the brunt of this follows from their position exactly in the middle of the body, where it is widest. In other insects also these legs are used for the same purpose as soon as the insects are put in the water. But the swimming legs of water-beetles are oars of quite peculiar construction. They are not turned about in the coxæ, as are other legs, but at the foot-joint. The coxa, namely, has grown entirely together with the thoracic partition. The muscles we have mentioned, exceeding in strength all the soft parts taken together, take hold directly of the large wing-shaped tendons of the upper thigh, and extend and retract the leg in one of the planes lying close to the abdominal partition. The foot forms the oar, however. It is very much lengthened and still more widened, and can be turned and bent in by separate muscles in such a way that in the passive movement, that is, the retraction, the narrow edge is turned to the fore, and therefore to the medium to be dislodged; however, as soon as the active push is to be performed and the leg is extended with greater force, it cuts down through the water with its whole width. These effective oar-blades are still considerably enlarged by the hairs arising on the side of the foot, which spread out at the decisive moment.

“Every one knows that the oar-blades of swimming beetles always go up and down simultaneously and in regular time. On the other hand, as soon as one puts a Dyticus on the dry land, i.e. on an unyielding medium, it uses its hind legs entirely after the manner of other land insects; that is, they are drawn in and extended again alternately, as takes place clearly enough from the footsteps in Fig. 119, A. We learn from this that water insects have not yet, from want of practice, forgotten the mode of walking of land insects.

“The forcing up of the water as a propelling power is added to the repulsion produced by the strong strokes of the oars. If the beetle stood up horizontally in the water, he would be lifted up.

“As the trunk, however, assumes an oblique position when the insect wishes to swim, one can then imagine the driving up of the water as being divided into two forces, one of which drives the body forward in a horizontal direction, while the other, that is, the vertical component, is supplied by the moving of the oars. The swimming insect is thus, as it were, a snake flying in the water.

“The long streamer-like hind legs of many water-bugs, for example Notonecta, approach more nearly our artificial oars. These legs are turned out from the bottom.

“There is no doubt but that the legs of insects, as regards the many-sidedness and exactitude of their locomotive actions, place the similar contrivances of other animals far in the shade. We shall be forced to admire these ingenious levers still more, however, when we take into consideration their energy and strength. That the force with which the locomotive muscles of insects is drawn together is enormous compared with that of vertebrates, we may learn if we try to subdue the rhythmical movements of the thorax of a large butterfly by the pressure of our finger or to open against the insect’s will the closed jumping leg of a grasshopper, or the fossorial shovel of a mole-cricket.”