/-I c ' -D i-n eO i-=) D D , CD : m a > i . TRANSFORMATIONS '.-, 3 GUIDE TO THE STUDY OF INSECTS, AND A TREATISE ON THOSE INJUEIOUS AND BENEFICIAL TO CROPS: FOR THE USE OF COLLEGES, FARM-SCHOOLS, AND AGRICULTURISTS. BY A. S. PACKARD, JR., M.D. \V1TH FIFTEEN PLATES AND SIX HUNDRED AND SEVENTY WOOD-CUTS. THIRD EDITION. SALEM : NATURALISTS' AGENCY. LONDON : TJJUBNER & CO. TASSEL: THEODOR FISCHER. 1872. Entered according to Act of Congress, in the year 18G9, by A. S. PACKARD, JR., in the Clerk's Office of the District Court of the District of Massachusetts. Printed at the SALEM PRESS. F. W. Putnam & Co., Salem, Ma?.- . ^ -. PREFACE TO THE THIRD EDITION. SEVERAL important changes, and many slight altera- tions, have been made in this edition. The author would especially call attention to the change in his views as to the number of segments in the head of hexapodous in- sects ; there being four instead of seven (see p. 19). On p. 52 will be found a notice of parthenogenesis in the pupa of Chironomus, and on p. Ill an account of the recently discovered glands for secreting wax ; while on p. 258 correct figures of the larva and pupa of Melitcea Harrisii are given, and a brief account of the Linguatulina is added. It is hoped that these changes, with the ad- dition of four more plates, and an appendix, will bring the work up to the present state of the science. The author should here state, in justice to himself, that the primary object in preparing the systematic portion of the work was to give as clear a view as possible of the larger groups of insects ; so that the groupings of the families into subdivisions of suborders has been omitted for the sake of perspicuity. Thus, the difference between the Heteropterous and Hemipterous divisions of the Hemiptera is not perhaps so clearly indicated as may seem desirable ; so also, the difference of the Tenthre- dinidas and Uroceridse from the rest of the Hymenoptera, of the Lice from the rest of the Hemiptera, or the Lepis- matidtie, Campodese and Poduridse, from the remainder of the Neuroptera. Perhaps in endeavoring to bring out clearly the essential unity of organization in the members IV PREFACE. of the larger groups, sufficient justice has not been paid to the frequent diversity observable. Certain small and unimportant families have also been omitted ; it is believed, without detriment to a work of this scope. Most authors regard the Hymeuoptera and equivalent groups as "orders" rather than "suborders." When the reader prefers, he might alter to suit his views. It is not improbable that the Hexapoda, Arachnida and My- riapoda are subclasses; hence, the Hymeuoptera, etc., may be considered as orders, and then, for example, the Hemiptera, Heteroptera and Lice (Pediculina and Mallo- phaga) might be regarded as suborders of the grand group Hemiptera. It matters little to the author, so long as the fact (or what he believes to be the fact) be recog- nized, that the Hexapods, Arachnids and Myriapods are subdivisions of a class, and not separate classes equivalent each to the Crustacea, for example. Salem, March, 1872. X, PREFACE . THIS introduction to the study of insects is designed to teach the beginner the elements of entomology, and to serve as a guide to the more elaborate treatises and memoirs which the advanced student may wish to consult. Should the book, imperfect as the author feels it to be, prove of some service in inducing others to study this most interesting and useful branch of natural history, the object of the writer will have been fully attained. In order to make it of value to farmers and gardeners, whose needs the writer has kept in view, and that it may be used as a text book in our agricultural colleges, concise ac- counts have been given of insects injurious or beneficial to vegetation, or those in any way affecting human interests. When the localities of the insects are not precisely given, it is to be understood that they occur in the Eastern Atlantic States from Maine to Pennsylvania, and the more northern of the Western States. When the family names occur in the text they are put in spaced Italics, to distinguish them from the generic and specific names which are Italicized in the usual way. The succession of the suborders of the hexapodous insects is that proposed by the author in 1863, and the attention of zoologists is called to their division into two series of sub- orders, which are characterized on page 104. To the first and highest may be applied Leach's term METABOLIA, as they all agree in having a perfect metamorphosis ; for the second and lower series the term HETEROMETABOLIA is pro- iv PREFACE. posed, as the four suborders comprised in it differ in the degrees of completeness of their metamorphoses, and are all linked together by the structural features enumerated on page 104. The classification of the ITymenoptera is original with the author, the bees (Apiclae) being placed highest, and the saw- flies and Urocericlae lowest. The succession of the families of the Lepidoptera is that now generally agreed upon by en- tomologists. Loew's classification of the Diptcra, published in the "Miscellaneous Collections" of the Smithsonian Institution, has been followed, with some modifications. Ilaliday's suggestion that the Pulicidoe are allied to the Mycetophilidne gives a clue to their position in nature among the higher Diptera. Leconte's classification of the Coleoptera is adopted as far as published by him, ?'. - tem; c is the dorsal vessel, or heart, from the side of \vlii !). in each ring, a small vessel is sent downwards and around to <, the ventral vessel. (h-ii/i'iirtl. FIG. 4. An ideal section of a Bee. Here the crust is dense and thick, to which strong muscles are attached. On the upper side of the ring the wings grow out, while the legs are inserted near the under side. The trachea? (i!> enter through the stif/nut, or breathing pore, situated just under the wing, and their brandies sub- divide and are distributed to the wings, with their live principal veins as iudic::to.l THE CLASS OF INSECTS. their young stages, from the time of their exclusion from the egg, until they pass into mature life. A more careful study of this period than we arc now able to enter upon would show us how much alike the young of all articulates are at first, and how soon they begin to differ, and assume the shape characteristic of their class. Most Worms, after leaving the egg, are at first like some infusoria, being little sac-like animalcules, often ciliated over nearly the entire surface of the infinitesimal body. Soon this sac-like body grows longer, and con- tracts at intervals ; the intervening parts become unequally enlarged, some segments, or rings, formed by the contraction of the bod^y-walls, greatly exceeding in size those next to them ; and it thus assumes the appearance of being more or less equally ringed, as in the young TerebeUa (Fig. 5), where the ciliae are restricted to a single circle surrounding o o the body. Gradually (Fig. G) the cilice disap- ' e pear and regular locomotive organs, consisting of minute paddles, grow out from each side ; feelers (antennae), jaws, and eyes (simple rudi- mentary eyes) appear on the few front rings of the body, which are grouped by themselves into a sort of head, though it is difficult, in a large proportion of the lower worms, for un- skilled observers to distinguish the head from the tail. Thus we see throughout the growth of the worm, no attempt at subdividing the body into regions, each endowed with its peculiar d functions ; but only a more perfect system of rings, each relatively veiy equally developed, in the figure, also to the dorsal vessel (f), the intestine (6), and the nervous cord (a], The tracheae and a nervous filament are also sent into the legs and to the wings. The tracheae are also distributed to the dorsal vessel and intestine by numerous branches which serve to hold them in place. Original. FIG. o. Young TerebeUa, soon after leaving the egg. From A. Ayassiz. FIG. (i represents the embryo of a worm (Autolytus cornutus) at a later stage of growth, a is the middle tentacle of the head ; c, one of the posterior tentacles; b, the two eye-spots at the base of the hinder pair of feelers ; c is one of a row of oar-like organs (cirri) at the base of which are inserted the locomotive bristles, THE CLASS OF INSECTS. 5 but all becoming respectively more complicated. For example, in the Earth-worm (Lumbricus) , each ring is distinguishable into an upper and under side, and in addition to these a well- marked side-area, to which, as for example in marine worms (e. g. Nereis) , oar-like organs are attached. In most worms eye-spots appear on the front rings, and slender tentacles grow out, and a pair of nerve-knots (ganglia) are apportioned to each ring. In the Crustaceans, such as the fresh-water Crawfish (Asta- CMS), as shown by the German naturalist Kathke ; and also in the earliest stages of the Insect, the body at once assumes a worm-like form, thus beginning its embryonic life from the goal reached by the adult worm. The young of all Crustaceans (Fig. 7) first begin life in the egg as oblong flattened worm-like bodies, each end of the body being alike. The .young of the lower Crustaceans, such as the Barnacles, and some marine forms (Copepoda), and some loAvly organized parasitic species inhabiting the gills of fishes, are hatched as microscopic embryos which would readily be mistaken for young mites (Acarina). In the higher Crus- taceans, such as the fresh-water Crawfish, the young, when hatched, does not greatly differ from the parent, as it has passed through the worm-like stage within the egg. Fig. 7 represents the 3'oung of the fresh- water Lobster (Crawfish) before leaving the egg. The body is divided into rings, ending in lobes on the sides, which are the rudiments of the limbs, b is the rudiment of the 63-6- Fig. 7. stalk, at the end of which is the e}-e ; a is the fore antennae ; c is the hind antennas ; d is one of the maxilla-feet ; e is the first pair of true feet destined in the adult to form the large "claw." Thus the eye-stalks, antenna?, claws, and legs are moulded upon a common form, and at first are scarcely distin- with the cirri serving as swimming and locomotive organs; d, the caudal styles, or tail-feelers. In this figure we see how slight are the differences between the feelers of the head, the oar-like swimming organs, and the caudal filaments; we can easily see that they are but modifications of a common form, and all arise from the common limb-bearing region of the body. The alimentary canal, with the proven triculus, or anterior division of the stomach, occupies the middle of the body; while the mouth opens on the under side of the head. From A. Agassi,. FIG. 7. Embryo of the Crawfish. From Rathke. 1* G THE CLASS OF INSECTS. guishable from each other. Here we see the embryo divided into a head-thorax and a tail. It is the same with Insects. Within the egg at the dawn of life they are flattened oblong bodies curved upon the yelk- mass. Before hatching they become more cylindrical, the limbs bud out on the sides of the rings, the head is clearly demarked, and the young caterpillar soon steps forth from the egg-shell ready armed and equipped for its riotous life. As will be seen in Fig. 8, the legs, jaws, and antennae are first started as buds from the side of the rings, being simply elongations of the body-wall, which bud out, become larger, and finally jointed, until the \x buds arising fro.n the thorax or le abdomen become legs, those from the base of the head be- come jaws, while the antennas and palpi sprout out from the front rings of the head. Thus while the bodies of all articulates Fig. s. are built up from a common em- bryonic form, their appendages, which are so diverse, when we compare a Lobster's claw with an Insect's antenna, or a Spider's .spinneret with the hinder limbs of a Centipede, are yet but modifications of a common form, adapted for the different uses to which they are put Joy these animals. FIG. 8. A Caddis, or Case-fly (Mysttttides) in the egg, with part of the yolk (.;) n,)t yet inclosed within the body-walls, n, antennas; between a and b the mandi- bles; li, maxilla; c, labium; <7, the separate eye-spots (ocelli), which afterwards in- crease greatly in number and unite to form the compound' eye. The "neck" or junction of the head with the thorax is seen at the front part of 'the yolk-mass; e, the three pairs of legs, which are folded once on themselves;/, the pair of anal legs attached to the tenth ring of the abdomen, as seen in caterpillars, which form long antenna-like filament:- in the Cockroach and May-fly, etc. The rings of the body are but partially formed; they are cylindrical, giving the body a worm-like form. Here, as in the other two figures, though not so distinctly seen, the antenna?, jaws, and last pair of abdominal legs are modifications of but a single form, and grow out from the side of the body. The head-appendages are directed forwards, as they are to be adapted for sensory and feeding purposes ; the legs are directed downwards, since they are to support the insect while walking. It appears that the two ends of the body arc perfected before the middle, and the under side before the upper, as we see the yolk-mass is not yet inclosed and the rings not yet formed above. Thus all articulates differ from all vertebrates in having the yolk-mass situated on the back, instead of on the belly, as in the chick, dog, or human em- bryo. From Zaddach. THE CLASS OF INSECTS. 7 The Worm is long and slender, composed of an irregular number of rings, all of very even size. Thus, while the size of the rings is fixed, their number is indeterminate, varying from twenty to two hundred or more. The outline of the body is a single cylindrical figure. The organs of locomotion are fleshy filaments and hairs (Fig. 3, /) appended to the sides. In one of the low intestinal worms, the Tape-worm (Tcenia), each ring, behind the head and "neck," is provided with organs of reproduction, so that when the body becomes broken up into its constituent elements, or rings (as often occurs naturally in these low forms for the more ready propagation of the species, since the 3 r oung are exposed to many dangers while living in the intestines of animals), they become living inde- pendent beings which "move freely and somewhat quickly like Leaches," and until their real nature was known they were thought to be worms. This and other facts prove, that, in the Worm, the vitality of the animal is very equally dis- tributed to each ring. If we cut off the head or tail of some of the low worms, such as the Flat Worms (Planar ia, etc.), each piece will become a distinct animal, but an Insect or Crab sooner or later dies when deprived of its head or tail (abdomen). Thus, in the Worm the vital force is very equally distributed to each zoological element, or ring of the body ; no single part of the body is much honored above the rest, so as to sub- ordinate and hold the other parts in subservience to its peculiar and higher ends in the animal economy. The Crustacean, of which the Shrimp (Fig. 9) is a typical example, is com- posed of a determinate number (21) of rings which Kg. 9. are gathered into two regions ; the head-thorax (cephalo- thorax) and hind -body, or abdomen. In this class there is a broad distinction between the anterior and posterior ends of the bod} r . The rings are now grouped into two regions, and the hinder division is subordinate in its structure and FIG. 9. A Shrimp. Pandalus annulicornis. a, cephalothorax ; b, abdomen. 8 ' THE CLASS OF INSECTS. uses to the forward portion of the body. Hence the nervous power is transferred in some degree towards the head ; the cephalothorax containing the nervous centres from which nerves are distributed to the abdomen. Nearly all the organs perform- ing the functions of locomotion and sensation reside in the front region ; while the vegetative functions, or those concerned in the reproduction and nourishment of the animal, are mostly carried on in the hinder region of the body (the abdomen). The typical Crustacean cannot be said to have a true head, in distinction from a thorax bearing the organs of locomotion, but rather a group of rings, to which are appended the organs of sensation and locomotion. Hence we find the appendages of this region gradually changing from antennae and jaws to foot-jaws, or limbs capable of eating and also of locomotion ; they shade into each other as seen in Fig. 9. Sometimes the jaws become remarkably like claws ; or the legs resemble jaws at the base, but towards their tips become claw-like ; gill-like bodies are sometimes attached to the foot-jaws, and thus, as stated by Professor J. D. Dana in the introduction to his great work on the Crustacea of the United States Exploring Expedi- tion, the typical Crustaceans do not have a distinct, head, but rat.her a "head-thorax" (cephalothorax). When we rise a third and last step into the world of Insects, we see a completion and final development of the articu- late plan which has been but obscurely hinted at in the two lowest classes, the Worms and Crustaceans. Here we first meet with a true head, separate in its structure and functions from the thorax, which, in its turn, is clearly distinguishable from the third region of the body, the abdomen, or hind-body. These three regions, as seen in the Wasp (Fig. 10), are each provided with three distinct sets of organs, each having distinct functions, though all are governed by and minister to the brain force, now in a great measure gathered up from the Fig. 10. posterior rings of the body, and in a more concentrated form (the brain being larger than in the lower articulates) lodged in the head. Here, then, is a centralization of parts headwards ; they are FIG. 10. Phllanthus ventilabris Fabr. A Wood-wasp. From Say. COMPOSITION OF THE INSECT-CRUST. 9 brought as if towards a focus, and that focus the head, which is the meaning of the term " capitalization," proposed by Pro- fessor Dana.* Riny distinctions have given way to regional distinctions. The former characterize the Worm, the latter the Insect. In other words, the division of the body into three parts, or regions, is in the insect, on the whole, better marked than the division of any one of those parts, except the abdo- men, into rings. COMPOSITION OF THE INSECT-CRUST. Before describing the composition of the body-wall, or crust, of the Insect, let us briefly review the mode in which the same parts are formed in the lower classes, the Worms and Crustaceans. We have seen that the typical ring, or segment (called by authors zoonule, zoonite, or somite, meaning parts of a body, though we prefer the term arthromere, denoting the elemental part of a jointed or articulate animal), consists of an upper (tergite), a side (pleurite), and an under piece (sternite). This is seen in its greatest simplicity in the Worm (Fig. 2), where the upper and ventral arcs are separated by the pleural region. In the Crus- tacean the parts, hardened by the deposition of chitine and therefore thick and unyielding, have to be farther subdivided to secure the necessary amount of freedom of motion to the body and legs. The upper arc not only covers the back of the ani- mal, but extends down the sides ; the legs are jointed to the epimera, or flanks, on the lower arc ; the episternnm is situated between the epimerum and sternum ; and the sternum, form- ing the breast, is situated between the legs. In the adult, there- fore, each elemental ring is composed of six pieces. It should, however, be borne in mind that the tergum and ster- * In two papers on the Classification of Animals, published in the American Journal of Science and Arts, Second Series, vol. xxxv, p. 65, vol. xxxvi, July, 18U3, and also in his earlier paper on Crustaceans, "the principle of cephalization is shown to be exhibited among animals in the following ways : 1. By a transfer of members from the locomotive to the cephalic series. 2. By the anterior of ttfe locomotive organs participating to some extent in ce- phalic functions. 3. By increased abbreviation, concentration, compactness, and perfection of structure, in the parts and organs of the anterior portion of the body. 4. By increased abbreviation, condensation, and perfection of structure in the posterior, or gastric and caudal portion of the body. 5. By an upward rise in the cephalic end of the nervous system. This rise reaches its extreme limit in Man." 10 THE CLASS OF INSECTS. num each consist, in the embryo, of two lateral parts, or halves, which, during development, unite on the median line of the body. Typically, therefore, the crustacean ring consists pri- marily of eight pieces. The same number is found in all insects which are wingless, or in the larva and pupa state ; this applies also to the Myriupods and Spiders. In the Myriopoda, or Centipedes, the broad tergum overlaps the small epimera, while the sternum is much larger than in the Spiders and Insects. In this respect it is like the broad flat under-surface of most worms. Hence the legs of the Centipede are inserted very far apart, and the "breast," or sternum, is not much smaller than the dorsal part of the crust. In the Jul us the dorsal piece (tergum) is greatly developed over the sternum, but this is a departure from what is ap- parently the more typical form of the order, /. e. the Centipede. In the Spiders there is a still greater disproportion in size between the tergum and the sternum, though the latter is very large compared with that of Insects. The epimera and epioterna, or side-pieces of the Spiders, are partially concealed by the over-arching tergum, and they are small, since the joints of the legs are very large, Audouin's law of development in Articu- lates showing that one part of the insect crust is always developed at the expense of the adjoining part. In the Spider we notice that the back of the thorax is a single solid plate consisting originally of four rings consolidated into a single hard piece. In like manner the broad solid sternal plate results from the reunion of the same number of sternites cor- responding, originally, to the number of thoracic legs. Thus the whole upper side of the head and thorax of the Spider is consolidated into a single hard horny immovable plate, like the upper solid part of the cephalothorax of the Crab or Shrimp. Hence the motions of the Spiders are very stiff' com- pared with those of many Insects, and correspond to those of the Crab. The ci'ust of the winged insect is modified for the per- fornrmce of more complex motions. It is subdivided in so different a manner from the two lower orders of the class, that it would almost seem to have nothing in common, structurally speaking, with the groups below them. It is only by examin- COMPOSITION OF THE IXSECT-CliUST. 11 Fig. ]2. ms scm ins" pt-. ing the lowest wingless forms such as the Louse, Flea, Poflnra, and Bark-lice, where we see a transition to the Or- ders of Spiders and Myriopods, that we can perceive the plan pervading all these forms, uniting them into a common class. . A segment of a winged six-footed insect (Ilexapod) consists typically of eight pieces which we will now examine more leisurely. Figure 12 represents a side-view of the thorax of the Telea Polyphemus, or Silk- worm moth, with the legs and wings removed. Each ring consists primarily of the teir/xm, the two side-pieces (epimerum and episternum) and the sternum, or breast-plate. But one of these pieces (sternum) remains simple, as in the lower orders. The tergum is divided into four pieces. They were named by Au- douin going from before backwards, the prcescutum, scutum, scutelliint, and postscutellum. The scutum is invariabty present and forms the larger part of the upper portion (tergum) of the tho- rax ; the scutellum is, as its name indicates, the little shield so promi- nent in the beetle, which is also uniformly present. The other two pieces are usually minute and crowded down out of sight, and placed between the two oppos- ing rings. As seen in Fig. 11, the proescutum of the moth is a small rounded piece, bent vertically down, so as not to be seen from above. In the lowly organized Hejnalus, and some FIG. 11. Tergal view of the middle segment of the thorax of Telea Polyphemus, prm, prrescntum; ms, scutum; scm, scntellum; ptm, postscutellum; pt, patagium, or shoulder tippet, covering the insertion of the wings. Original. FIG. 12. Side view of the thorax of T. Polyphemus, the hairs removed. 1, Pro- thorax ; 2, Mesothorax ; 3, Metathorax, separated by the wider black lines. Tergum of the prothorax not represented, ms, mesoscutum ; scm, mesoscutellum; ms" , metascutnm; scm", metascutellum; pt, a supplementary piece near the inser- tion of patagia; w, pieces situated at the insertion of the wings and surrounded by membrane; em, epimerum of prothorax, the long upright piece above being the episternum; 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", coxse; te, le", le"', trochautiu.es; tr, tr, tr, trochanters. Original. - epm tr te c" tr c'" tr 123 12 THE CLASS OF INSECTS. Neuroptera, such as the Polystcecliotes (Fig. 13 a), the pne- scutum is large, well developed, triangular, and wedged in between the two halves of the scutum. The little J , piece succeeding the scutellum, i. e. the postscu- tellum, is still smaller, and rarely used in descrip- tive entomology. Thus far we have spoken of the 3 j middle, or mesothoracic, ring, where these four pieces are most equally developed. In the first, a1 ' or prothoracic, ring, one part, most probably the scutum, is well developed, while the others are aborted, and it is next to impossible to trace them in most insects. The prothorax in the higher in- sects, such as the Hymenoptera, Lepidoptera, and Diptera is very small, and often intimately soldered to the succeeding or mesothoracic ring. In the lower insects, however, such as the Coleoptera, the bugs (Hemiptera), grasshoppers and their allies (Orthoptera), and the Neuroptera, the large broad pro- thorax consists almost entirely of this single piece, and most writers speak of this part under the name of "thorax," since the two posterior segments are concealed by the wings when the animal is at rest. The metathorax is usually very broad and short. Here we see the scutum split asunder, with the pnescutum and scutellum wedged in between, while the post- scutellum is aborted. On the side are two pieces, the upper (epimerum) placed just beneath the tergum, which is the collective name for the four tergal, or dorsal, pieces enumerated above. In front of the epimerum and resting upon the sternum, as its name im- plies, is the epistermim. These two parts (pleurites) compose the flanks of the elemental ring. To them the legs are articu- lated. Between the two episterna is situated the breast-piece (sternum), which shows a tendency to grow smaller as we ascend from the Neuroptera to the Bees. In those insects provided with wings, the epimera are also subdivided. The smaller pieces, hinging upon each other, as it were, give play to the very numerous muscles of flight FIG. 13. A tergal view of thorax of Ucpinlux (Sthcnnpis) ; 1, prothorax ; 2, meso- thorax ; 3, metathorax. The prothorax is very small compared with that of Poly- stcechutea (13 , 1), where it is nearly as long as broad. Oriyiiwl. COMPOSITION OF THE INSECT-CRUST. 13 needed by the insect to perform its complicated motions while on the wing. The insertion of the fore wing is concealed by the "shoulder tippets," or patagia (Fig. 11), which are only present in the mesothorax. The external opening of the spiracles just under the wing perforates a little piece called by Audouin the peri- treme, A glance at Figures 11 and 12 shows how compactly the various parts of the thorax are agglutinated into a globular mass, and that this is due to the diminished size of the first and third rings, while the middle ring is greatly enlarged to support the muscles of flight. There are four tergal, four pleural, two on each side (and these in the Hymenoptera, Lepi- doptera, and Diptera subdivide into several pieces), and a single sternal piece, making nine for each ring and twenty- seven for the whole thorax, with eight accessory pieces (the three pairs of peritremes and the two patagia), making a total of thirty-five for the entire thorax ; or, multiplying the four tergal pieces by two, since they are formed by the union of two primitive pieces on the median line of the body, we have thirty-nine pieces composing the thorax. TABLE OF THE PARTS OF THE THORAX APPLIED TO THE PRO-, MESO-, AND METATIIORAX, RESPECTIVELY. , Pnescutum, fDorsnl S Scutum, I Surface J Scutcllum. - Postscutellum. Thorax P.c,,,,, i w 1 1 vf n r* o ' ( Episterual apophysis, Stigma, Peritreme. We must remember that these pieces are rarely of precisely the same form in any two species, and that they differ, often in a very marked way, in different genera of insects. How sim- ple, then, is the typical ring, and how complex are the va- rious subdivisions of that ring as seen in the actual, living insect, where each part has its appropriate muscles, nerves, and tracheae ! We have seen how the thorax is formed in Insects generally, let us now advert to the two types of thorax in the six-footed 14 THE CLASS OF INSECTS. insects. In the higher series of suborders, comprising the Dip- tera, Lepicloptera and Hymenoptera, placing the highest last, the thorax shows a tendency to assume a globular shape ; the upper side, or tergum, is much arched, the pleural region bulges out full and round, while the legs conceal at their insertion the sternum which is minute in size. In the lower series, embracing the Coleoptera, Hemiptera, Orthoptera, and Neuroptcra, the entire body tends to be more flattened ; in the thorax the tergum is broad, especially that of the prothorax, while the pleurites (episterna and epimera) are short and bulge out less than in the higher series, and the ster- num is almost invariably well developed, often presenting a large thick breast-plate bearing a stout spine or thick tubercle, as in (Edipoda. We can use these characters, in classifying insects into suborders, as they are common to the whole order. Hence the use of characters drawn from the wings and mouth- parts (which are sometimes wanting), leads to artificial dis- tinctions, as they are 2wipheral organs, though often convenient in our first attempts at classifying and limiting natural groups. The abdomen. In the hind body, or third region of the trunk, the three divisions of the typical ring (arthromere), are entire, the tergum is broad and often not much greater in ex- tent than the sternum ; and the pleurites also form either a single piece, or, divided into an epimerum and episternum, form a distinct lateral region, on which the stigmata are sit- uated. The segments of the abdomen have receiA'ed from Lacaze-Duthiers a still more special name, that of wite, and the different tergal pieces belonging to the several rings, but especially those that have been modified to form the genital armor have been designated by him as tergites. We have applied this last term to the tergal pieces generally. The typi- cal number of abdominal segments is eleven. In the lowest insects, the Neuroptera, there are usually eleven ; as we have counted them in the abdomen of the embryo of Diplax. In others, such as the Hymenoptera and Lepidoptera, there may never be more than ten, so far as present observation teaches us. The formation of the sting, and of the male intromittent organ, may be observed in the full-grown larva and in the in- COMPOSITION OF THE OVIPOSITOR. 15 complete pupa of the Humble-bee, and other thin-skinned Hvmenopterous larvae, and in a less satisfactory way in the young Dragon-flies. If the larva of the Humble-bee be taken just after it has become full-/ed, and as it is about to enter upon the pupa state, the elements (sterno - rhah- clites Lacaze- Dnthiers), or tubercles, destined to rig. ic. form the ovipositor, lie in separate pairs, in two groups, Fig. 15. exposed distinctly to view, The ovipositor thus consists of three Fig. 17. 11 a. Fig. 14. as in Figures 14-18. pairs of slender non-articulated tubercles, situated in juxta- position on each side of the mesial line of the body. The first pair arises from the eighth abdominal ring, and the second and third pair grow out from the ninth ring. The ends of the first pair scarcely reach beyond the base of the third pair. "With the growth of the semi-pupa, the end of the abdomen decreases in size, and is FIG. 14. Rudiments of the sting, or ovipositor, of the Humble-bee. 8, 9, 10, sternites of eighth, ninth, and tenth abdominal rings in the larva, a, first pair, situ- ated on the eighth sternite; 6, second and inner pair; and c, the outer pair. The let- tering is the same in figures 14-22. The inner pair (fc), forms the true ovipositor, through which the eggs are supposed to pass when laid by the insect, the two outer pairs, a and c, sheathing the inner pair. Ganin shows that in the embryo of Folyne7>ia(Fig.G;>5), the three pairs of tubercles arise from the 7th, 8th and 9th seg- ments respectively. FIG. 15, 1G. The same a little farther advanced. FIG. 17. The three pairs now appear as if together growing from the base of the ninth segment; 17, side view of the same, showing the end of the abdomen grow- ing smaller through the diminution in size of the under side of the body. FIG. 18. The three pairs of rhabdites now nearly equal in size, and nearly ready to unite and form a tube; 18 a, side view of the same; the end of the abdo- men still more pointed ; the ovipositor is situated between the seventh and tenth rings, and is partially retracted within the body. 16 THE CLASS OF INSECTS. gradually incurved toward the base (Fig. 18), and the three pairs of rhabdites approach each other so closely that the two outer ones completely ensheath the inner, until a complete extensible tube is formed, which is gradually withdrawn entire!}' within the body. The male genital organ is originally composed of three pairs (two pairs, apparently, in u?Es- 1 chna, Fig. 19) of tubercles all arising from the ninth abdominal ring, being sternal outgrowths and placed on each side of the mesial line of the body, two be- Fi s- 20. ing anterior, and very unequal in size, and the Fig. 19. third pair nearer the base of the abdomen. The ex- ternal genital organs are to be considered as probably homologous with the limbs, as Ganin has shown that they bud out in the same manner from (sec p. 704 fig. 655) the arthromere.* ~ b This view will apply to the Fi-. 21. genital armor of all Insects, so far as we have been able to observe. It is so in the pupa of ^Eschna (Fig. 21), and the pupa of Agrion (Fig. 22), which com- pletely repeats, in its essential features, the structure of the ovipositor of Bombus. Thus in jtEsclma and Agrion the ovipositor consists of a pair of closely appressed ensi- form processes which grow out from under the posterior edge of the eighth abdominal ring, and are embraced between two pairs *This term is proposed as better defining the ideal ring, or primary zoological element of an articulated animal than the terms somite or zoonite, which seem too vague; we also propose the term arthroderm for the outer crust, or body walls, of Articulates, and arthropleura for the pleural, or limb-bearing region, of the body, being that portion of the arthromere situated between the tergite and sternite. FIG. 19. The rudiments of the male intromittent organ of the pupa of ^Eschna, consisting of two flattened tubercles situated on the ninth ring; the outer pair large and rounded inclosing the smaller linear oval pair. FIG. 20. The same in the Humble-bee, but consisting of three pairs of tubercles, a, y, z ; 8, 9, 10, the last three segments of the abdomen. FIG. 21. The rudimentary ovipositor of the pupa of ^Eschna, a Dragon-fly. FIG. 22. The same in pupa of Agrion, a small Dragon-fly. Here the rudiments of the eleventh abdominal ring are seen, tl, the base of one of the abdominal false gills. The ovipositor of Cicada is formed in the same way. figs. 11-22 original. 22 . COMPOSITION OF THE OVIPOSITOK. 17 of thin lamelliform pieces of similar form and structure, arising from the sternite of the ninth ring. These outgrowths appar- ently also honaologize with the filiform, antennae-like, jointed appendages of the eleventh ring, as seen in the Perlidoe and most Neuroptera and Orthoptera (especially in Mantis tes- sellata where they (Fig. 23) closely c^? e * resemble antennae), which, arising as ^Ggga^gftv^ "A _ they do from the arthropleural, or limb- [^ _( ^ J~ bearing region of the body, i. e, between rig. 23. the sternum and episternum, are strictly homologous with the abdominal legs of the Myriapoda, the "false legs" of cater- pillars, and the abdominal legs of some Neuropterous larva? (CorydaMs, Phryganeidce, etc.). It will thus be seen that the attenuated form of the tip is produced by the decrease in size of certain parts, the actual disappearance of others, and the perfection of those parts to be of future use. Thus towards the extremity of the body the pleurites are absorbed and disappear, the tergites OA T erlap on the sternites, and the latter diminish in size and are withdrawn within the body, while the last, or eleventh sternite, entirely disappears.* Meanwhile the sting grows larger and larger, until finally we have the neatly fashioned abdominal tip of the bee concealing the complex sting with its intricate system of visceral ves- sels and glands. The ovipositor, or sting, of all insects, therefore, is formed on a common plan (Fig. 24). The solid elements of the arthro- *In Ranatra, however, Lacaze-Duthiers has noticed the curious fact that in order to ibnn the long respiratory tube of this insect, the tergite and sternite of the pregenital (eighth) segment are aborted, Avhile the pleiirites are enormously en- larged and elongated, so as to carry the stigmata far out to the end of the long tube thus formed. FIG. 23. End of the abdomen of Mantis tesseUatn ; p, many-jointed anal style resembling an antenna. 5-11, the last seven abdominal segments; the 8-1 1th ster- nites being obsolete. From Lacaze-Duthiers. FIG. 24. Ideal plan of the structure of the ovipositor in the adult insect. 'i-"t, the tergites, connected by dotted lines with their corresponding sternites. b, the eighth tergite, or anal scale; c, epimerum; ', , two pieces forming the outer pair of rhabdites; i, the second pair, or stylets; and /, the inner pair, or sting; d, the 31- 18 THE CLASS OF INSECTS. mere are modified to form the parts supporting the sting alone. The external opening of the oviduct is always situated between the eighth and ninth segments, while the anal opening lies at the end of the eleventh ring. 80 that there are really, as Lacaze-Duthiers observes, three segments interposed between the genital and anal openings. The various modifications of the ovipositor and male organ will be noticed under the different suborders. THE STRUCTURE OF THE HEAD. After studying the com- position of the thorax and abdomen, where the constituent parts of the elemental ring occur in their greatest simplicity, we may attempt to unravel the intricate structure of the head. We are to determine whether it is composed of one, or more, segments, and if several, to ascertain how many, and then to learn what parts of the typical arthromere are most largely developed as compared with the development of similar parts in the thorax or abdomen. In this, perhaps the most difficult problem the entomologist has to deal with, the study of the .head of the adult insect alone is only guesswork. We must trace its growth in the embryo. Though many writers consider the head as consisting of but a single segment, the most emi- nent entomologists have agreed that the head of insects is com- posed of two or more segments. Savigny led the way to these discoveries in transcendental entomology by stating that the appendages of the head are but modified limbs, and homol- ogous with the legs. This view at once gave a clue to the complicated structure of the head. If the antennas and biting organs are modified limbs, then there must be an elemental segment present in some form, however slightly developed in the mature insect, to which such limbs are attached. But the best observers have differed as to the supposed number of such theoretical segments. Burmeister believed that there were two only ; Cams and Audouin thought there were three ; McLeay and Newman four, and Straus-Durckheim recognized seven. From the study of the semipupa of the Humble-bee (Bombus) support of the sting; e, the support of the stylet (i). R, the anus : O, the outlet of the oviduct. The seventh, eighth, and ninth sternites are aborted. From Lacaze,- Duthiers. THE STRUCTURE OF THE HEAD. 19 and several low Neuropterous forms, as the larva of Ephemera, but chiefly the embryos of Diplax, CJirysopa, Attddbus, Nema- tus, and Pulex, we have concluded that there are four such ele- mental segments in the head of hexapodous insects. On reference to fig. 57 it will be seen that there is a sternal portion on the under side of the two posterior segments of the head, and in the embryo of Attelabus we have seen sterna also developed in the antennal and manclibular segments, so that we may conclude that there are four segments in the head of all six footed insects, corresponding to the jointed appendages, i. e. the labium, or second maxilke, the first maxilhe, the man dibles, and the antennae. Though having, in accordance with the generally received opinions of Milne-Edwards, Dana, and others, believed that the eyes of Crustacea, and therefore of Insects, were the homologues of the limbs, and developed on separate segments placed in front of the antennal segment, as stated in the previous editions of this work ; I have, however, on farther study of the subject, been led to reconsider the mat- ter, and decide that the eyes are but modified dermal sense cells, and in certain articulates developed on limb-bearing seg- ments. Thus in the King Crab (Limulns) a pair of ocelli are situated on the first segment of the body, and the lai-ge com- pound eyes grow out on the back of the third segment, both bearing limbs. In the embryos of all the insects yet exam- ined, the eyes are groups of specialized cells of the skin which grow out on the upper, or tergal, side of the same segment which bears the antennae. In certain mites, as Hydrachna, and its allies, the simple eyes are situated over the second pair of legs, and at a considerable distance behind the head. Among the Avorms, also, organs of sight, as in Potyophthalmus, are developed on each segment of the body ; or, as in certain Pla- narians, scattered irregularly over the body. The three ocelli, when present, are developed after the eyes appear. Each of these three ocelli is situated upon a distinct piece ; but we must consider the anterior single ocellus as in reality formed of two, since in the immature pupa of Bombus the anterior ocellus is transversely ovate, resulting from the fusion of two originally distinct ocelli. There are, therefore, apparently two pairs of ocelli. The clypeus and labrum are 20 THE CLASS OF INSECTS. simply a fold of the skin of the front part of the antennary segment, and are not to be compared with the tergite or rudi- ment of the eleventh segment of the abdomen. Now, since the arthropleural is the limb-bearing region in the thorax, it must follow that this region is quite well devel- oped in the head, while the tergal region, bearing the organs of sight, sometimes of enormous size, is perhaps still more largely developed ; and as all the parts of the head are subordinated in their development to that of the appendages of which they form the support, it must follow logically that the larger por- tion of the body of the head is pleural and tergal, and that the sternal parts are very slightly developed. Thus each region of the body is characterized by the relative development of the three parts of the arthromere. In the abdomen the upper (tergal) and under (sternal) surfaces are most equally devel- oped, while the pleural line is reduced to a minimum. In the thorax the pleural region is much more developed, either quite as much, or often more than the upper, or tergal portion, while the sternal is reduced to a minimum. In the head the tergites form the main bulk of the region, and the sternites are reduced to a minimum. TABLE ov THE SEGMENTS OF THE HEAD AND THEIR APPENDAGES, BEGINNING WITH THE MOST ANTERIOR. Preoral. -r,. ) C Antennae, together with First Segment t TQT ^ \ ^ ^^.^ ^ harv > I clypeus, eyes, and ocelli. Pastoral. Second Segment ) ,, I Pleural, Mandibles. (Mandibular), ) Third Segment ) ,, > Pleural, First maxilla?. (First Maxillary), ) Fourth Segment } Tergal (occiput), ( Serond Maxillary, or} > Pleural (gena), '-''"'"0, ) Sternal (guhi), The Appendages. We naturally begin with the thoracic appendages, or legs, of which there is a pair to each ring. The leg (Fig. 25) consists of six joints, the basal one, the coxa, in the Hymenoptera, Lepidoptera, and Diptera, consisting of two THE APPENDAGES. 21 pieces, i. e. the coxa and trochantine (see Fig. 12) ; the tro- chanter; the femur; the tibia, and, lastly, the tarsus, which is subdivided into from one to five joints, the latter being the normal number. The terminal joint ends in a pair of claws between which is a cushion-like sucker called the pulvillus. This sucking disk enables the Fly to walk upside down and on glass. In the larva, the feet are short and horiry, and the rig. 25. joints can be still distinguished. In Myriopods, each segment of the abdomen has a pair of feet like the thoracic ones. We must consider the three pairs of spinnerets of Spiders, which are one to three-jointed, as homologous with the jointed limbs of the higher insects. In the six-footed insects (Hexapoda), the abdominal legs are deciduous, being present in the Coleopterous grub, the Dipterous maggot, the caterpillar, and larva of the Saw-fly, but disappearing in the pupa state. They are often, as in most maggots, either absent, or reduced in number to the two anal, or terminal pair of legs ; while in the Saw-flies, there are as many as eight pairs. These "false" or "prop-legs" are soft and fleshy, and without articulations. At the retrac- tile extremity is a crown of hooks, as seen hi caterpillars or the hind-legs of the larva of Chironomus (Fig. 26), in which the prothoracic pair of legs is reduced to inarticu- late fleshy legs like the abdominal ones. The position of the different pairs of legs deserves notice in connection with the principle of " antero-posterior symmetry." The fore- legs are directed forwards like the human arms, but the two hinder pairs are directed backwards. In the Spiders, three pairs of abdominal legs (spinnerets) are retained through- out life ; in the lower Hexapods, a single pair, which is ap- pended to the eleventh segment, is often retained, but under a form which is rather like an antenna, than limb-like. In some Neuropterous larvae (Phryganea, Corydalus, etc.) the anal pair of limbs are very well marked ; they constitute the " anal forceps " of the adult insect. They sometimes become true, many-jointed appendages, and are then remarkably like FIG. 25. A, coxa; B, trochanter ; C, femur; D, tibia; F, tibial spurs; E, tarsus, divided into five tarsal joints, the fifth ending in a claw. from Sanborn. 22 THE CLASS OF INSECTS. antennae, as in the instance of Mantis tessellata described by Lacaze-Duthiers (Fig. 23). In the Cockroach these append- ages, sometimes called "analcerci," resemble the antennae of the same insect. In the Lepidoptera and Hymenoptera they do not appear to be jointed, and are greatly aborted. The Wings. The wings of insects first appear as little soft vascular sacs permeated by trachea?. They grow out in the preparatory stages (Fig. 27) of the pupa from the side of the /. thorax and above the insertion of the legs, i.e. between the epimerum and tcrgum. During the pupa state they are pad-like, but when the pupa skin is thrown off they expand with air, and in a few minutes, as in the Butterfly, enlarge to many times their original size. The wings of insects, then, are simple expansions of the crust, spread over a framework of horny tubes. These tubes are really double, consist- ing of a central trachea, or air tube, "C 1 j fy. QY inclosed within a larger tube filled with blood, and which performs the functions of the veins. Hence the aeration of the blood is carried on in the wings, and thus they serve the double purpose of lungs and organs of flight. The number and situation of these veins and their branches (veinlets) are of great use in separating genera and species. The typical number of primary veins is five. They diverge outward at a slight angle from the insertion of the wing, and are soon divided into veinlets, from which cross veins are tin-own out connecting with others to form a net-work of veins and veinlets, called the venation of the wing (Figs. 28, 29). The interspaces between the veins and veinlets are called cells. At a casual glance the venation seems very irregular, but in many insects is simple enough to enable us to trace and name the veinlets. The five main veins, most usually present, are FIG. 27. The semipupa of Bombus, the larva skin having been removed, show- ing the two pairs of rudimentary wings growing out from the mesothorax (1:), and metathorax (m}. n and the seven succeeding dots represent the eight abdominal stigmata, the first one (n) being in the pupa situated on the thorax, since the first ring of the abdomen is in this stage joined to the thorax. Original. THE WINGS. 23 called, beginning at the costa, or front edge, the costal, subcostal, median, submedian, and internal, and sometimes the median divides into two, making six veins. The costal vein is un- divided ; the subcostal and me- dian are divided into several branches, while the submedian and internal are usually simple. The venation of the fore- wings affords excellent marks in separating genera, but that of the hind wings varies less, and is consequently of less use. The wings of many insects are divided by the veins into three well-marked areas ; the costal, median, and internal. The costal area (Fig. 31 b) forms the front edge of the wing and is the strongest, since the veins are nearer together than elsewhere, and thus afford the greatest resistance to the air Fig. 29. FIG. 28. Foreand hind wings of a Butterfly, showing the venation. I. fore wing: a, costal vein; 6, subcostal vein; 61, 62, 63, 6-4, 65, five subcostal veinlets; c, inde- pendent vein (it is sometimes a branch of the subcostal, and sometimes of the me- dian vein) ; d, median vein ; d i, d 2, <>, the middle or peclicd, and the terminal part or ftujdlinii, Fl - :: ' ; - Fig. 35. or davola, which usually comprises the greater part of the antenna. It is believed by some that the sense of hearing is lodged in the antenna?, though Siebold has discovered an auditory apparatus situated at the base of the abdomen of some, and in the fore-legs of other species of Grasshoppers. Mr. J. B. Hicks has made the latest studies 011 the auditory apparatus. According to him "it consists first of a cell, sac, or cavity filled with fluid, closed in from the air by a mem- brane analogous to that which closes the foramen orale in the higher animals ; second, that this membrane is, for the most part, thin and delicate, but often projects above the surface, in either a hemispherical, conical, or canoe-shaped, or even hair- like form, or variously marked ; thirdly, that the antennal nerve gives off branches Avhich come in contact with the inner wall of the sacs ; but whether the nerve enters, or, as is most probable, ends in the small internally projecting papilla which I have shown to exist in many of these sacs, it is very difficult to say. The principal part of the nerve proceeds to these organs, the remaining portion passing to the muscles, and to the roots of the hairs, at least to those of the larger sort." On the other hand, Lefebvre, Leydig, and Gerstaecker regard this so-called "auditory apparatus" as an organ of smell. The antennae have also the sense of touch, as may readily be observed in Ants, Bees, and the Grasshopper and Cockroach. "The Honey-bee, when constructing its cells, ascertains their proper direction and size by means of the extremities of these 4 FIG. 35. Filiform antenna of Ampliisiin. From Horn. FIG. 3(5. A, lamellate antenna ol a LamelHoorn Beetle; B, antenna of a Fly, with the bristle thrown off from the terminal joint; C, bristle-like antenna of a Dragon-lly, Libcllula. From Sunbcrn. THE APPENDAGES OF THE HEAD. 27 organs ; while the same insect, when evidently affected by sounds, keeps them motionless in one direction, as if in the act of listening." (Newport.) After cutting off one or both antennae of the June beetle, Lachnostema, the insect loses its power of directing its flight or steps, wheeling about in a senseless manner. Dr. Clemens observed that the Cecropia moth was similarly affected after losing its antennae. The Mandibles (Fig. 37) are inserted on each side of the mouth-opening. They usually consist of but a single joint, H G o Fig. 37. representing probably the basal part of the ideal limb. This part, however, is often subdivided by two longitudinal furrows into three parts, each ending in a "tooth" of unequal size for tearing and cutting the food. This tripartite form of the man- dibles, to which attention has been called by Mr. Scudder, is more fully carried out in the maxilla, where each portion is highly specialized. The mandibles vary greatly in form and size. The two cutting edges are usually opposed to each other, or frequently overlap in the carnivorous forms. Their base is often concealed by the clypeus and labrum. Their motion is transverse, being the reverse of the motion of the jaws of Ver- tebrates. The Maxilke (Figs. 386, 39) are much more complicated organs than the mandibles. Fig. 39. They are FIG. 37. Different forms of mandibles. A, mandible of Cicindda purpitrea; B, f]jUopterci,a.green grasshopper; C, Libellula trimaculata ; D, ]'espa maculata, or paper-making Wasp ; E, " rostrum " or jointed sucker of the Bed-bug, Cimex lectu- lariits, consisting of mandibles, maxillae, and labium; F, proboscis, or sucker, of a Mosquito, Culex, in which the mandibles are long and bristle-like. From Sanborn. G, mandible of Amphizoa ; H, mandible of Acratus, a genus of Cockchafers. From Horn. FIG. 38. a, mentum and labial palpi; b, one maxilla, with its palpus, of Acra- tus. From Horn. FIG. 3!>. Maxilla of Amphizoa, with the two lobes (stipes and lacinia), and the palpifer bearing the four-jointed palpus. From Horn. 28 THE CLASS OF INSECTS. inserted on the under side of the head and just behind the mouth. The maxilla consists of a basal joint, or cardo, beyond which it is subdivided into three lobes, the stipes, or footstalk ; the palpifer, or palpus-bearer ; and the lacinia, or blade. The stipes forms the outer and main division of the organ. The lacinia is more membranaceous than the other parts, and its upper surface is covered with fine hairs, and forms a great part of the side of the mouth. It is divided into two lobes, the superior of which is called the galea, or helmet, which is often a thick double-jointed organ edged with stiff hairs, and is used as a palpus in the Orthoptera and many Coleoptera. The inferior lobe is attached to the internal angle of the lacinia. It terminates in a stiff minute claw, and is densely covered with stout hairs. The maxillary palpi are long, slender, one to four-jointed organs. In Perla I have found that both pairs of palpi bear organs probably of smell. The maxillae vary greatly in the different groups. Their office is to seize the food and retain it within the mouth, and also to aid the mandibles in comminuting it before it is swallowed. This function reminds us of that of the tongue of vertebrate animals. The labium, or second maxillae (Fig. 40), is placed in front of the gula, which forms the under part of the head, and is bounded a on each side by the gcnce, or cheeks, and /~] f\ posteriorly by the occiput. The gense are I \j\J 1 v J bounded laterally by the epicranium and 1 1 the under side of the eyes. In front are rig. 40. situated the basal parts of the labium, or second maxilla 3 , which embraces the submentum and mentum (or labium proper). The labial palpi are inserted into the mentum, but often the latter piece is differentiated into two, the anterior of which takes the name of pulpiyer, called by Dr. Leconte (Smithsonian Miscellaneous Collections) the ligula, and the palpi originate from them. The lifjula is the front edge of the labium, being the piece forming the under lip. It is often a fleshy organ, its inner surface being continuous FIG. 40. Ligula and labial palpi of Amphizoa, an aquatic beetle. It is quadrate and without paraglossre ; a, mentum of the same, being deeply incised, and with a tooth at the bottom of the excavation. .From Horn. THE APPENDAGES OF THE HEAD. 29 with the soft membrane of the mouth. In the Bees, it is enor- mously developed and covered with soft hairs. It is often confounded with the palpiger. In Hydrous it is divided into two lobes. Inmost of the Carabidce and Bees it is divided into three lobes, the two outer ones forming the paraglossai (Fig. 41m), and acting as feelers, while the middle, usually much longer, forms the lingua, or tongue, being the continuation of the ligula. In the bees, where the ligula is greatly developed, it performs the part of the tongue in Vertebrates, and aids the max- illse in collecting nectar and pollen. The roof of the mouth is formed by the labrum and the epipharynx (Fig. 42 c), a small fleshy tubercle concealed beneath the labrum. It is seen in the bees on turning up the labrum. It probably corresponds to the "labellum" of Schiodte. The labrum (Fig. 41 aljjij_ 1, palpifer; j, labial palpi; m, paraglossae ; k, ligula. From Newport. 30 THE CLASS OF INSECTS. (These lobes will be explained farther on when speaking of their development in the embryo.) Behind the epicra- nium is the occiput, or base of the head. It belongs to the la- bial, or second max- illary segment, and helps to form a com- plete ring, articulat- ing with the thorax. It is perforated by a foramen to afford a connection between the interior of the head and thorax. It is sometimes, as in many Coleoptera, Or- thoptera, and Hemip- tera, elongated be- hind and constricted, It will be seen beyond, that the thus formin; Fig. 42. a "neck." labrum and clypeus are in the embryo developed from a "tongue-like process whose inferior part eventually becomes the labrum, while superiorly it sends a triangular process (the rudiment of the clypeus) into the interval between the proce- phalic lobes."* This part (i.e. the ch/pcus and labrum) is the most anterior part of the head, and in the embryo, as in the adult, is normally situated in front of the ocelli, but is not to be compared Avith the " anal plate," or eleventh tergite, of the larva, or with the telson of the scorpion, as Huxle3 r t supposes. FIG. 42. Side view of the front part of the head, together with the mouth- parts of the Humble-bee (Bomlnis). , clypeus covered with hairs; b, labrum; c, the fleshy epipharynx partially concealed by the base of the mandibles (, " cere- brellum ; " c, thoracic ganglia, which distribute a nerve to each leg; d, eight pairs of abdominal ganglia. The dotted lines represent the wings. From Li'idy. 3 Fig. 43. "The under or external 34 THE CLASS OF INSECTS. close approximation to them." Newport also believes that the ganglionless upper, or internal, column of fibres is analogous to the motor column of Vertebrata, while the external, or under one, corresponds to the sensitive column, thus representing the cerebro-spinal system of the Vertebrata. From each pair of ganglia are distributed special nerves to the various organs. In the larva of /Sphinx the normal num- ber of double ganglia is thirteen, and the nervous cord of the Neuroptera and other lowly organized and attenuated forms of insects corresponds in the main to this number. In the adult insect, especially in the Coleoptera, Diptera, Lcpidoptera, and Hymenoptera, the three thoracic ganglia are fused together, following the fusion and general headwise development of the segments of the tegument. Besides the central nervous cord, corresponding to the spinal cord of the Vertebrates, there is a vagus, or visceral nerve, representing the sympathetic nerve of higher animals. This nerve "arises, in the larva, from the anterior part of the cerebrum, and, forming a ganglion on the upper surface of the phaiynx, always passes backward beneath the brain, along the middle line of the oesophagus." In its microscopic structure the nervous cord, like that of Vertebrata, consists of a central "white" substance, and an outer or peri- pheral part, the "gray" substance. In the embryo the ganglia are very large and close together, the commissures, or connecting filaments being very short, and small in proportion. ORGANS OF NUTRITION. These consist of the alimentary canal and its appendages, or accessory glands (Fig. 44). We have already treated of the external appendages (mouth-parts) which prepare the food for digestion. The simplest form of the alimentary canal is that of a straight tube. In the larva of Stylops and the sedentary young of Bees, it ends in a blind sac, as they live on liquid food and expel no solid excretions. When well developed, as in the adult insect, it becomes a long convoluted thick muscular tube, subdivided into different parts which perform different functions and have distinct names, taken from analogous organs in the vertebrate animals. This digestive tube is composed of three coats, the outer, or peri- ORGANS OF NUTRITION. 35 toneal; the middle, or muscular; and the inner, or mucous. The mucous coat is variously modified, being plaited or folded ; or, as in the Orthoptera and carnivorous Coleoptera, it is solidified and covered with rows of strong horny teeth, forming a sort of gizzard. The alimentary canal is held in place by retractor muscles, but principally by exceedingly numerous branches of the main tracheae. This canal (Fig. 45) is subdivided into the mouth and pha- rynx, the oesophagus, supplementary to which is the crop, or " sucking stomach" of Diptera, Lepidoptera, and Hymenoptera ; \\\Qproventriculus, or gizzard ; theventriculus, or true stomach, and the intestine, which consists of the ilcum, or short intes- FIG. 44. Anatomy of Sphinx ligustri. m, i, q, the nervous cord resting on the floor of the body; at c, the ganglia form a brain-like organ, much larger than the ganglia of the thorax (m) and abdomen (q). From the brain is sent off the subossophageal nerve which surrounds the gullet into which the food is conveyed by the maxilla?, or spiral tongue (), which, when at rest, is rolled up between the labial palpi (&). From the nervous cord is also thrown off a pair of nerves to each pair of legs (as at n, o, />) and a branch, d, is sent off from above, distribiiting nerves to the muscles of flight. The heart, or dorsal vessel (e,f), lies just beneath the median line of the body, and is retained in place by muscular bands (as at /) as well as by small tracheal branches. The alimentary canal (7j, j, basal ones. 2, the mandibles, and also enlarged, showing the cutting edge divided into four teeth. 3, maxilla? divided into two lobes : d, the outer and anterior lobe, 2-jointed, the basal joint terminating in two seta? ; and , the inner lobe concealed from view, in its natural position, by the outer lobe, d. 4, the base or pedicel of the second maxilla?, or labium, the expanded terminal portion being drawn sepa- rately; d and a, two movable stout styles representing, perhaps, the labial palpi; the lobe to which they are attached is multidentate, and adapted for seizing prey ; on the right side the two styles are appressed to the lobe, x represents, perhaps, the ligula; but we have not yet studied its homologies carefully: this part is attached to a transversely linear piece soldered to the main part of the labium. /, the llth abdominal ring, with its pair of conical anal styles, z, the last tarsal joint and pair of long slender claws. FIG. 64. The pupa of Diplax, having rudimentary wings, in which the eyes are much larger, and the legs much shorter than in the recently hatched larva; in- troduced to be compared with the young larva. Figs. 57-64, original. TRANSFORMATIONS OF THE INSECT. 61 On review it will be seen how remarkable are the changes in form of the insect before it is hatched, and that all are the result of simple growth. We have seen that the two ends of the body are first formed, and that the under side of the body is formed before the back ; that the belly is at first turned out- wards, and afterwards the embiyo reverses its position, the back presenting outwards. All the appendages are at first simple protrusions from the body-walls, and new segments are 'interpolated near the tip of the abdomen. These changes take place very rapidly, within a very few days, and some of the most important and earlier ones in a few hours. We can now better understand that the larva and pupa stages are the result of a similar mode of growth, though very marked from being in a different medium, the insect having to seek food and act as an independent being. TRANSFOKMATIONS OF THE INSECT. We have seen that during the growth of the embryo, the insect undergoes remark- able changes of form, the result of simple growth. The meta- morphoses of the animal within the egg are no less marked than those which occur after it has hatched. It will also be seen that the larva and pupa stages are not always fixed, defi- nite states, but only pauses in the development of the insect, concealing beneath the larva and pupa skins the most impor- tant changes of form. The process of hatching. No other author has so carefully described the process of hatching as Newport, who observed it in the larva of Meloe. "When the embryo larva is ready for its change, the egg-shell becomes thinned and concave on that side which covers the ventral surface of the body, but is much enlarged, and is more convex on the dorsal, especially towards the head. The shell is then burst longitudinally along the middle of the thoracic segments, and the fissure is ex- tended forwards to the head, which then, together with the thoracic segments, is partially forced through the opening, but is not at once entirely withdrawn. The antennae, parts of the mouth, and legs are still inclosed within separate envelopes, and retain the larva in this covering in the shell. Efforts are then made to detach the posterior segments of the body, which 62 THE CLASS OF INSECTS. are gradually released, and with them the antennas, palpi, and legs, and the larva removes itself entirely from the shell and membranes. In this process of evolution the young Meloe throws off two distinct coverings : first, the shell with its lining membrane, the analogue of the membrane in which, as I have elsewhere shown,* the }"oung Myriopod is inclosed, and re- tained several days after the bursting of the ovum, and which represents in the Articulata, not the allantois, but apparently the amnion, of Vertebrata ; next, the first, or foetal deciduation of the tegument, analogous probably to the first change of skin in the Myriopod, after it has escaped from the amnion, and also to the first change which the young Arachnidan invariably undergoes a few days after it has left the egg, and before it can take food. This tegument, which, perhaps, may be analo- gous to the vernix caseosa of Vertebrata, thrown off at the instant of birth, is left by the young Meloe with the amnion in the shell ; and its separation from the body, at this early period, seems necessary to fit the insect for the active life it has commenced." (Linn. Trans, xx. p. 306, etc.) The larva state. The larva (Latin larva, a mask) was so called because it was thought to mask the form of the perfect insect. The larvae of Butterflies and Moths are called cater- pillars; those of Beetles, gru bs; and those of the two-winged Flies (Diptera) maggots ; the larvae of other groups have no distinctive common names. As soon as it is hatched the larva feeds voraciously, as if in anticipation of the coming period of rest, the pupa state, for which stores of fat (the fatty bodies) are developed for the supply of fat globules out of which the tissues of the new body of the pupa and imago are to be formed. Most larvae moult, or change their skin, four or five times. In the inactive thin-skinned larva?, such as those of Bees, Wasps, and Gall-flies, the moults are not apparent ; as the larva increases in size it out-grows the old skin, which comes off in thin shreds. But in the active larva?, such as cater- pillars, grasshoppers, and grubs, from the rapid absorption of vessels in the outer layer of the skin, just before the change, * Philosophical Transactions, Pt. 2, 1841, p. 111. TRANSFORMATIONS OF THE INSECT. C3 it becomes hard and dry, and too small for the growing in- sect, and is then cast off' entire. A series of bee-larvae can be selected showing a graduation in size and form from the egg and recently hatched larva up to the full-grown larva. In the caterpillar and other active larvae, there are usually four or five stages, each showing a sudden and marked increase in size. Newport states that the caterpillar of Sphinx ligustri moults six times, and at the last moult be- comes a third larger than at any earlier period ; the larva of Arctia caja moults from five to ten times. A few days before the assumption of the pupa state, the larva becomes restless, stops eating, and deserts its food, and usually spins a silken cocoon, or makes one of earth, or chips, if a borer, and there prepares for the change to the pupa state. During this semipupa period (lasting, in many insects, only for a day or several days, but in some Saw-flies through the winter) the skin of the pupa grows beneath that of the quies- cent larva. While the worm-like larva exhibits no trire- gional distinctions, the muscles of the growing pupa contract and enlarge in certain parts so as to modify the larva form, until it gradually assumes the triregional form of the adult insect, with the differentiation of the body into a head, thorax, and abdomen. In a series of careful studies, abundantly illustrated with excellent plates^ Weismann has recently shown that Swammer- dam's idea that the pupa and imago skins were in reality already concealed under that of the larva is partially founded in truth. Swammerdam states, " I can point out in the larva all the limbs of the future nymph, or Culex, concealed beneath the skin," and he also observed beneath the skin of the larvae of bees just before pupating, the antennae, mouth-parts, wings, and limbs of the adult. ("VVeismann.) During its transformations the pupa skin is developed from the hypodermis, or inner layer of skin. This peals off, as it were, from the inner layer of the old larva skin, which soon dries and hardens, and is thrown off. Meanwhile the muscles of the body contract and change in form, thus causing the origi- nal segments of the larva to infold and contract at certain parts, gradually producing the pupa form. If, during this period, the 64 THE CLASS OF INSECTS. insect be examined at intervals, a series of slight changes of form may be seen, from the larva to the imago state. In some cases each change is accompanied by a moult, as in the ' ' ac- tive" Ephemera, where Lubbock counted twenty one moults. As a general rule, then, it may be stated that the body of the larva is transformed into that of the imago ; ring answer- ing to ring, and limb to limb in both, the head of the one is homologous with that of the other, and the appendages of the larva are homologous with the appendages of the imago. "Weismann has shown that in the larva of the Meat-fly, Musca vomitoria, the thorax and head of the imago are developed from what he calls "imaginal disks." These disks are minute isolated portions of the hypodermis, which are formed in the embryo, before it leaves the egg, and are held in place within the body-cavity of the larva by being attached either to nerves or tracheae, or both. After the outer layer of the larva skin dries and hardens, and forms the cask-shaped pvparium, the use of which corresponds to the cocoon of moths, etc., these imaginal disks increase in size so as to form the tegument of the thorax and head. The abdomen of the Meat-fly, however, is formed by the direct conversion of the eight hinder segments of the body of the larva, into the corresponding segments of the imago. Accompanying this change in the integument there is a destruction of all the larval system of organs ; this is either total or effected by the gradual destruction of tissues. Now we see the use of the "fatty body;" this breaks up, setting free granular globules of fat, which, as we have seen in the embryo, produces by the multiplication of cells the new tissues of the pupa. Thus the larva-skin is cast aside, and also the softer organs within, but the formation of new tissues keeps even pace with the destruction of the old, and the insect pre- serves its identity throughout. The genital glands, however, are indicated even in the embryo, and are gradually developed throughout the growth of the insect, so that this lu'stolysis, or destruction of tissues, is not wholly complete. The quiescent pupa-state of Musca is long-continued, and its vitality is latent, the acts of respiration and circulation being almost suspended. (Weismann.) TRANSFORMATIONS OF THE INSECT. 65 In the metamorphosis of CoretJira, a Mosquito-like Fly, which is active both in the larva and pupa states, "the segments of the larva are converted directly into the corresponding seg- ments of the body of the imago, the appendages of the head into the corresponding ones of the head of the imago ; those of the thorax are produced after the last moult of the larva as cUverticula of the hypodermis round a nerve or trachea, from the cellular envelope of which the formation of tissue in the interior of the appendages issues. The larval muscles of the abdominal segments are transferred unchanged into the imago ; the thoracic muscles peculiar to the imago, as also some additional abdominal muscles, are developed in the last larval periods from indifferent cellular cords which are indi- cated even in the egg. The genital glands date back to the embryo, and are gradually developed ; all the other systems of organs pass with little or no alteration into the imago. Fatty body none or inconsiderable. Pupa-state short and active." (Weismann.) As the two types are most clearly discriminated by the presence or absence of true imaginal disks, Weismann suggests that those insects which undergo a marked metamorphosis might be divided into Insecta discota (or Insects with imaginal disks), and those without, into Insecta adiscota. The metamorphosis of Corethra may prove to be a type of that of all insects which are active in their preparatory stages ; and that of Musca typical of all those that are quiescent in the pupa-state, at least the Lepidoptera and those Diptera which have a coarctate * pupa, together with the Coleoptera and those Neuroptcra in which the metamorphosis is complete, as Pliry- yl(( j )ite)-a, soon after it has taken its flight. Westwood also considers this subimago skin identical with that covering the bodies of coarctate Diptcra, as in Eristalis. Newport states, that when the imago of ^hinx is about to cast off the pupa-skin the abdominal segments are elongated beyond their original extent, this being the first part of the insect that is entirely freed from its attachment within the pupa-case. After this the thorax slits down, and the body is drawn out of the rent. In the Butterfly 'the wings mature in a few moments, but those of Fliinx being thicker, require two or three hours. Newport (Philosophical Transactions, London, 1832 and ls:U) has detailed with great minuteness the internal changes of JSpkinx h'f/ttstri while transforming. The most marked changes are in the nervous and digestive systems. Several anomalous modes of metamorphosis have been ob- served, one in Diptera and the other in S items and Mdoe. The development of the latter insect will be noticed be} T ond. Sir John Lubbock has described the singular metamorphosis of Lonchoptera, which he considers to be allied to Sarr/us, though the adult stages differ greatly. The larvre are oblong TRANSFORMATIONS OF THE INSECT. 69 ovate, flattened, with four long setae in front and two behind, with the sides of the body emarginate and spinulated. They were found under logs. "When the larva is full grown, it de- taches itself from the skin, which retains its form, and within which the insect changes into a white opaque fleshy grub con- sisting apparently of thirteen segments which gradually dimin- ish in size from one end to the other. There are no limb-cases. According to analogy the pupa should be ' incomplete ; ' it is probable, therefore, that the legs and wings make their appear- ance at a later stage. If this be so the perfect form is only attained after passing through three well-marked stages. I re- gret, however, that the specimens at my disposal did not enable me to decide this point." (Trans. Ent. Soc. London, Third Ser. i, 18G2.) Haliday states that Tlirips goes through a propupa and pupa stage. There are five well-defined stages in the Homopterous Typhlocyba, and more than three in Aphis. Yersin has noticed several stages in the development of Gryllus campestris, and the genus Psocus has four such stages. The duration of the different stages varies with the changes of the seasons. Cold and damp weather retards the process of transformation. Re'aumur kept the pupa of a Butterfly two years in an ice-house before, on being removed to a warm place, it changed to a butterfly. Chrysalids survive great alter- nations of heat and cold ; they may be frozen stiff on ice, and then, on being gradually exposed to the heat, thaw out and finish their transformations. Retrograde Development. There are certain degradational forms among the lowest members of each group of Insects which imitate the group beneath them. The Tardigrades (which are considered by some authors to be allied to the Mites) are mimicked by the low parasitic worm-like Demodex foUiculorum ; the low Neuroptera, such as Lepisma. imitate the Myriopoda ; and the wingless Lice remind us of the larvge of the Neuropter- ous Hemerobius. Among the Coleoptera, the history of Stylojis affords a strik- ing example. The active six-footed' larva is transformed into the strange bag-like female which takes on the form of a cylin- drical sac, the head and thorax being consolidated into a 70 THE CLASS OF INSECTS. minute flattened portion. The process of degradation here seems carried out to its farthest limit. Tims the degraded forms of the lower series of Hexapods take on a Myriopod aspect. In the more highly cephalized Diptera, Lepidoptera, and Hymenoptera the degraded forms are modelled on a higher articulate type. The idea of a divis- ion into three regions is involved. Thus the wingless forms of Flies, such as the Bird-louse, Nirmus; the Bat-tick, Nycte- ribia ; the Bee-louse, Braula; and Chionea resemble strikingly the biregional Arachnids. In the wingless female of Orgyia and the Canker-worm moth, the head is free, but the thorax is merged' into the abdomen. The resemblance to the lower insects is less striking. The worker ants and wingless Ichneumons, Pezomachus, still more strictly adhere to the type of their suborder, and in them the triregional form of the body persists. Among the first of the examples here cited we have seen the workings of a law, by which most degraded forms of insects (and this law is exerted with greater force in Crustacea) tend to revert to the worm-like, or, as we ma}- call it, the archetypal, form of all Articulata. "VVe have seen that many winged forms mimic the groups above them, whereas the wingless degraded species revert to a worm-like form. In either case, the progress is towards a higher or a lower form. The latter is the more exceptional, as the evolution and growth of all animals is upwards towards a more specialized, diiferentiated form. The Imago. After completing its transformations the adult insect immediately seeks to provide for the propagation and continuance of the species. The sexes meet, and, soon after, the male, now no longer of use in the insect economy, perishes. The female hastens to lay her eggs either in, upon, or near what is to be the food of the 3'oung, and then dies. This period generally occurs in the summer and autumn, and during the winter the species is mostly represented by the egg alone. Rarely does the adult insect hibernate, but in many species the pupa hibernates to disclose the adult in early summer. The larva seldom, as such, iives through the winter. Re'aumur. kept a virgin butterfly for two years in his hot- house. From this it would seem that the duration of the life GEOGRAPHICAL DISTRIBUTION. 71 of an insect may be in this way greatly prolonged. Most in- sects live one year. Hatching from the egg in early summer, they pass through the larva state, and in the autumn become pupae, to appear as imagos for a few days or weeks in the succeeding summer. Many Lepidoptera are double-brooded, and some have even three broods, while the parasitic insects such as Lice and Fleas, and many Flies, keep up a constant succession of broods. Warmth, Mr. R. C. R. Jordan remarks in the Ento- mologists' Monthly Magazine, has much to do with rapidity of development, as insects may be forced artificially into hav- ing a second brood during the same season. Some Coleoptera, such as the Lamellicorns, are supposed to live three years in the larva state, the whole time of life being four years. The Cockchafer (Melolontha) of Europe is three years in arriving at the perfect state, and the habits of the Goldsmith Beetle (Cotalpa lanigera), according to Rev. Samuel Lockwood (American Naturalist, vol. 2, p. 18G), and of the June Beetle, and allied genera, are probably the same. GEOGRAPHICAL DISTRIBUTION. The insect-fauna of a coun- try comprises all the insects found within its limits. The Polar, Temperate, and Tropical zones each have their distinct insect-fauna, and each continent is inhabited by a distinct assemblage of insects. It is also a curious fact that the insect- fauna of the east coast of America resembles, or has many an- alogues in, that of the Eastern hemisphere, and the west coast of one repeats the characteristics of the west coast of the other. Thus some California insects are either the same spe- cies or analogues (i.e. representative species) of European ones, and the Atlantic coast affords forms of which the ana- logues are found in Eastern Asia and in India. This is corre- lated with the climatic features which are repeated on alternate sides of the two hemispheres. The limits of these faunae are determined by temperature and natural boundaries, i.e. the ocean and mountain ranges. Thus the insect-fauna of the polar regions is much the same in Europe, Asia, and North America ; certain widely spread polar species being common to all three of these continents. When we ascend high mountains situated in the temperate 72 THE CLASS OF INSECTS. zone, whose summits nearly reach the snow-line, we find a few insects which are the same or very similar to those of the polar regions ; such an assemblage is called an Alpine fauna. The insect-fauna of each great continent may be divided into an Arctic, or polar, a Temperate, and a Tropical fauna, and an Alpine fauna if there are mountains in the warm latitudes which reach near the snow-line. Mountain barriers, inland seas, des- erts, and peculiarities in the flora (or collection of plants peculiar to a certain district), are boundaries of secondary importance in limiting the distribution of species. On the other hand insects are diffused by winds, rivers, oceanic currents, and the agency of man. By the latter im- portant means certain insects become cosmopolitan. Certain injurious insects become suddenly abundant in newly cultivated tracts. The balance of nature seems to be disturbed, and insects multiplying rapidly in newly settled portions of the country, become terrible pests. In the course of time, how- ever, they seem to decrease in numbers and moderate their attacks. Insect-faunae are not limited by arbitrary boundaries, but fade into each other by insensible gradations corresponding in a general way to the changes of the temperature of different portions of the district they inhabit. The subject of the geographical distribution of insects, of which we have as yet but given the rudiments, may be studied to great advantage in North America. The Arctic insect-fauna comprises Greenland, the arctic American Archipelago, and the northern shores of the continent beyond the limit of trees. A large proportion of the insects found in this region occur in arctic Europe and arctic Asia, and are hence called circnm- polar, while other species are indigenous to each country. Again, the arctic fauna of Labrador and Hudson's Bay differs from that of the arctic portions of the region about Behring's Straits, certain species characterizing one side of the continent being replaced by representative species which inhabit the opposite side. The Alpine fauna of the White Mountains consists, besides a very few peculiar to them, of circumpolar species, which are now only found in Labrador and Greenland, and which are GEOGRAPHICAL DISTRIBUTION. 73 supposed to be relics of a glacial fauna which formerly inhab- ited the northern part of the temperate zone, and in former times followed the retreat of a glacial, or arctic climate from the low-lands to the Alpine summits. These patches, or out- liers, of an Arctic fauna, containing however a preponderance of subarctic forms, also occur in the colder parts of New England. The subarctic fauna is spread over British North America, stretching north-westerly from the interior of Labrador and the northern shores of the St. Lawrence, following the course of the isothermal lines which run in that direction, and north of which no cereals grow. There are subarctic forms which inhabit the shores of the Bay of Functy, 'especially about Eastport, Maine, where the fogs and cold arctic marine currents lower the climate. Dr. J. L. Leconte, in a paper on the Coleoptera of Kansas and Eastern New T Mexico (Smithsonian Contributions to Knowl- edge), thus subdivides the Coleopterous fauna of the United States, and gives a useful map to which the reader is referred. "The whole region of the United States is divided by merid- ional, or nearly meridional lines into three, or perhaps four, great zoological districts, distinguished each b}' numerous peculiar genera and species, which, with but few r exceptions, do not extend into the contiguous districts. The eastern one of these extends from the Atlantic Ocean to the arid prairies on the west of Iowa, Missouri, and Arkansas, thus embracing (for convenience merely) a narrow strip near the sea-coast of Texas. This narrow strip, however, belongs more properly to the eastern province of the tropical zoological district of Mexico. "The central district extends from the western limit of the eastern district, perhaps to the mass of the Sierra Nevada of California, including Kansas, Nebraska, Utah, New Mexico, Arizona, and Texas. Except Arizona, the entomological fauna of the portion of this district west of the Rocky Mountains, and in fact that of the mountain region proper, is entirely un- known ; and it is very probable that the region does in reality constitute two districts bounded by the Rocky Mountains, and the southern continuation thereof. 74 THE CLASS OF IXSECTS. "The western district is the maritime slope of the continent to the Pacific, and thus includes California, Oregon, and Wash- ington Territories.. "These great districts are divided into a number of prov- inces, of unequal size, and which are limited by changes in climate, and therefore sometimes distinctly, sometimes vaguely defined." "The method of distribution of species in the Atlantic and Pacific district?, as already observed by me in various memoirs, is entirely different. In the Atlantic district, a large number of species are distributed over a large extent of country ; many species are of rare occurrence, and in passing over a distance of several hundred miles, but small variation will be found in the species obtained. In the Pacific district, a small number of species are confined to a small region of country ; most species occur in considerable numbers, and in travelling even one hundred miles, it is found that the most abundant species are replaced by others, in many instances very similar to them ; these small centres of distribution can be limited only after careful collections have been made at a great number of locali- ties, and it is to be hoped that this very interesting and im- portant subject of investigation may soon receive proper atten- tion from the lovers of science of our Pacific shores. "In the Central district, consisting, as it does to a very large extent, of deserts, the distribution seems to be of a mod- erate number of species over a large extent of country, with a considerable admixture of local species ; such at least seems to be the result of observations in Kansas, Upper Texas, and Arizona." There are a very few species which range from New England to Brazil, and fewer still (Xjjleutes robintoe, according to Bois- duval, is found in California) range from New England to California. Junonia ccetu'a, according to authors, is found both in the Southern States and California, and Pt/rrJiarctia Isabella of the Eastern States would be easity confounded with P. Cali- fonn'ca. Variation. Islands afford more variable forms than conti- nents ; the Madeiran insects and those of Great Britain vary more than the same species found on the continent of Europe. GEOGRAPHICAL DISTRIBUTION. 75 A species spread through two zones of temperature also varies ; many European species, according to McLachlan, becoming "melanized" in going northward, while others become paler. Such varieties have been described as different species. Mr. Alfred Wallace finds that the most constant forms of species are those the most limited in their geographical range as to a particular island, while those species, which range over a large part of the Mala}'an Archipelago, vary very consider- ably. It is a general rule throughout the animal and vegetable- world, that the most widely spread species are those capable of withstanding the greatest climatic changes, and adapting them- selves to the greatest diversities of topography. While the most widely distributed species are thought to be the most variable, Mr. Scudder finds in the genus Cl, are of life size, and borrowed from the new edition of Dr. Dawson's Acadian Geology. Plato 1. Fig. 1. Fisr.2. Fig. 3. Fig. -1. Fix. 5. Fig. 7. FOSSIL INSECTS. GEOLOGICAL DISTRIBUTION. 79 the characters of the Neuroptcra and Hemiptera. It is a large insect, spreading about two inches ; its body must have measured over an inch in length. In the Mesozoic rocks, the celebrated Solenliofen locality in Bavaria is rich in Liassic insect-remains. Dr. Ilagen (Ento- mologist's Annual, London, 1862) states that among the Solen- liofen fossils the Neuroptera and Orthoptera are most largely represented ; as out of four hundred and fifty species of insects, one hundred and fifty are Neuroptera, of which one hundred and thirty-six are Dragon-flies, and besides "there is a Cory- dahis, one CJirysopa, a large Apochrysa, and a beautiful Xf/m2)hes. The last two genera, which do not seem very remote from Chrysopa, are now found only in the Southern Hemi- sphere, Nymplies is peculiarly an Australian genus." The Lias of England is very rich in fossil insects, especially the Purbeck and Rhoetic Beds (see Brodie's Work on Fos- sil Insects and also Westwood in the Geological Journal, etc. Vol. X.). In the Trias, or New-Red Sandstone of the Connecticut Valley, Professor Hitchcock has found numerous remains of the larva of an aquatic insect. The insects of the Tertiary formation more closely resemble those of the present da}-. The most celebrated European locality is CEningen in Switzerland. According to Professor O. Heer, over five thousand specimens of fossil insects have been found at CEningen, comprising 844 species, of which 518 are Coleopterous. From all Tertiary Europe there are 1,322 species, as follows: 166 Hymenoptera, 18 Lepidoptera, 166 Diptera, 660 Coleoptera, 217 Hemiptera, 39 Orthoptera, and 56 Neuroptera. "If we inquire to what insect-fauna of the present period the Tertiary fauna is most analogous, we shall be surprised to find that most of the species belong to genera actually found in the old and the new world. The insect-fauna of CEningen con- tains 180 genera of this category, of which 114 belong to the Coieoptera. Of these last, two (Dineutes and Caryborus) re- main in Europe, while all the others are now found living both in Europe and in America. The whole number of Coleopterous genera furnished by CEningen, and known to me, amount to M 80 THE CLASS OF INSECTS. 158 ; those that are common to "both hemispheres forming then more than two-thirds of the whole number, Avhile of the actual Coleopterous fauna of Europe, according to the calculation of M. Lacordaire, there is only one-third. The genera found to-day in both parts of the world have then during the Tertiary epoch played a more important part than is the case now ; hence the knowledge of the character of the fauna is rendered more difficult. We find at QSningen but a very small number (five) of genera exclusively European ; seventeen are found to-day in Europe, in Asia, and in Africa, but not in America. For the most part they belong to the Mediterranean fauna (comprising eight genera) and give to the insect-fauna of CEningeii a strong proportion of Mediterranean forms. In this fauna I only know of one exclusively Asiatic genus ; two are peculiar to Africa, and two others (Anoplites and Naupactus) are American. "There are now living, however, in Europe certain genera which, without being exclusively American, since they are found in Asia and in Africa, belong more peculiarly to America ; such are Belostomwn, Hypselonotus, Diplonyclms, Evagorus, Sten- opocfct, Plecia, Caryborus, and Dineutes. . . . The genera peculiar to our fauna of Tertiary insects amount to forty-four, of which twenty-one belong to the Coleoptera ; among the Orthoptera there is one, and six Hymenoptera, six Diptera, and eleven Hemiptera. They comprise 140 species." (Heer.) An apparently still richer locality for Tertiaiy insects has been discovered by Professor Denton west of the Rocky Moun- tains, near the junction of the White and Green Elvers, Colo- rado. According to Mr. Scudder "between sixty and seventy species of insects were brought home, representing nearly all the different suborders ; about two-thirds of the species were Flies, some of them the perfect insect, others the maggot-like larvre, but, in no instance, did both imago and larva of the same insect occur. The greater part of the beetles were quite small ; there were three or four kinds of Homoptera (allied to the tree-hoppers), Ants of two different genera, and a poorly preserved Moth. Perhaps a minute Thrips, belonging to a group which has never been found fossil in any part of the world, is of the greatest interest." He thus sums up what is known of American fossil insects. THE DISEASES OF INSECTS. 81 "The species of fossil insects now known from North America, number eighty-one : six of these belong to the Devonian, nine to the Carboniferous, one to the Triassic, and sixty-five to the Tertiary epochs. The Hymenoptera, Homoptera, and Diptera occur only in the Tertiaries ; the same is true of the Lepidop- tera, if we exclude the Morris specimen, and of the Coleoptera, with one Triassic exception. The Orthoptera and Myriopods are restricted to the Carboniferous, while the Neuroptera occur both in the Devonian and Carboniferous formations. No fossil Spiders have yet been found in America." (American Nat- uralist, vol. 1, p. 630.) One species of Spider has been found in the Coal-measures of Europe, and a large number in Prus- sian Amber. THE DISEASES OF INSECTS have attracted but little atten- tion. They are so far as known mostly the result of the attacks of parasitic plants and animals, though epidemics are known to break out and carry off myriads of insects. Dr. Shinier gives an account of an epidemic among the Chinch bugs, which "was at its maximum during the moist warm weather that fol- lowed the cold rains of June and the first part of July, 1865." Species of microscopic plants luxuriate in infinitesimal for- ests within the alimentary canal of some wood-devouring insects, and certain fungi attack those species which are exposed to dampness, and already enfeebled by other causes. Among the true entopliyta , or parasitic plants, which do not however ordi- narily occasion the death of their host, Professor Leidy describes Enterobryus eleyans, E. spiralis, E. alternatus, Arthromitus crlstatus, Cladopliytum comatum, and Corynodadns radiatus, which live mostly attached to the mucous Avails of the interior of the intestine of Jnhts manjinatus and two other species of Julus, and Passahis cormitus. Eccrina longa Leidy, lives in Polydesmus Virginiensis ; and E. moniliformis Leidy in P. granulatus. But there are parasitic fungi that are largely destructive to their hosts. Such are Spliaeria and It-aria. "These fungi grow Avith great rapidity within the body of the animal they attack, not only at the expense of the nutritive fluids of the latter, but, after its death, all the interior soft tissues appear 6 82 THE CLASS OF INSECTS. to be converted into one or more aerial receptacles of spores." (Leidy.) These fungi, so often infesting caterpillars, are hence called "caterpillar fungi." They fill the whole body, distend- ing even the legs, and throw out long filaments, sometimes longer than the larva itself, giving a grotesque appearance to the insect. Leidy has found a species which is veiy common in the Seventeen-year Locust, Cicada septendecim. He found "among myriads of the imago between twelve and twenty specimens, which, though living, had the posterior third of the abdominal contents converted into a dry, powdery, ochreous- yellow, compact mass of sporuloid bodies." He thinks this Cicada is very subject to the attacks of these fungi, and that the spores enter the anal and genital passages more readily than the mouth ; thus accounting for their development in the abdomen. The most formidable disease is the " Muscardine," caused by a fungus, the Botrytus Bassiana of Balsamo. It is well known that this disease has greatly reduced the silk crop in Europe. Balbiani has detected the spores of this fungus in the eggs of Bombyx mori as well as in the different parts of the body of the insect in all stages of growth. Extreme cleanliness and care against contagion must be observed in its prevention. Among plants a disease like Muscardine, due to the presence of a minute fungus (Mucor mellitophortis) , fills the stomach of some insects, including the Honey-bee, with its colorless spores, and greatly weakens those affected. Another fungus, Sporendonema muscce, infests the common House-fly. Another Silk-worm disease called ' ' Pebvine" carries off many silk-worms. Whether it is of pathological or vegetable origin is not yet settled. There are also a few intestinal worms known to be para- sitic in insects. The well-known "Hair-worm" (Gordius) in its young state lives within the body of various insects in- cluding the Spiders. The tadpole-like young differs greatly from the parent, being short, sac-like, ending in a tail. Upon leaving the egg they work their way into the body of insects, and there live on the fatty substance of their hosts, where they undergo their metamorphosis into the adult hair-like worm, and make their way to the pools of water in which they live THE DEFORMITIES OF INSECTS. 83 and beget their species, and lay "millions of eggs connected together in long cords." Leidy thus writes regarding the habits of a species which infests grasshoppers. "The number of Gordii in each insect varies from one to five, their length from three inches to a foot ; they occupy a position in the visceral cavity, where they lie coiled among the viscera, and often extend from the end of the abdomen forward throuo-h ~ the thorax even into the head ; their bulk and Aveight are fre- quently greater than all the soft parts, including the muscles, of their living habitation. Nevertheless, with this relatively immense mass of parasites, the insects jump about almost as freely as those not infested. "The worms are milk-white in color, and undivided at the extremities. The females are distended with ova, but I have never observed them extruded. When the bodies of Grass- hoppers, containing these entozoa, are broken and lain upon moist earth, the worms gradually creep out and pass below its surface." Goureau states that Filar ia, a somewhat similar worm, in- habits Hibernia brumata and Vanessa prorsa. (Ann. Ent. Soc. France.) Siebold describes Gordius snbbifurcus which infests the Honej-bee, especially the drones, though it is rather the work- ers, which frequent the pools where the Gordii live, that we would expect to find thus infested. Another entozoan is J/er- mis albicans of Siebold, which is a very slender whitish worm much like Gordius, and about five inches long. It is found in the drone of the honey-bee and in some other insects. Deformities of Insects. Numerous instances of supernume- rary legs and antennae are recorded. The antennae are some- times double, but more commonly the legs. "Of these As- muss has collected eight 'examples, and it is remarkable that in six of them the parts on one side are treble." Newport, from whom we have quoted, states that "the most remarkable ex- ample is that given by Lefebvre of Scarites Pyrachmon in which from a single coxa on the left side of the prosternum two tro- chanters originated. The anterior one, the proper trochanter, supported the true prothoracic leg ; while the posterior one, in the form of an oblong lanceolate body, attached to the base of 84 THE CLASS OF INSECTS. the first, supported two additional legs equally well formed as the true one." The wings are often partially aborted and deformed ; this is especially noticeable in the wings of butterflies and moths. Mr. F. G. Sanborn has described and figured a wing of a female of LiWlnla o o htctuosa Bnrm. (Fig. 69), in which among other deformities "the ptero- stigma is shorter and broader than that of the opposite wing, and is situated about one-eighth of an inch only from the nodus, only one cubital vein occurring between them, instead of fourteen as in the opposite wing." (Proceed- ings of the Boston Society of Natural History, vol. xi, p. 326.) DIRECTIONS FOR COLLECTING AND PRESERVING INSECTS. Insects differ sexually in that the female generally appears to have one abdominal ring less (one ring disappearing during the cemi-pupa state, when the ovipositor is formed), and in being larger, fuller, and duller colored than the males, while the lat- ter often differ in sculpture and ornamentation. In collect- ing, whenever the two sexes are found united the}" should be pinned upon the same pin, the male being placed highest. When we take one sex alone, we may feel sure that the other is somewhere in the vicinity ; perhaps while one is frying about so as to be easily captured, the other is hidden under some leaf, or resting on the trunk of some tree near by, which must be examined and every bush in the vicinity vigorousl}" beaten by the net. Many species rare in most places have a metropolis where they occur in great abundance. During seasons when his favorites are especially abundant the collector should lay up a store against years of scarcity. At no time of the year need the entomologist rest from his labors. In the winter, under the bark of trees and in moss he can find many species, or on trees, etc., detect their eggs, which he can mark for observation in the spring when they hatch out. He need not relax his endeavors day or night. Mothing is night employment. Skunks and toads entomologize at night. Early in the morning, at sunrise, when the dew is still on the leaves, insects are sluggish and easily taken with the hand ; COLLECTING AND PRESERVING INSECTS. 85 so at dusk, when many species are found flying, and in the night, the collector will be rewarded with many rarities, many species flying then that hide themselves by day, while many caterpillars leave their retreats to come out and feed, when the lantern can be used with success in searching for them. Wollaston (Entomologist's Annual, 1865) states that sandy districts, especially towards the coast, are at all times prefer- able to clayey ones, but the intermediate soils, such as the loamy soil of swamps and marshes are more productive. Near the sea, insects occur most abundantly beneath pebbles and other objects in grassy spots, or else at the roots of plants. In many places, especially in Alpine tracts, as we have found on the summit of Mt. Washington and in Labrador, one has to lie down and look carefully among the short herbage and in the moss for Coleoptera. The most advantageous places for collecting are gardens and farms, the borders of woods and the banks of streams and ponds. The deep, dense forests, and open, treeless tracts are less prolific in insect life. In winter and early spring the moss on the trunks of trees, when carefully shaken over a newspaper or white cloth, reveal many beetles and Hymenoptera. In the late summer and autumn, toadstools and various fungi and rot- ten fruits attract many insects, and in early spring when the sap is running we have taken rare insects from the stumps of freshly cut hard-wood trees. Wollaston says, " Dead animals, partially-dried bones, as well as the skins of moles and other vermin which are ordinarily hung up in fields are magnificent traps for Coleoptera ; and if any of these be placed around or- chards and inclosures near at home, and be examined every morning, various species of Nitidulce, Silpliiclce, and other insects of similar habits, are certain to be enticed and cap- tured. "Planks and chippings of wood may be likewise employed as successful agents in alluring a vast number of species which might otherwise escape our notice, and if these be laid down in grassy places, and carefully inverted every now and then with as little violence as possible, many insects will be found adhering beneath them, especially after dewy nights and in showery weather. Nor must we omit to urge the importance 86 THE CLASS OF INSECTS* of examining the under sides of stones in the vicinity of ants' nests, in which position, during the spring and summer months, many of the rarest of our native Coleoptera may be occasion- ally procured." Excrementitious matter always contains many interesting forms in various stages of growth. The trunks of fallen and decaying trees offer a rich harvest for many wood-boring larvoe, especially the Longicorn beetles, and weevils can be found in the spring, in all their stages. Nu- merous carnivorous Coleopterous and Dipterous larvae dwell within them, and other larvee which eat the dust made by the borers. The inside of pithy plants like the elder, raspberry, blackberry, and syringa, are inhabited by many of the wild Gees, Osmia, Ceratina, and the wood-wasps, Crabro, Stigmus, etc., the habits of which, with those of their Chalcid and Ich- neumon parasites, offer endless amusement and study. Ponds and streams shelter a vast throng of insects, and should be diligently dredged with the water-net, and stones and pebbles should be overturned for aquatic beetles, Ile- miptera, and Dipterous larvae. The various sorts of galls should be collected in spring and autumn and placed in vials or boxes, where they may be rear- ed, and the rafters of out-houses, stone-walls, etc., should be carefully searched for the nests of Mud-wasps. Collecting Apparatus. First in importance is the net. This is made by attaching a ring of brass wire to a handle made to slide on a pole six feet long. The net may be a foot in diameter, and the bag itself made of thin gauze or mosquito- netting (the finer, lighter, and more durable the better), and should be about twenty inches deep. It should be sewed to a narrow border of cloth placed around the wire. A light net like this can be rapidty turned upon the insect with one hand. The insect is captured by a dexterous twist which also throws the bottom over the mouth of the net. The insect should be temporarily held between the thumb and fore-finger of the hand at liberty, and then pinned through the thorax while in the net. The pin can be drawn through the meshes upon opening the net. The beating-net should be made much stouter, with a sluil- Jower cloth bag and attached to a shorter stick. It is used for beating trees, bushes, and herbage for beetles and Hemiptera COLLECTING AND PRESERVING INSECTS. 87 and various larvae. Its thorough use we would recommend in the low vegetation on mountains and in meadows. The water- net may be either round or of the shape indicated in Fig. 70. The ring should be made of brass, and the shallow net of grass-cloth or coarse millinet. It is used for collecting aqua- tic insects. Various sorts of forceps are indispen- Fig. 70. sable for handling insects. Small delicate narrow-bladed for- ceps with fine sharp points in use by jewellers, and made either of steel or brass, are excellent for handling minute specimens. For larger ones long curved forceps are very con- venient. For pinning insects into boxes the forceps should be stout, the blades blunt and curved at the end so that the insect can be pinned without slanting the forceps much. The ends need to be broad and finely indented by lines so as to firmly hold the pin. With a little practice the forceps soon take the place of the fingers. They will have to be made to order by a neat workman or surgical-instrument maker. Some persons use the ordinary form of pliers with curved handles, but they should be long and slender. A spring set in to separate the handles when not grasped by the hand is a great convenience. Various pill-boxes, vials, and bottles must always be taken, some containing alcohol or whiskey. Many collectors use a wide-mouth bottle, containing a sponge saturated with ether, chloroform, or benzine, or bruised laurel leaves, the latter be- ing pounded with a hammer and then cut with scissors into small pieces, which give out exhalations of prussic acid strong enough to kill most small insects. Besides these the collector needs a small box lined with corn-pith, or cork, and small enough to slip into the coat- pocket ; or a larger box carried by a strap. Most moths and small flies can be pinned alive without being pinched (which injures their shape and rubs off the scales and hairs), and then killed by pouring a little benzine into the bottom of the box. Killing Insects for the Cabinet. Care in killing affects very sensibly the looks of the cabinet. If hastily killed and dis- torted by being pinched, with the scales rubbed off and other- wise mangled, the value of such a specimen is diminished 88 THE CLASS OF INSECTS. I either for purposes of study or the neat appearance of the col- lection. Besides tke vapor of ether, chloroform, and benzine, the fumes of sulphur readily kill insects. Large specimens may be killed by inserting a pin dipped in a strong solution of ox- alic acid. An excellent collecting bottle is made by putting into a wide-mouth bottle two or three small pieces of cyanide of potassium, which may be covered with cotton, about half- filling the bottle. The cotton may be covered with paper lightly attached to the glass and pierced with pin-holes ; this keeps the insect from being lost in the bottle. For Uiptera, Loew recommends moistening the bottom of the collecting box with creosote. This is excellent for small flies and moths, as the mouth of the bottle can be placed over the insect while at rest ; the insect flies up into the bottle and is immediately suffocated. A bottle well prepared will, according to Labortlbine, last several months, even a year, and is vastly superior to the old means of using ether or chloroform. He states, "the incon- venience of taking small insects from a net is well known, as the most valuable ones usually escape ; but by placing the end of the net, filled with insects, in a wide-mouthed bottle, and putting in the cork for a few minutes, they will be suffocated." Pinniny Insects. The pin should be inserted through the thorax of most insects. The Coleoptera, however, should be pinned through the right wing-cover ; many Hemiptera are best pinned through the scutellum. The specimens should all be pinned at an equal height, so that about one-fourth of the pin should project above the insect. The best pins are those made in Berlin by Klager. They are of five sizes, No. 1 being the smallest; Nos. 1, 2, and 5 are the most convenient. For very minute insects still smaller pins are made. A very good but too short pin is made by Edles- ton and Williams, Crown Court, Cheapside, London. Their Nos. 19 and 20 may be used to impale minute insects upon, and then stuck through a bit of cork, or pith, through which a No. 5 Klager pin may be thrust. Then the insect is kept out of the reach of devouring insects. Still smaller pins are made by cutting off bits of very line silvered wire at the right length, which may be thrust by the forceps into a piece of pith, after the insects have been impaled upon them. COLLECTING AND PRESERVING INSECTS. 89 Small insects, especially beetles, may be mounted on cards or pieces of mica through which the pin may be thrust. The French use small oblong bits of mica, with the posterior half covered with green paper on which the number may be placed. The insect may be gummed on the clear part, the two sexes to- gether. The under side can be seen through the thin mica. Others prefer triangular pieces of card, across the end of which the insect may be gummed, so that nearly the whole un- der side is visible. Mr. Wollaston advocates gumming small Coleoptera upon cards. Instead of cutting the pieces of cards first, he gums them promiscuously upon a sheet of card-board. "Having gummed thickly a space on your card-board equal to, at least, the entire specimen when expanded, place the beetle upon it, drag out the limbs with a pin, and, leaving it to dry, go on with the next one that presents itself. As the card has to be cut after- wards around your insect (so as to suit it), there is no advan- tage in gumming it precisely straight upon your frame, though it is true that a certain amount of care in this respect lessens your after labor of cutting-off very materially. When } T our frame has been filled, and you are desirous of separating the species, cut out the insect with finely pointed scissors." For mending broken insects, i. e. gumming on legs and an- tenna? which have fallen off, inspissated ox-gall, softened with a little water, is the best gum. For gumming insects upon cards Mr. Wollaston recommends a gum "composed of three parts of tragacanth to one of Arabic, both in powder ; to be mixed in water containing a grain of corrosive sublimate, without which it will not keep, until of a consistency just thick enough to run. As this gum is of an extremely absorbent nature, nearly a fortnight is required before it can be properly made. The best plan is to keep add- ing a little water (and stirring it) every few days until it is of the proper consistency. It is advisable to dissolve the grain of corrosive sublimate in the water which is poured first upon the gum." Preservative Fluids. The best for common use is alco- hol, diluted with a little water ; or whiskey, as alcohol of full strength is too strong for caterpillars, etc., since it shrivels them 90 THE CLASS OF INSECTS. up. Glycerine is excellent for preserving the colors of cater- pillars, though the internal parts decay somewhat, and the specimen is apt to fall to pieces on being roughly handled. Laboulbene recommends for the preservation of insects in a fresh state plunging them in a preservative fluid consisting of alcohol with an excess of arsenions acid in fragments, or the common white arsenic of commerce. A pint and a halt' of al- cohol will take about fourteen grains (troy) of arsenic. The living insect, put into this preparation, absorbs about T o 3 ,-.j of its own weight. When soaked in this liquor and dried, it will be safe from the ravages of Moths, Anthrenus, or Dcnneatex. This liquid will not change the colors of blue, green, or red beetles if dried after soaking from twelve to twenty-four hours. He- miptera and Orthoptcra can be treated in the same way. A stay of a month in this arseniated alcohol mineralizes the insect, so that it appears very hard, and, after diying, becomes glazed with a white deposit which can, however, be washed oil' with alcohol. In this state the specimens become too hard for dissection and study, but will do for cabinet specimens designed for permanent exhibition. Another preparation recommended by Laboulbene is alcohol containing a variable quantity of corrosive sublimate, but the latter has to be weighed, as the alcohol evaporates easily, the liquor becoming stronger as it gets older. The strongest solu- tion is one part of corrosive sublimate to one hundred of alco- hol ; the weakest and best is one-tenth of a part of corrosive sublimate to one hundred parts of alcohol. Insects need not re- main in this solution more than two hours before drying. Both of these preparations are A r ery poisonous and should be handled with care. The last-named solution preserves specimens from mould, which will attack pinned insects during damp summers. A very strong brine will preserve insects until a better liquor can be procured. Professor A. E. Yen-ill recommends two sim- ple and cheap solutions for preserving, among other specimens, the larvae of insects "with their natural color and form remark- ably perfect." The first consists of two and a half pounds of common salt and four ounces of nitre dissolved in a gallon of water, and filtered. Specimens should be prepared for perma- nent preservation in this solution by being previously immersed COLLECTING AND PRESERVING INSECTS. 91 in a solution consisting of a quart of the first solution and two ounces of arseniate of potasli and a gallon of water. (Pro- ceedings Boston Society Nat. Hist., vol. x, p. 257.) The nests, cocoons, and chrysalids of insects may be pre- served from injury from other insects by being soaked hi the arseniated alcohol, or dipped into benzine, or a solution of car- bolic acid or creosote. Pi'cjHir/iir/ Insects for the Cabinet. Dried insects may be moistened by laying them for twelve or twenty-four hours in a box containing a layer of wet sand, covered with one thick- ness of soft paper. Their wings can then be easily spread. Setting-boards for spreading the wings of insects may be made by sawing deep grooves in a thick board, and placing a strip of pith or cork at the bottom. The groove may be deep enough to allow a quarter of the length of the pin to project above the insect. The setting-board usually consists of thin parallel strips of board, leaving a groove between them wide enough to receive the body of the insect, at the bottom of which a strip of cork or pith should be glued. The ends of the strips should be nailed on to a stouter strip of wood, raising the surface of the setting-board an inch and a half so that the pins can stick through without touching. Several setting-boards can be made to form shelves in a frame covered with wire gauze, so that the specimens may be preserved from dust and destructive in- sects, while the air may at the same time have constant access to them. The surface of the board should incline a little to- wards the groove for the reception of the insect, as the wings often gather a little moisture, relax and fall down after the insect is dried. Moths of medium size should remain two or three days on the setting-board, while the larger thick-bodied Sphinges and Bombycidce require a week to dry. The wings can be arranged by means of a needle stuck into a handle of wood. They should be set horizontally, and the front mar- gin of the fore-wings drawn a little forward of a line perpen- dicular to the body, so as to free the inner margin of the hind wings from the body, that their form may be distinctly seen. When thus arranged, they can be confined by pieces of card pinned to the board as indicated in figure 71, or, as we prefer, by square pieces of glass laid upon them. 92 THE CLASS OF IXSECTS. After the insects have been thoroughly dried they should not be placed iu the cabinet until after having been in quarantine to see that no eggs of Dermestes or Anthrenus, etc., have been deposited on them. For preserving dried insects in the cabinet Laboulbene recommends plac- ing a rare insect (if a beetle or any Fig. 7i. other hard insect) in Avater for an hour until the tissues be softened. If soiled, an insect can be cleansed under water with a fine hair-pencil, then submit it to a bath of arseniated alcohol, or, better, alcohol with corrosive sublimate. If the insect becomes prune-colored, it should be washed in pure alcohol several times. This method will do for the rarest insects ; the more common ones can be softened on wet sand, and then the immersion in the arseniated alcohol suffices. After an immersion of an hour or a quarter of an hour, according to the size of the insect, the pin is not affected by the corrosive sublimate, but it is better to unpin the insect previous to immersion, and then pin it when almost dry. For cleaning insects ether or benzine is excellent, applied with a hair-pencil ; though care should be taken in using these substances which are very inflammable. After the specimens are placed in the cabinet, they should be farther protected from destructive insects by placing in the drawers or boxes pieces of camphor wrapped in paper perfo- rated by pin-holes, or bottles containing sponges saturated with benzine. The collection should be carefully examined every month ; the presence of insects can be detected by the dust beneath them. Where a collection is much infested with destructive insects, benzine should be poured into the bottom of the box or drawer, when the fumes and contact of the ben- zine with their bodies will kill them. The specimens them- selves should not be soaked in the benzine if possible, as it renders them brittle. Insect-cabinet. For permanent exhibition, a cabinet of shal- low drawers, protected by doors, is most useful. A drawer may be eighteen by twenty inches square, and two inches deep in the clear, and provided with a tight glass cover. For constant COLLECTING AND PRESERVING INSECTS. 93 use, boxes made of thin, well-seasoned wood, with tight-fitting covers, are indispensable. For Coleoptera, Dr. Leconte recom- mends that they be twelve by nine inches (inside measurement). For the larger Lepidoptera a little larger box is preferable. Others prefer boxes made in the form of books, which may be put away like books on the shelves of the cabinet, though the cover of the box is apt to be in the way. The boxes and drawers should be lined with cork cut into thin slips for soles ; such slips come from the cork-cutter about twelve by four inches square, and an eighth of an inch thick. A less expensive substitute is paper stretched upon a frame. Mr. E. S. Morse has given in the American Naturalist (vol. I, p. 156) a plan which is very neat and useful for lining boxes in a large museum, and which are placed in horizontal show-cases (Fig. 72). "A box is made of the re- quired depth, and a light frame is fitted to its in- terior. Upon the upper and under surfaces of this frame, a sheet of white paper (drawing or log- paper answers the pur- pose) is securely glued. Fig. 7-2. The paper, having been previously dampened, in drying con- tracts and tightens like a drum-head. The frame is then secured about one-fourth of an inch from the bottom of the box, and the pin is forced down through the thicknesses of paper, and if the bottom of the box be of soft pine, the point of the pin may be slightly forced into it. It is thus firmly held at two or three different points, and all lateral movements are prevented. Other advantages are secured by this arrangement besides firmness ; when the box needs cleaning or fumigation, the entire collection may be removed by taking out the frame, or camphor, tobacco, or other material can be placed on the bottom of the box, and concealed from sight. The annexed figure represents a transverse section of a portion of the side and bottom of the box with the frame. A, A, box ; B, frame ; 94 THE CLASS OF INSECTS. P, P, upper and under sheets of paper ; C, space between lower sheet of paper and bottom of box." Other substitutes are the pith of various plants, especially of corn ; and palm wood, and "inodorous felt" is used, being cut to fit the bottom of the box. Leconte recommends that "for the purpose of distinguish- ing specimens from different regions, little disks of variously colored paper be used ; they are easily made by a small punch, and should be kept in wooden pill-boxes ready for use ; at the same time a key to the colors, showing the regions em- braced by each, should be made on the fly-leaf of the catalogue of the collection." He also strongly recommends that the "specimens should all be pinned at the same height, since the ease of recognizing species allied in characters is greatly in- creased by having them on the same level." He also states that "it is better, even when numbers with reference to a catalogue are employed, that the name of each species should be written on a label attached to the first speci- men. Thus the eye is familiarized with the association of the species and its name, memory is aided, and greater power given of identifying species when the cabinet is not at hand." For indicating the sexes the astronomical sign $ (Mars) is used for the male, and (Venus) for the female, and 9 for the worker. Transportation of Insects. "While travelling, all hard-bodied insects, comprising many H3 T menoptera, the Coleoptera, He- miptera, and many Neuroptera should be thrown, with their larve, etc., into bottles and vials filled with strong alcohol. When the bottle is filled new liquor should be poured in, and the old may be saved for collecting purposes ; in this way the specimens will not soften and can be preserved indefinitely, and the colors do not, in most cases, change. Leconte states that "if the bottles are in danger of being broken, the specimens, after remaining for a day or two in alcohol, may be taken out, partially dried by exposure to the air, but not so as to be brit- tle, and these packed in layers in small boxes between soft paper ; the boxes should then be carefully closed with gum- paper or paste, so as to exclude all enemies." Lepidoptera and Dragon-flies and other soft-bodied insects may be well preserved by placing them in square pieces of pa- REARING LARV^:. 95 per folded into a triangular form with the edges overlapping. Put up thus, multitudes can be packed away in tin boxes, and will bear transportation to any distance. In tropical climates, chests lined with tin should be made to contain the insect- boxes, which can thus be preserved against the ravages of white ants, etc. In sending live larvae by mail, the} r should be inclosed in lit- tle tin boxes, and in sending dry specimens, the box should be light and strong, and directions given at the post-office to stamp the box lightly. In sending boxes by express they should be carefully packed in a larger box, having an inter- space of two inches, which can be filled in tightly with hay or crumpled bits of paper. Beetles can be wrapped in pieces of soft paper. Labels for alcoholic specimens should consist of parchment with the locality, date of capture, and name of collector written in ink. A temporary label of firm paper with the locality, etc., written with a pencil, will last for several years. Preservation of Larvae. Alcoholic specimens of insects, in all stages of growth, are very useful. Few collections contain al- coholic specimens of the adult insect. This is a mistake. Many of the most important characters are effaced during the drying process, and for purposes of general study alcoholic speci- mens, even of Bees, Lcpidoptera, Diptera, and Dragon-flies are very necessary. Larvce, generally, may be well preserved in vials or bottles of alcohol. They should first be put into whiskey, and then into alcohol. If placed in the latter first, they shrivel and become distorted. Mr. E. Burgess preserves caterpillars with the colors unchanged, by immersing them in boiling water thirty or forty seconds, and then placing them in equal parts of alcohol and water. It is well to collect larvae and pupae indiscriminately, even if we do not know their adult forms ; we can approximate to them, and in some cases tell very exactly what they must be. REARING LARVAE. More attention has been paid to rearing Caterpillars than the young of any other suborder of insects, and the following remarks apply more particularly to them, but 96 THE CLASS OF INSECTS. very much the same methods may be pursued in rearing the larvae of Beetles, Flies, and Hymenoptera. Subterranean larvae have to be kept in moist earth, aquatic larvae must be reared in aquaria, and carnivorous larvae must be supplied with flesh. The larvae of Butterflies are rare ; those of moths occur more frequently, while their images may be scarce. In some years many larvae, which are usually rare, occur in abundance, and should then be reared in numbers. In hunting for caterpillars bushes should be shaken and beaten over newspapers or sheets, or an umbrella ; herbage should be swept, and trees examined carefully for leaf-rollers and miners. The best specimens of moths and butterflies are obtained by rearing them from the egg, or from the larva or pupa. In confinement the food should be kept fresh, and the box well ventilated. Tumblers covered with gauze, pasteboard boxes pierced with holes and fitted with glass in the covers, or large glass-jars, are very convenient to use as cages. The bot- tom of such vessels may be covered with moist sand, in which the food-plant of the larva may be stuck and kept fresh for several days. Larger and more airy boxes, a foot square, with the sides of gauze, and fitted with a door, tlirough which a bot- tle of water may be introduced, serve well. The object is to keep the food-plant fresh, the air cool, the larva out of the sun, and in fact everything in such a state of equilibrium that the larva will not feel the change of circumstances when kept in confinement. Most caterpillars change to pupae in the autumn ; and those which transform in the earth should be covered with earth, kept damp by wet moss, and placed in the cellar until the following summer. The collector in seeking for larvae should carry a good number of pill-boxes, and especially a close tin box, in which the leaves may be kept fresh for a long time. The different forms and markings of caterpillars should be noted, and they should be drawn carefully together with a leaf of the food-plant, and the drawings and pupa skins, and per- fect insect, be numbered to correspond. Descriptions of cat- erpillars cannot be too carefully made, or too long. The relative size of the head, its ornamentation, the stripes and spots of the body, and the position and number of tubercles, and the hairs, or fascicles of hairs, or spines and spinules, ENTOMOLOGICAL WORKS. 97 which arise from them, should be noted, besides the general form of the body. The lines along the body are called dorsal, if in the middle of the back, subdorsal; if upon one side, lat- eral, and ventral when on the sides and under surface, or stig- matal if including the stigmata or breathing pores, which are generally parti-colored. Indeed, the whole biography of an insect should be ascertained by the observer ; the points to be noted are : 1. Date, when and how the eggs are laid ; and number, size, and marking of the eggs. 2. Date of hatching, the appearance, food-plant of Zarua, and number of days between each moulting ; the changes the larva undergoes, which are often remarkable, especially before the last moulting, with drawings illustrative of these ; the hab- its of the larva, whether solitary or gregarious, whether a day or night feeder ; the Ichneumon parasites, and their mode of attack. Specimens of larvae in the different moultings should be preserved in alcohol. The appearance of the larvae when full-fed, the date, number of days before pupating, the forma- tion and description of the cocoon, the duration of larvae in the cocoon before pupation, their appearance just before changing, their appearance while changing, and alcoholic specimens of larvae in the act, should all be studied and noted. 3. Date of pupation ; description of the pupa or chrysalis ; duration of the pupa state, habits, etc. ; together with alcoholic specimens, or pinned dry ones. Lepidopterous pupae should be looked for late in the summer or in the fall and spring, about the roots of trees, and kept moist in mould until the imago appears. Many Coleopterous pupae may also occur in mould, and if aquatic, under submerged sticks and stones, and those of borers under the bark of decaying trees. 4. Date when the insect escapes from the pupa, and method of escape ; duration of life of the imago ; and the number of broods in a season. ENTOMOLOGICAL WORKS. The titles of a few of the most im- portant works on Insects are given below. The more advanced student should, however, possess Dr. Hagen's Bibliotheca En- tomologica, 8vo, 2 vols., Leipzig, 1862-3, which contains a 7 98 THE CLASS OF INSECTS. complete list of all entomological publications up to the year 1862. Besides these he should consult the annual reports on the progress of Entomology published in Wiegmann's Archiv fur Naturgeschichte, begun in 1834, and continued up to the present time ; and also Gimther's Zoological Record (8vo, Van Voorst, London), beginning with the year 18G4. Occasional articles are also scattered through the various government re- ports, and those of agricultural societies and agricultural papers, GENERAL WORKS. The works of Swammerdam, ftfalphir/hi, Leemvcnhoek, Lyomiet, Serres, MecJtel, Ramdohr, Suckow, J\feria.n, and Hcrbst. Reaumur, Rend Ant. 7. Practical Entomologist. Entomological Society of Philadelphia. Vols. 1, 2, 4to, 1865-G7. Harris, T. W. A Treatise on some of the Insects of New England, which are injurious to Vegetation. Third edition, illustrated. Boston, 1862. Leconte, J. L. Classification of the Coleoptera of North America. Part 1, 1861-2. Smithsonian Institution. . List of Coleoptera of North America. 8vo, 1863-6. Smithsonian Institu- tion. New Species of North American Coleoptera. 8vo. Part 1, 1863-6. Smith- sonian Institution. Coleoptera of Kansas and Eastern New Mexico. 4to. 3 plates. 1859. Smithsonian Institution. Hagen, IT. Synopsis of the Neuroptera of North America. Svo. 1861. Smith- sonian Institution. Morris, J. G. Catalogue of the described Lepidoptera of North America. Svo, I860. Smithsonian Institution. ENTOMOLOGICAL WORKS. 101 Osten Sacken, R. Catalogue of the described Diptera of North America. 1858. Smithsonian Institution. Loew, H., and Osten Sacken, K. Monograph of the Diptera of North America. Parts 1,2, Svo, 1862-64. Smithsonian Institution. Trimble, I. P. A Treatise on the Insect Enemies of Fruit and Fruit Trees. The Curculio and Apple moth. 4to. Plates. New York, 1805. MORPHOLOGY. Savigny, J. C. Memoires sur les Animaux sans Vertebras. 1 Partie. Description et Classification des Auimaux invertebres et articules, 1. Fascicule. Theorie des Organes de la Bouche des Crustaces et des Insectes. Paris, 181(j. Audouin, J. V. Recherches anatomiques sur le Thorax des aniniaux articules et celui des Insectes hexapodes en particulier. (Anuales d. Scieuc. natur. 1, 1824, p. 97 and 416.) JEschscholtz, J. F. Beschreibung des inneren Skeletes eiuiger Insekteii aus ver- schiedenen Ordnungen. Dorpat, 1820, Svo, p. 24-49, 2 Taf. JBaer, K. E. V. Ueber das aussere tmd innere Skelet (Meckel's Archiv. f. Anatom. u. Physiol. 1S2G, p. 327-374). Erichson, jr. F. Ueber zoologische Charaktere der Insekten, Arachniden und Crustaceen. (Eutomographieu, S. 1-28.) Berlin, 1840, Svo. Brulle, A. Recherches sur les Transformations des Appendices dans les Arti- cules (Annales des Sciences naturelles, 3. ser. II, 1844, p. 271-374). Leuckart, R. Ueber die Morphologic und die Verwaudtschaftsverhaltnisse der Wirbelloseu Thiere. Braunschweig, 1848, Svo. ANATOMY AND PHYSIOLOGY. Straus-Diirckheim, H. Considerations generales sur 1' Anatomic comparee des Animaux articules, auxquelles on a joint 1' Anatomic descriptive du Melolontha vulgaris. Paris, 1828, 4to. 10 pi. Ditfour, L. Numerous anatomical papers in the Annales des Sciences naturelles, Paris. Siebold, C. Th.v. Lehrbuch der Vergleichenden Anatomic der wirbellosen Thiere. Berlin, 1848, Svo. Translated by W. I. Burnett. Boston, 1851, Svo. Gegenbaur, C. Gruudziige der vergleiehenden Anatomic. Leipzig, 1859, Svo. Geoffrey St. ITilaire, Etienne. Considerations philosophiqnes sur la determination du Systeme solide et du Systeme nerveux des Animaux articuk-s. (Auual. d. scienc. natur. II, 1824, p. 295 ff., Ill, p. 199 u. p. 453 ff.) Newport, G. On the Structure, Relations, and Development of the nervous and circulatory Systems, and on the existence of a complete Circulation of the Blood in Vessels, in Myriapoda and Macrourous Arachnida. (Philosoph. Transact- 1843, p. 243-302.) . On the Structure and Development of the Blood, I. ser. The Development of the Blood Corpuscle in Insects and other Invertebrata, and its Comparison with that of Man and the Vertebrata. (Annals of Nat. Hist. XV, 1845, p. 281-284. ) . On the Nervous System of the Sphinx ligustri Lin. and on the Changes which it undergoes during a Part of the Metamorphoses of the Insect. (Philo- soph. Transact. 1832, p. 383-398, and 1834, 389-423.) On the Temperature of Insects and its Connexion with Functions of Res- piration and Circulation in this class of Invertebrated Animals. (Philosoph. Transact. 1837, p. 259-338.) JBlanckard, E. Recherches anatomiques et zoologiques sur le Systeme nerveux des Animaux sans vertebres. Du systeme uerveux des Insectes. (Aunal. d. scieuc. natur. 3. ser. V, 1846, p. 273-379.) 102 THE CLASS OF INSECTS. Blanchard, E. Du Systeme nervcux chcz les Inverte'bres dans ses rapports avec la Classification de ces Animaux. Paris, 1840, 8vo. Milne- Edicurds, II. Legons surlaPhysiologie etl' Anatomic compare de 1'Homme et des Auimaux. Paris, Masson 1857-'J4, 8vo. EMBRYOLOGY. JlaihJ:e, IT. Untersuchungen liber die Bildung und Entwickelung des Flusskreb- scs, Leipzig, Voss. 1820, Fol. mil 5 Taf. . Zur Morphologic, Reisebemerkungen ans Taurien. Riga, 1837, 4to, mit 5 Taf. Herald, J. M. Exercitationes de animnlinm vertebris carentium in ovo formationc I. De generatione Aranearum in ovo. Untersuchnngen liber die Bildungsge- schichtc dcr Wirbellosen Tliiere im Ei. 1. Th. Von der Erzeuguug der Spiuuen im Ei. Marburg, Krieger, 1824, fol. mit 4 Taf. . Disquisitioues de animnlinm vertebris carentium in ovo formatione. De generatione Insectorum in ovo. Fasc. I, II, Frankfurt a Main, 1835-38, fol. KiJUlkcr, A. Observationes de prima Insectorum genesi, adjecta articulatorum . evolntionis cum vertebratorum comparatione. Dissert, inaug. Turici, Meyer et Zeller, 1842, 4to, c. tab. 3. Zadiliich, G. Untersnchung liber die Entwickeliing und den Ban der Gliederthiere. Heft 1. Die Eutwickeluug des Phryganiden-Eies. Berlin, Reimer. 1854, 4to, c. tab. 5. Leuckart, R. Die Fortpflanzung und Entwickelung der Pupiparen nach Beobach- tungen an Melophagns ovinns. (Abhandl. d. naturf. Gesellseh. zu Halle IV, 1858> S. 145-22G.) Huxley, T. On the agamic Reproduction and Morphology of Aphis (Transact. Linnean Soc. of London, XXII, p. 103-23fi.) Lullock, J. On the Ova and Psendova of Insects (Philosophical Transactions of the Royal Soc. 1859, p. 341-3G9. Claparetle, E. Recherches sur 1'evolution des Araignees. 4to. Utrecht, 18G2. ireisnMnn, A. Ueber die Entstehung des vollendeten Insekts in Larvennd Pnppe. Ein Beitrag zur Metamorphose der Insekten, Frankfurt a Main, 18fi3, 4 to. . Die Entwickeliing der Dipteren im Ei, nach Beobachtungen an Chirono- nuis, Musca vomitoria und Pulex canis (Zeitschrift fiir Wissenschal'tliche Zo- ologieXIII,p. 107-204.) . Die narhembryonale Entwickelung der Musciden nach Beobachtungen an Musca vomitoria uud Sarcophaga carnaria. (The same, XIV, p. 187-33(3.) FOSSIL INSECTS. Giebel, C, Fauna der Vorwelt mit steter Beriicksichtigung der lebenden Thiere. 2. Bd. Gliederthiere. 1. Abtheilung. Die Insekten und Spinnen der Vorwelt mit steter Beriicksichtigung der lebenden Insekten und Spinnen. Leipzig, 185G, 8vo. Berendt, C. ft. Die im Bernstein befindlichen organischen Reste der Vorwelt, ge- sammelt und in Verbindung mit Mehreren herausgegeben. 1. Band. 2, Abth. Die im Bernstein befindlichen Crnstaceen, Myriapodcn, Arachniden und apteren der Vorwelt, bearbeitet von C.L.Koch und C. G. Jierendt. 2. Band. Die im Bernstein befindlichen Hemipteren, Orthopteren, nd Neuropteren der Vorwelt, bearbeitet von E. F. Germar, F. J. Pictet, und //. Hay en. Berlin, 1854-5(i, fol. Heer, O. Die Insecten-faunader Tertiaergebilde von CEniugenuml Radoboj. Leip- zig, 1849, 4to, 3 vols. Scwhler, S. H. An inquiry into the Zoological Relations of the first discovered Traces of fossil Xenropterous Insects in North America. From the Memoirs of the Boston Society of Natural History, Vol. 1, 18U7, with a plate. ENTOMOLOGICAL JOURNAL. 103 PERIODICAL WORKS (now in course of publication). Edw nrds, W. H. Butterflies of North America. Colored plates. Commenced 1868. Annnles de la Societe entomologique de France, Paris. Commenced 1832. Transactions of the Entomological Society of London. Commenced 1834. V Insectologie Agricole, Monthly Journal, Paris. Commenced 1867. Zeitung. Entomologische Verein, Stettin. Commenced 1840. Llnncea entomologica. Entomologische Verein, Berlin. Commenced 184G. Zeitschrift. Entomologische Verein, Berlin. Commenced 1857. Annales de la Societe entomologique Beige, Brussels. Commenced 1857. Proceedings of the Academy of Natural Sciences, Philadelphia. Commenced 1819. Journal of the Academy of Natural Sciences, Philadelphia. Commenced 1817. Transactions of the American Philosophical Society.. New Series. Commenced 1818. Proceedings of the Boston Society of Natural History. Commenced 1834. Journal of the Boston Society of Natural History. Commenced 1834. Annals of the Lyceum of Natural History of New York. Commenced 1824. Proceedings and Transactions of the American Entomological Society, Philadel- phia. Commenced 1861. Proceedings and Communications of the Essex Institute, Salem. Commenced 1848. American Naturalist, Salem. Commenced March, 1807. ENTOMOLOGICAL JOURNAL. Every collector should keep a daily journal of his captures and observations, noting down every fact and hint that falls under his notice. In this book, commenced as soon as the season opens in early spring, can be placed on record the earliest appearance, the time of great- est abundance, and the disappearance of every insect in any of its stages. Also the descriptions of larvae, with sketches, and observations upon their habits ; though drawings had better be kept upon separate pieces of paper for easier reference. The insects, when captured and unnamed should be numbered to agree with corresponding numbers in the note-book. At the close of the season one will be surprised to see how much material of this kind has accumulated. He can then make a calendar of appearances of perfect insects and larvae, so as to have the work of the next season portioned out to him ; he will thus know when and where to look for any particular insect or caterpillar. THE NUMBER OF SPECIES OF INSECTS. Oswald Heer estimates that the Insects comprise four-fifths of the whole animal king- dom. While there are about 55,000 species of animals known, excluding the Insects, the number of this last single class amounts to upwards of 190,000 known species, according to 104 THE CLASS OF INSECTS. Gerstaecker's estimate. He reckons that there are at least 25,000 species of tfymenoptera, from 22,000 to 24,000 Lepidop- tera, about 24,000 Diptera, and 90,000 Coleoptera ; the number of the other suborders cannot be easily estimated. Besides these there are about 4,600 Araclmida, and 800 Myriopods. GROUPING OF INSECTS INTO ORDERS AND SUBORDERS. Be- fore beginning an account of the Six-footed Insects, we present the following tabular view of the Classification of In- sects. The idea that the Myriopods, Spiders, and Six-footed Insects formed orders and not classes was first proposed by R. Leuckart in 1848, and afterwards supported by Agassiz and Dana. The arrangements proposed by these and other authors are put in tabular form on page 106. THE CLASS OF INSECTS. Order I. Segments grouped into three distinct re-") gions ; eyes compound and simple ; two pairs of | HEXAPODA wings;* three pairs of thoracic legs; one pair of } (Six-footed In- jointed abdominal appendages. A more or less | sects), complete metamorphosis, . . . . . J Order II. Segments grouped into two regions, a") false cephalotho rax f and an abdomen; noantennre; | . VCHNID eyes simple ; wingless ; four pairs of thoracic legs; J> ,A' -i \ three pairs of jointed abdominal appendages (spin- | nerets) often present. No metamorphosis, . . J Order III. Body cylindrical, worm-like. Segments ~) not grouped into regions. Head free ; eyes sim- | MYPIOPODA pie; antennae present; wingless; numerous ab- (Ceut : dominal legs present; yelk-sac present for a | short period after hatching. No metamorphosis. J THE ORDER OF SIX-FOOTED INSECTS J (Hexapoda}. Metabola. The body usually cylindrical ; prothorax ") small; mouth-parts more generally haustellate j HYMENOPTERA. (formed for sucking) ; metamorphosis complete ; } LKPIDOPTEHA. pupa inactive ; larva usually cylindrical, very | DIPTKRA. unlike the adult, . . . . . . J Heterometabol'i. The body usually flattened ; pro- ^ ~ thorax large and squarish ; mouth-parts usually | HEMIPTERA adapted for biting; metamorphosis in a large YC)>- '''' number incomplete; pupa often inactive; larva | -^ ' flattened, often resembling the adult, . . J J * The number of wingless forms is comparatively few. The Diptera have but one pair. fTtie so-called " cephalothovax" of Spiders is not like that region in the Crabs, the head being much freer from the thorax. J Leuckart's classification is an advance on others in his considering the Hexa- poda, Araclmida, and Myriapoda as orders instead of classes, but he says nothing GROUPING OF INSECTS. 105 The following diagram shows, in a rude way, the relative rank and affinities of the seven suborders, and of the two series of Six-footed Insects. e. o a i-^ P Nenroptera. Through Lepisma, and Podura which are wingless Neuropter- ous insects, the lower series is connected with the Myriopods, the minute degraded myriopod, Pauropus of Lubbock, per- haps forming the connecting link ; and through the wingless flies, Bmula, Chionea, and Nycteribia, the. Diptera, belonging to the higher series, assume the form of the Spiders, the head being small, and sunken into the thorax, Avhile the legs are long and slender. The first and highest series culminates in Apis, the Honey-bee ; and the second, or lower, in Cicindela, the Tiger-beetle. regarding the rank and value of the minor groups. Professor Agassiz extended Leuckart'S views in considering the seven grand divisions of the order of Hexapods as suborders. In 1803 (How to Observe and Collect Insects, Maine Scientific Sur- vey, and Synthetic Types of Insects, Boston Journal of Natural History), we proposed a new classification of these suborders, by which they are thrown into two main groups headed by the Hymenoptera and Coleoptera respectively. These two groups, as represented in the diagram, are nearly equivalent in value, and stand in a somewhat parallel relation. There is nothing like a linear series in the animal kingdom, but it is like a tree. The higher series of suborders form more of a linear series than the lower series, so that in the diagram the Neuroptera, Orthoptera, Hemiptera, and Coleoptera form a more broken series than the Hy- menoptera, Lepidoptera, and Diptera. A Bee, Butterfly, and House-fly are much more closely allied to each other than a Beetle, a Squash-bug, a Grasshopper, and a Dragon-fly are among themselves. The Neuroptera are the most indepen- dent, and stand at the bottom of and between the two series, though by the Orthop- tera they are very intimately linked with the Hemiptera and Coleoptera. 106 THE CLASS OF INSECTS. O rH 4-i "7* O ^ ^ +j r ^ ^ c" c'.S 4 3 3 5 y''S _s Myriopot 1 6 o u - . 2 c ^ . 2 rt ^ ^j* ^ d ^2 g oS "^ cT ^ ** 3 - cS* -^ -* i 4_> cr -^ s " | ||| 1 1 | 1 1| ^J O G ?l) o i <*4 5 J^ ? ^-i -2 J^'a i I " a^ccSo ^ o ^ < g o .2 ^ || L-i c, a % 3 ^ aT ^ |"~j ^- 2 Ci . """* i'*^ . _i P^ ^ ^ "t^ / ^ N O ^ "^ O ^ ^ " ^i ^ ' -H ^ 3 I ~ H _, | -1 +? " |- rH ^ . co .S S o ^ ^ ^ o a . - f r" 1 3-3j ^d tfi ^ C ^ ,4 J ^ 2 ^ o 2 ^ &H "^ +3 ^* ^- JX t-i "* +- -O ^ 1 '""' _ i- . O 2 s i ^5 5 a ?x^ a M, =f ^ 2" 5 " ^r S of ' ^* f" *^ 7S ^ " C/3 ~-~ & ci r ~ 1 & D f t -^ r - ^i *-* "t^ "" ^-^ r~t ^ ' -f\ ^ v ^ ^5 " CT* i4 'S.'^o'~'2;^ i -s ^ t- z p t~ P'~ ^ . p g E'S ^5 o '3 ,3 '3 ~ H ' -^ ^ (5 "o 3 & 3 >..S* a"^ ^ *? 5 ;_ "5" | O 2 1-5 ^H S 3 r. CJ ^ * g. ^ g- -J >< ^ "*" P, "* ' - 1 -4-3 CO Q ^ ~ ^ M-I r ~ < 1 1 J" o 5 % | O . | O ISIIS | ^ ~S ^ *O ^ ! O a o "d H O d I o ^1 t-J H M t> M HYMENOPTERA. 107 HYMENOPTERA. THE Bees, "Wasps, Saw-flies, Ants, and other members of this suborder differ from all other insects in having, in the higher and more typical forms, the basal joint of the abdomen thrown for- ward upon and intimately united with the thorax. The head is large, with large compound eyes, and three ocelli. The mouth-parts are well developed both for biting, and feeding on the sweets of plants, the ligula especially, used in lapping nectar, being greatly developed. The other regions of the body are more distinct than in other insects ; the wings are small but powerful, with comparatively few and somewhat irregular veins, adapted for powerful and long-sustained nights ; and the genital appendages retracted, except in the Ichneu- mon parasites and Saw-flies, within the bod}^, are in the female modified into a sting. The transformations of this suborder are the most complete of all insects ; the larvre in their general form are more unlike the adult insects than in any other suborder, while the pupae, on the other hand, most clearly approximate to the imago. The larvae are short, cylindrical, footless (excepting the young of the Saw-flies, the lowest family, which are provided with abdominal legs like Lepidopterous larvoe), w r orm-like grubs, which are helpless, and have to be fed by the prevision of the parent. The pupa has the limbs free, and is generally contained in a thin silken cocoon ; that of the Saw-flies, however, being thick. The Hymenoptera exhibit, according to Professor Dana, the normal size of the insect-type. "This archetypic size is be- NOTE to page 10G. Ray divided the Hexapods into Coleoptera and Aneloptera, the latter division embracing all the other suborders except the Coleoptera. His Ametamorphota Hexapoda contained the wingless hexapoda; while the Ametamor- phota polypoda comprise the Myriopods, and the A. octopmla the Arachnids. Lin- nanis' Aptera (with numerous feet) are equivalent to the Myriopods, and his Aptera (with S-U feet) to the Arachnids. In Fabricius' system the Eleutherata are equiva- lent to the Coleoptera ; the Ulonata to the Orthoptera; the Synistata to the Neurop- tera; the riezata to the Hymenoptera; the Odonata to theLibellulida?; the Glossata to the Lepidoptera; the Wiynyota to the Hemiptera; the Antlinta to the Diptera. The Mitosnta are the Myriopods, and the Unogata, the Arachnids. In Latreille's system the Suctoria, or Fleas, are now referred to the Diptera; the Parasita or Lice, to the Hemiptera, and the Thysanura to the Xeuroptera. 108 HYMENOPTERA. tween eight and twelve lines (or twelfths of an inch) in length, and two and a half and three lines in breadth." This size is probably a smaller average than in any other suborder ; thus the Hymenoptera while being the most cephalized, consequently comprise the most compactly moulded insectean forms. Besides these structural characters, as animals, endowed with instincts and a kind of reason differing, perhaps, only in di-ijree from that of man, these insects outrank all other Articu- lates. In the unusual differentiation of the individual into males and females, and, generally sterile workers, with a farther dimor- phism of these three sexual forms, such as Huber has noticed in the Humble-bee, and a consequent subdivision of labor among them ; in dwelling in large colonies, thus involving new and intricate relations with other insects (such as Aphides, ant-hill-inhabiting beetles, and the peculiar bee-parasites) ; their wonderful instincts, their living principally on the sweets and pollen of flowers, and not being essentially carnivorous (i.e. seizing their prey like the Tiger-beetle) in their hr.bits, as are a large proportion of the other suborders, with the exception of Lepidoptera ; and in their relation to man as a domestic an- imal, subservient to his wants, the Bees, and Hymenoptera in general, possess a combination of characters which are not found existing in any other suborder of insects, and which rank them first and highest in the insect series. The body-wall of the Hymenoptera is unusually dense and hard, smooth and highly polished, and either naked, or covered with hair as in a large proportion of the bees. The head is large, not much smaller than the thorax, and its front is verti- cal. The antennae are short, filiform, often geniculate, very rarely pectinated. The mandibles are large, stout, toothed, and the maxillae are well developed into their three subdivisions, the palpi being usually six-jointed ; the labial palpi are usually four-jointed, and the prolongation of the under lip, or ligula, is highly developed, being furnished with a secondary pair of palpi, the paraglossoe, while in the pollen-gathering species the ligula is of great length, and thus answers much the same purpose as the spiral tongue (maxilla?) of the Lepidoptera. Reaumur states that the Bee does not suck up the liquid sweets, but laps them up with its long slender hairy tongue. HYMENOPTERA. 109 "Even in the drop of honey the bee bends the end of its tongue about, and lengthens and shortens it successively, and, indeed, Avitlulraws it from moment to moment." The liquid passes along the upper surface of the pilose tongue, which is withdrawn between its sheaths, the palpi and maxilhe, and thus "conveys and deposits the liquid with which it is charged within a sort of channel, formed by the upper surface of the tongue and the sheaths which fold over it, by which the liquid is conveyed to the mouth." (Shuckard.) The thorax forms a rounded compact oval mass, with the prothorax and metathorax very small, the mesothorax being large, and also the propodeum, to which the pedicel of the ab- domen is attached. The pleurites are large and bulging, while the sternum is minute. The coxae and trochantines are large, and quite free from the thorax ; and the trochanters are small, while the rather slender legs are subject to great modifications, as they are devoted to so many different uses by these insects ; thus, in the Sand-wasps they are strongly bristled for the purpose of digging, and in the Bees, the basal joint of the tarsi is much enlarged for carrying pollen. "The manner in which the bee conveys either the pollen, or other material it purposes carrying home, to the posterior legs, or venter, which is to bear it, is very curious. The rapidity of the motion of its legs is then very great ; so great, indeed, as to make it very difficult to follow them ; but it seems -first to collect its material gradually with its mandibles, from which the anterior tarsi gather it, and that on each side passes successively the grains of which it consists to the inter- mediate legs, by multiplicated scrapings and twistings of the limbs ; this, then, passes it on by similar mano3iivres, and de- posits it, according to the nature of the bee, upon the pos- terior tibiae and tarsi, or upon the under side of the abdomen. The evidence of this process is speedily manifested by the pos- terior legs gradually exhibiting an increasing pellet of pollen. Thus, for this purpose, all the legs of the bees are more or less covered with hair. It is the mandibles which are chiefly used in their boring or excavating operations, applying their hands, or anterior tarsi, only to clear their way ; but by the construc- tive, or artisan bees, they are used both in their building and 110 HYMENOPTERA . mining operations, and are worked like trowels to collect moist clay, and to apply it to the masonry of their habitations." (Shuckard.) The four wings are present, except in rare instances. They are small ; the hinder pair long, narrow, ovate, lanceolate. The costal edge of the fore-wing (Fig. 29), is generally straight, becoming a little curved towards the apex, which is obtusely subrectangular ; the outer edge is bent at right angles, while the inner edge of the wing is long and straight. The veins are often difficult to trace, as in the outer half of the wing they break up into a system of net-veins, which are few in number, yet the continuations of the subcostal, median, and submedian A-eins can be distinguished after careful study. In some low Ichneumonidce, the Proctotrupidce, and Chalcididce, the veins show a tendency to become obsolete, only the simple subcostal vein remaining ; and in Pteratomns, the veins are entirely obliterated, and the linear feather-like wings are in one pair fissured, reminding us of the Plume- moths, Pteropliorus. The abdomen is composed in the larva state of ten segments, but in the adult stinging Hymenoptera, of six complete seg- ments in the females, and seven in the males ; while in the lower families the number varies, having in the Tenthredi- n itla>, eight tergites on the upper side and six sternites on the lower side. The remaining segments are, during the transfor- mations of the insect, aborted and withdrawn within the body. The ovipositor and corresponding parts in the male haA'e been described on pp. 14-18. The nervous system consists in the larvae of eleven ganglia, in the adult five or six of these remain as abdominal ganglia, while the remainder, excluding the cephalic ganglia, are placed in two groups in the thorax. The cerebral ganglia are well developed, evincing the high intellectual qualities necessary in presiding over organs with such different uses as the simple and compound eyes, the antennae, and lingua and palpi, and mandibles, especially in those sociable species which build complete nests. The digestive system, in those bees which sip up their food, consists, besides the external mouth-parts, of a "long cesoph- HYMENOPTERA . Ill agns which dilates into a thin-walled sucking stomach," which in the Apia rice and Vespidce may be simply a lateral fold, or, as in many CrabronidcB, "attached solely by a short and narrow peduncle." In Formica, Cynips, Leucospis, and Xyijhid- ria there is a globular uncurved callous gizzard, which is en- veloped by the base of the stomach, according to Siebold, who also states that "those Hymenoptera which are engaged during a long and active life in labors for the raising and support of their young, have a pretty long and flexuous stomach and in- testine, and the first has, usually, many constrictions ; " while the Cynipidce, Iclineumonidae, and Tenthredinidce, which take no care of their young, have only a short small stomach and intestine. The salivary glands consist of two rather short ramified tufts, often contained entirely in the head. The tracheae consist, as in other insects, of two main branches, from which numerous transverse anastomosing branches are given off, with numerous vesicular dilatations. Two such vesi- cles of immense volume are situated at the base of the abdo- men, which according to Hunter and Newport "serve chiefly to enable the insect to alter its specific gravity at pleasure dur- ing flight, and thus diminish the muscular exertion required during these movements." The urinary vessels are very numerous in the Hymenoptera ; they are usually short and surround the pylorus in numbers of from twenty to one hundred and fifty. The two poison glands (Fig. 54, h,g) are composed of long ramose tubes, resembling the salivary glands in their minute structure. The poison is poured from these into a pyriform sac lodged near the base of the sting, which is provided with a peculiar muscular apparatus for its sudden extension and with- drawal. The poison, in the Ants, Bees, and Wasps, consists, according to Will, of "formic acid, and a whitish, fatty, sharp residuum, the former being the poisonous substance." (Bur- nett.) The wax-secreting apparatus consists of special dermal glands, as Milne- Ed wards supposed. Glaus has shown (see Gegenbaur's Verg. Anatomic) that these minute glands are mostly unicellular, the external opening boing through a fine chiiinous tube on the outer surface of tin iutogunaent. la the 112 H YMENOPTERA. wax-producing insects these glands are developed in great numbers over certain portions of the body. In the Aphides, whose bodies are covered with a powder consisting of flue waxy threads, these glands are collected in groups. Modifications of them appear in the Coccidne. In the wax-producing Hymen- optera the apparatus is somewhat complicated. The bees secrete wax in thin, transparent, membranous plates on the under side of the abdominal segments. Polygonal areas are formed by the openings of an extraordinarily large number of fine pore-canals, in which, surrounded by very numerous tra- cheal branches, the cylindrical gland-cells are densely piled upon each other. These form the wax organs, over which a fatty layer spreads. In those bees which do not produce wax, the glands of the wax organs are slightly developed. Wax organs also occur in the Humble bees. The honey is elaborated by an unknown chemical process, from the food contained in the proventriculus, or crop, and which is regurgitated into the honey-cells. The ovaries consist of many-chambered, four, six, or a hun- dred, short tubes. "The receptacula seminis is nearly always simple, round or ovoid, and necked, and is prolonged into a usually short seminal duct." The gldndula appendicularis con- sists of a bifurcate tube which opens into the dnctus seminalis, and only rarely into the capsula seminalis itself. In the Tenthredinidce , "this apparatus is formed on a different type ; the seminal vesicle is a simple diverticulum of the vagina, and more or less distinct from it, besides it is defi- cient in the accessory gland. The copulatory pouch is absent in all the Hymenoptera, as are also the sebaceous glands with those females which have a sting and a poison gland," while in other insects the sebaceous glands are present, and it would be nat- urally inferred, therefore, that the two are homologous, but modified for diverse functions. The two testes of the male are "composed of long follicles, fasciculate and surrounded, together with a portion of the torose deferent canal, by a common envelope ; but more com- monly the two testes are contained in a capsule situated on the median line of the body." (Siebold.) The eggs are usually long, cylindrical, and slightly curved in HYMENOPTERA. 113 the Bees ; in the "Wasps they are more globular, and affixed by their smaller somewhat pedicelled end to the side, near the bot- tom of the cell in which they are laid. The eggs of the lower families tend to assume a spherical form. The eggs of dif- erent species of Bombus present no appreciable differences. The larvae of the Bees and Wasps, especially the social species, which live surrounded by their food, are of a very persistent form, the various genera differing but slightly, while the species can scarcely be separated. Such we have found to be the case in the Bees and Wasps ( Vespiclce) and Fossorial Wasps. The sexes of the species with a very thin tegument, such as Apis, Bombus, and Vespa, can be quite easily distin- guished, as the rudiments of the genital armor can be seen through. The Hymenoptera are mostl}' confined to the warmer and temperate regions of the earth ; as we approach the poles, the Bees disappear, with the exception of Bombus, and perhaps its parasite Apatlius ; a species of Vespa is found on the Lab- rador coast, which has a climate like that of Greenland. No fossorial species of Wasps are known to us to occur in the arc- tic regions, while a few species of Ants, and several Chaleidi- dce and Ichneumonidce are not uncommon in Northern Labrador and Greenland. Our alpine summits, particularly that of Mt. Washington, reproduces the features of Northern Labrador and Greenland as regards its Hymenopterous fauna. The tropics are, however, the home of the Hymenoptera, and especially of the Bees. There are estimated to be about twenty-five thousand living species of this suborder, and this is probably a much smaller number than are yet to be discovered. In geological history, the Hymenoptera do not date far back compared with the Neuroptera and Orthoptera, and even the Coleoptera. Indeed they were among the last to appear upon the earth's surface. The lower forms, so far as the scanty records show, appeared first in the Jura formation ; the Ants appear in the Tertiary period, especially in amber. As we have noticed before, the Hymenoptera are more purely terrestrial than any other insects. None are known to be aquatic in the early stages, and only two genera have been found 8 114 HYMEXOPTERA. swimming in the adult state on the surface of pools, and they are the low, minute, degraded Proctotrupids, Presticichia natans and Polt/nema natans described by Mr. Lubbock. The Hymenoptera do not imitate or mimic the forms of other in- sects, but, on the contrary, their forms are extensively copied in the Lepidoptera, and especially the Diptera. A partial excep- tion to this law is seen in the antennae of the Australian genus Thaumatosoma, where they are long and slender, and knobbed as in the butterfly, and also in Tetralonia mirabilis of Smith, from Brazil. The Hymenoptera, also, show their superiority to all other in- sects in the form of their degraded wingless species, such as Pezomachus, the workers of Formica and the female of Mutilla. In these forms we have no striking resemblances to lower orders and suborders, but a strong adherence to their own Hymenop- terous characters. Again ; in the degradational winged forms, we rarely find the antenna? pectinated ; a common occurrence in the lower suborders. In a low species of the A.piance, Lamprocolletes dadocerus, from Australia, that land of anom- alies, the antenna? are pectinated. This, Mr. F. Smith, the best living authority on this suborder, says, "is certainly the most remarkable bee that I have seen, and the only in- stance, to my knowledge, of a bee having pectinated antennae ; such an occurrence, indeed, in the Aculeate Hymenoptera is only known in two or three instances, as in Psammotherma JJab- ellata amongst the Mutillidce, and again in Ctenocerus Klugii in the Pompilidce. ; there is also a modification of it in one or two other species of Pomp Hi dee." Among the Tenthre- dinidce, the male Lopliyrus has well-pectinated antennae, as also has Cladomacra macropus of Smith, from New Guinea and Celebes. The wings of perhaps the most degraded H^ymenoptera, the Proctotruftidce, arc rarely fissured ; when this occurs, as in Pleratomus Putnamii, they somewhat resemble those of Ptero- phorus, the lowest moth. It is extremely rare that the com- pound e} r es are replaced by stemmata, or simple eyes ; in but one instance, the genus Anthophorabia, are the eyes in the male sex reduced to a simple ocellus. This species lives in the darkness of the cells of Anthophora. APIARLE. 115 By reason of the permanence of the type, due to the high rank of these insects, the generic and specific characters are founded on very slight differences, so that these insects, and particularly the two higher families, the Wasps ( Vespidce) and Bees (Apiarice} are the most difficult insects to study. The easiest characters for the recognition of the genera, lie in the venation of the wings ; though in the fossorial families the legs vary greatly. The best specific characters lie in the sculptur- ing and style of coloration, but the spots and markings are apt to vary greatly. The great differences between the sexes art- liable to mislead the student, and hence large collections are indispensable for their proper study. Bees act as "marriage priests" in the fertilization of plants, conveying pollen from flower to flower, and thus insuring the formation of the fruit. It is said that many plants could not be fertilized without the interposition of Bees. Their interesting habits deserve long and patient study ; it is for their observations on the insects of this suborder that the names of Reaumur, the two Hubers, and Latreille will be ever held in special remembrance. Most Hymenoptera love the sun, and they may be caught while flying about flowers. The nests of bees, wasps, and ants should be sought for and the entire colony captured, together with the parasites. The hairy species should be pinned while in the net, and the naked ones can be put in the collecting-bot- tle. The larger species may be pinned, like other insects, through the thorax ; but the minute Chalcids, etc., should be gummed, like small Coleoptera, upon cards. The nests of bees and of wasps and ants and the young in various stages of growth should be collected, and in such num- bers as to show their different stages of construction, to serve as illustrations of insect architecture. APIARI^E Latreille (Apidce, Leach). This and those families succeeding which are provided with a true sting, were called by Latreille Hymenoptera Aculeata. The male antennas are mostly thirteen-jointed, while in the female they are twelve- jointed. The females (and the workers, when they exist) feed the larvae, which mostly live in nests or cells. 116 HYMEXOPTERA. In the social Bees, besides the normal male and female forms, there are asexual females, whose inner genital organs arc partly aborted, though externally only differing in their smaller size from the true females. The male antennae are longer, tapering more towards the tips, and the eyes of the male approach each other closer over the vertex than in the opposite sex, though these are characters which apply to other Hymenoptera. The mouth-parts are in the higher genera greatly elongated, the labium being long, with the lingua of great length, and the lobes of the maxillae long and knife-shaped ; but these parts, as well as the form of the jaws, arc subject to great modifications in the different genera : the labial palpi are four-jointed, and the maxillary palpi are from one to six-jointed. The hind tibia and basal joint of the tarsi are, in the pollen-gathering species, very broad ; the tibia is in Apis and Bombus hollowed on the outside, and stiff' bristles project over the cavity from each side of the joint, forming the honey-basket (corbiatlnm) , on which the "clodden masses of hone} T and pollen" are con- ve3"ed to their nests. In the parasitic genera, such as Apathus, the tibia is, on the contrary, convex, rather than concave, though of the usual width; while in Nomada, also parasitic, the legs are narrow, the tibia not being dilated. In Andrena and its allies, Hal ictus and Colletes, the mouth- parts, especially the tongue, are much shortened, thus afford- ing a passage into the Vespidce. . In these genera the tongue is folded back but once between the horny encasement of the maxilhe, but in the higher Apiariw the part formed by the union of the lingua and maxilla is twice bent back, and thus protected by the horny lobes of the maxillae. The fore- wings have two or three subcostal (cubital) cells. There are two thousand species of this family. The differ- ences between the larvae of the various genera of this family are very slight, those of the parasitic species are, however, readily distinguished from their hosts. The higher 'Ap tar ice, comprising the subfamily Apinw, have the ligula long, cylindrical, while the labial palpi have two very long, slender, compressed basal joints, and two short terminal joints. The genus Apis has no terminal spurs on the hind tibiae, APIARLE. 117 while the fore-wings have three subcostal (cubital) cells, the middle of which is elongated and acutely wedge-shaped. The eyes in the male are united above ; the mouth-parts are nearly aborted, and the hind legs are smooth. In the female there are two paraglossoe on the ligula, and the maxillary palpi are one-jointed. The worker only differs externally from the female in the shorter abdomen. The larva of the Honey-bee closely resembles that of Bom- bus, but the body is shorter, broader, and more flattened, while the head is less prominent, and the lateral tubercles along the body arc, perhaps, less prominent than in the. young Humble- bee, otherwise the two genera are, in the larval state, much alike. In its natural position, the larva lies at the bottom of the cell doubled upon itself. Though the larvre are said usually to feed upon pollen, Mr. Desborough states that honey alone is the food of the grub, as he reared 729 larvae with no other food than honej-. But as with the wild bees they may extract honey from the pollen provided for them. He says the matured bees may be observed feeding at night on the bee-bread (pollen). Lang- stroth (The Hive and Honey-bee), however, states that "pol- len is indispensable to the nourishment of the 3 T oung. It is very rich in the nitrogenous substances which are not contained in the honey." The Honey-bee, Apis mellrftca, is now distributed over the civilized world. It was introduced into this country during the seventeenth century, and into South America in 1845 (Ger- staecker). The Italian, or Ligurian, bee is considered by F. Smith as being a climatic variety. The cultivation of the Honey-bee is rapidly increasing in this country, but the German Bee-masters have made the most pro- gress in theoretical and practical Bee-culture. Convenient hives are now constructed by which all the operations of the bees can be observed at leisure. Gersttecker thus sums up the habits of the Honey-bee : A fertilized queen which, with a few workers, has wintered over, lays its eggs in the spring first in the worker, and afterwards, at a later period, in the drone- cells (both arranged in two perpendicular rows of cells). Early in summer, the workers construct the larger flask- shaped queen- 118 HYMENOPTERA. cells, which are placed on the edge of the comb, and in these the queen-larvae are fed with rich and choice nourishment. As soon as the first of the HCAV brood of qneens is excluded from its cell, which it indicates by a peculiar buzzing noise, the old queen deserts the nest, carrying away with her a part of the swarm, and thus forms a new colony. The recently excluded queen then takes its marriage flight high in the air with a drone, and on its return undertakes the management of the hive, and the duty of laying eggs. When another queen is disclosed, the same process of forming a new colony goes on. When the supply of young queens is exhausted, the workers fall upon the drones and destroy them without mercy. The first brood of workers live about six weeks in summer, and then give way to a new brood. Mr. J. G. Desborough states that the maximum period of the life of a worker is eight months. The queens are kno .vn to live five years, and during their whole life lay more than a million eggs (V. Berlepsch). Langstroth states that "during the height of the breeding season, she will often, under favorable circumstances, lay from 2,000 to 3,000 eggs a day." According to Von Siebold's discovery only the queens' and workers' eggs are fertilized by sperm- cells stored in the receptaculum seminis, and these she can fertilize at will, retaining the power for four or five years, as the muscles guarding the duct leading from this sperm-bag are subject to her will. Drone eggs are laid by unfertilized queen-bees, and in some cases even by worker-bees. This last fact has been confirmed by the more recent observations of Mi % . Tegetmeier, of London. Principal Leitch, according to Tegetmeier, has suggested the theory that a worker egg may develop a queen, if transferred into a queen-cell. "It is well known that bees, deprived of their queen, select several worker-eggs, or very young larvae, for the purpose of rearing queens. The cells in which these eggs are situated are lengthened out and the end turned down- OO * ' ward." He suggests that the development into a queen was caused by the increased temperature of the queen-cell, above that of the worker-cells. But Messrs. F. Smith and Woodbuiy (Proceedings of the Entomological Society of London, January 2, 18G2) support F. APIARLE. 119 Huber's theory, that the change is due to "the quality as well as quantity of food with which the royal larva is supplied," though Dr. Leitch objects, that it has been by no means con- clusively proved " that the so-called royal jelly differs hi any respect from the ordinary food supplied to the worker larva ; " and Mr. Woodbury cites the experiments of Dzierzon, as quoted by Kleine, "that as Huber, by introducing some royal jelly in cells containing worker-brood, obtained queens, it may be possible to induce bees to construct royal cells, when the Apiarian prefers to have them, by inserting a small portion of royal jelly in cells containing worker-larvae." Kleine takes "an unsealed royal cell which usually contains an excess of roj^al jelly and removes from it a portion of the jelly, on the point of a knife or pen, and by placing it on the inner margin of any worker cell, feels confident that the lan'oe in them will be reared as queens." Before these points are settled we must study the habits of the Wild Bees, and of the other social Hymenoptera and White Ants, together with the social Aphides more carefully. Mr. F. W. Putnam pertinently states, "at "present I cannot believe that the peculiarity of food, or the structure of the cells, pro- duces a difference of development in Humble-bees, for the lar- vae, as has been previously stated, were seen to make their own cells from the pollen paste. Is it not more natural to believe, as has been suggested to me by Professor J. Wyman, that the difference in the development of the eggs is owing to their be- ing laid at various times after impregnation? Thus, if I am right in supposing that the queens are impregnated by the males late in the summer, the eggs, laid soon after, produce the large queen larvse ; * the next set of eggs, laid in the spring, produce the workers, or undeveloped females, while from those deposited still later, male bees are principally developed." (Proceedings of the Essex Institute, Salem, vol. iv, 1864, p. 103.) Referring to Mr. Putnam's statement that there are both small and large queens (besides the workers), Dr. Gerstaecker infers, * Dr. Gerstsecker, on the other hand, states that "from the brood-cells of a nest of Hiimbus muscorum, found by him on the 18th of September, there were devel- oped at the end of the same month only workers." 120 HYMENOPTERA. "from the examination of numerous individuals found flying in the spring after hibernation, that these could not be considered as true queens, since their ovaries were only moderately devel- oped, though larger than those of the workers, while in the true queen, captured in the summer, the ovaries were perfectly developed. This corresponds almost entirely to what we find in the wasps, whose spring females have only moderately de- veloped ovaries." How the Honey-bee builds its cells, and whether they are ex- actly hexagonal, are questions that have interested the best observers from Maraldi who wrote in 1712, and Reaumur, whose Memoires appeared in 1740, down to the present date. Their solution involves not only the closest observation of the' insect while at work, but also the shrewdest judgment to ex- plain the facts observed and deduce a legitimate tlieoiy. Does the bee intelligently plan her work out beforehand, or does she follow the guidance of what is called instinct? Does she construct hexagonal cells which are mathematically exact, or does she vary the proportions of each cell, so that it is per- fect only in its general idca; Fig. 5, larva. Fig. 6, Tri- chodes apiiiriim . a, larva ; b, pupa. Fig. 7, .I/Hoe anr/ustlcoUis : Fig. 8, freshly hatched larva; Fig. 0, second stage of larva; Fig. 10, first stage of semi-pupa; Fig. 11, pupa. Fig. 12, Stylops CMMreni in the body of a wild bee, Andrena; Fig. 13, top view of the same removed from its host; Fig. 14, male of the same; a, side view. Fig. 15, Mucor mellitophorus, a parasitic fungus. Fig. 16, unknown larva found in nest of Humble-bee. Descriptions of the insect parasites will be given beyond. Plate 2. Fi?. 1. Fijr. ft. Ffc. 10. Fip;. r,. Fig. -t. Fig. 8. Fipr. 14. Fitr. I.'). z APIARI^. 129 the species is also confirmatory of the same supposition ; in- deed, the great diversity in this respect observable in these bees, appears to me to be analogous to a similar diversity in the length of the bills of humming-birds, which, it is well known, are always adapted for reaching the nectaries of the particular flowers which they usually frequent." In regard to the immense numbers of individuals in a col- ony, Mr. Stretch, who collected them at Panama, "found a nest several feet in length in the hollow of a tree, containing thousands of individuals, their numbers being, as he informs me, apparently countless. "Gardner, in his travels, gives a list of such species (of Melipona) as he met in the provinces of Piauhy and Goyaz, where he found them numerous; in every house, he says, 'you find the honey of these bees ; ' many species, he tells us, build in the hollow trunks of trees, others in banks ; some suspend their nests from branches of trees, whilst one species constructs its nest of clay, it being of large size ; the honey of this spe- cies, he says, is very good." (Smith.) In a nest of Trigona carbonaria from Eastern Australia, Smith, of the British Museum, found from 400 to 500 dead workers crammed in the spaces between Ilie combs, but he did not find a female among them. The combs are arranged precisely similar to those of the common wasp. The number of honey-pots, which are placed at the foot of the nest, amounted to 250. Smith inclines to the opinion that the hive of Trigona con- tains several prolific females ; "the accounts given of the mul- titudes inhabiting some nests is too great, I think, to render it possible that one female could produce them all. Mr. Stretch described a hive that he saw, occupying the interior of a decay- ing tree, that measured six feet in length, and the multitude of bees he compared to a black cloud. M. Guerin found six fe- males in a nest of Melipona fulvipes" Hill states, in Gosse's Naturalist's Sojourn in Jamaica, "that the wax of these bees [Trigona] is very unctuous and dark colored, but susceptible of being whitened by bleaching. The honey is stored in clusters of cups, about the size of pigeon's eggs, at the bottom of the hive, and always from the 9 130 HYMEXOPTEEA. brood-cells. The brood-cells are hexagonal ; they are not deep, and the young ones, when ready to burst their casement, just lill the whole cavity. The mother bee is lighter in color than the other bees, and elongated at the abdomen to double their length." Smith also states that the female of this genus has the abdomen greatly distended, reminding one of the gravid female of the White Ant. (Smith, Proc. Ent. Soc., London, Dec. 7, 18G3.) In North America, our nearest ally, as regards its habits, of the true Honey-bee, is the Humble-bee (Bombtcs), of which over forty species are known to inhabit North America. The economy of the Humble-bee is thus : the queen awakens in early spring from her winter's sleep beneath the leaves or moss, or in deserted nests, and selects a nesting-place generally in an abandoned nest of a field-mouse, or beneath a stump or sod, and "immediately," according to Mr. F. W. Putnam, "collects a small amount of pollen mixed with honey, and in this deposits from seven to fourteen eggs, gradually adding to the pollen mass until the first brood is hatched. She does not wait, however, for one brood to be hatched before laying the eggs of another ; but, as soon as food enough has been collected, she lays the eggs for a second. The eggs [Plate 4, Fig. 2] are laid, in contact with each other, in one cavity of the mass of pollen, with a part of which they are slightly covered. They are very soon developed ; in fact, the lines are nowhere dis- tinctly drawn between the egg and the larva, the larva and pupa, and again between the latter and the imago ; a perfect series, showing this gradual transformation of the young to the imago, can be found in almost every nest. "As soon as the larvae are capable of motion and commence feeding, they eat the pollen by which they are surrounded, and, gradually separating, push their way in various directions. Eating as they move, and increasing in size quite rapidly, they soon make large cavities in the pollen mass. When they have attained their full size, they spin a silken wall about them, Avhich is strengthened by the old bees covering it with a thin layer of wax, which soon becomes hard and tough, thus form- ing a cell. [Plate 4, Figs. 1, 2.] The larvae now gradually attain the pupa stage, and remain inactive until their full devel- 3/7 APIARLE. 131 opmcnt. They then cut their way out, and are ready to assume their duties as workers, small females, males or queens. "It is apparent that the irregular disposition of the cells is due to their being constructed so peculiarly by the larva?. After the first brood, composed of workers, has come forth, the queen bee devotes her time principally to her duties at home, the workers supplying the colony with honey and pollen. As the queen continues prolific, more workers are added, and the nest is rapidly enlarged. "About the middle of summer eggs are deposited which produce both small females and males." . . . "All eggs laid after the last of July produce the large females, or queens ; and, the males being still in the nest, it is presumed that the queens are impregnated at this time, as, on the approach of cold weather, all except the queens, of which there are several in each nest, die." (Putnam, Com. Essex Inst., vol. iv, p. 98, 1864.) Besides Apathus, the larvae of various moths consume the honey and waxen cells ; the two-winged flies, Volucella and Conops, and the larvte of what is either an Anthomyia or Tachina-like fly ; several species of Anthrax, the Coleopterous Anobium paniceum of Europe, Meloe, Stylops, and Anthero- phagus ochraceus are parasitic on Humble-bees.* The habits of the genus Apatlms are not clearly known, but they are supposed to prey, in the larva state, upon the larvae of Bombus, being found in their nests ; their habits, so far as known, ally them with Nomada. The species are distinguished by the tibia? being convex, instead of concave, as in Bombus, while the mandibles of the females are acute, triangular, biden- tate, being spatulate and three-toothed in Bombus, and they have no pollenigerous organs. There are males and females only, as in all the remaining genera of the family. Apathus Ashtonii (Plate 3, Fig. 1) is found in the Northern States. * EXPLANATION OF PLATE 3. Parasites of the Humble and Leaf-cutter Bees. Fig. I, Apathus Ashtonii. Fig. 2, Nephopteryx Edmandsii; a, larva; b, pupa. Fig. 3, 3, Microgaster nephoptericis, an Ichneumon parasite of Xephopteryx. Fig. 4, Antherophfigus ochraceus. Fig. 5, Anthomyial larva; a, side view. Fig. 6, Re- cently hatched larva of Stylops Childrenii; a, side view. Fig. 7, larva; a, pupa of Anthophorabia megachilis, a Chalcid parasite on Megachile. Fig. 8, Fteratomus Putnamii, an exceedingly minute Proctotrupid fly, supposed to be parasitic on An- thorphorabia megachilis ; a, a hiud wing. Fig. 9, a Mite found ill the nests of Humble-bees. 132 HYMEXOPTERA. Xylocopa, the Carpenter-bee, is "the largest and most bulky of all known bees," but less hirsute than Bombus, while the basal joint of the labial palpi is almost four times as long as the second ; and the maxillary palpi are six-jointed, the mouth- parts being very highly organized. The larva of A". Virginica (Plate 4, Fig. 3, adult ; Fig. '4, larva ; Fig. 5, nest) is slenderer than that of Bombus, the body tapering more rapidly towards each end. The power of boring the most symmetrical tunnels in solid wood reaches its perfection in the large Virginian Carpenter- bee (Xylocopa Virginica). We have received from Mr. James Angus, of West Farms, N. Y., a piece of trellis for a grape- vine, made of pine wood, containing the cells and young in various stages of growth, together with the larvse and chiysa- lids of Anthrax shniosa (Plate 4, Fig. 6, larva; Fig. 7, pupa), a species of fly parasitic on the larva of the bee, and which buries its head in its soft body and feeds on its juices. Mr. Angus thus writes us regarding its habits, under date of July 19 : "I asked an intelligent and observing carpenter yes- terday, if he knew how long it took the Xylocopa to bore her tunnel. He said he thought she bored about one-quarter of an inch a day. I don't think myself she bores more than one- half inch, if she does that. If I mistake not, it takes her about two da}'s to make her own length at the first start ; but this being across the grain of the wood may not be so easily done as the remainder, which runs parallel with it. She always follows the grain of the wood, with the exception of the en- trance, which is about her own length. The tunnels run from one to one and a half feet in length. They generally run in opposite directions from the opening, and sometimes other gal- leries are run above the first, using the same opening. I think they only make new tunnels when old ones are not to be found, and that the same tunnels are used for many years. Some of the old tunnels are very wide. I have found parts of them about an inch in diameter. I think this is caused by rasping otf the sides to procure the necessary material for con- structing their cells. The partitions are composed of wood- raspings, and some sticky fluid, probably saliva, to make it adhere. APIAKI.E. 133 "The tunnels are sometimes taken possession of by other bees and wasps. I think when this is the case, the Xylocopa prefers making a new cell to cleaning out the mud and rubbish of the other species. I frequently find these bees remaining for a long time on the wing close to the opening, and bobbing their heads against the side, as if fanning air into the opening. I have seen them thus employed for twenty minutes. Whether one bee, or more, makes the tunnel, that is, whether they take turns in boring, I cannot say at present. In opening the cells, more than one are generally found, even at this season. About two weeks ago, I found as many as seven, I think, in one." * The hole is divided by partitions into cells about seven-tenths of an inch* long. These partitions are constructed of the dust or chippings made by the bee in eating out her cells, for our active little carpenter is provided with strong cutting jaws, moved by powerful muscles, and on her legs are stiff brushes of hair for cleaning out the tunnel as she descends into the heart of the solid wood. She must throw out the chips she bites off from the sides of the burrow with her hind legs, pass- ing the load of chips backwards out of the cell with her fore- limbs, which she uses as hands. The partitions are built most elaborately of a single flattened band of chips, which is rolled up into a coil four layers deep. One side, forming the bottom of the cell, is concave, being * " Since writing the above I have opened one of the new holes of Xylocopa which was commenced between three and four weeks ago, in a pine slat used in the staging of the greenhouse. The dimensions were as follows: Opening fully 3-S wide ; depth 7-10 ; whole length of tunnel G and 5-16 inches. The tunnel branched both ways from the hole. One end, from opening, was 2 and 5-8, containing three cells, two with larva and pollen, the third empty. The other side of the opening, or the rest of the tunnel, was empty, with the exception of the old bee (only one) at work. I think this was the work of one bee, and, as near as I can judge, about twenty-five days' work. Width of tunnel inside at widest !)-!(> inch. For some days this bee has been discharging a great quantity of saw-dust and pollen, which I had collected by placing a vessel under it. It would seem that she had cells constructed also in the opposite side of the hole, and that she removed them to enlarge the tunnel. Among the stuff thrown out, I find a partition of a cell nearly entire. I have just found a Xylocopa bobbing at one of the holes, and in order to ascer- tain the depth of the tunnel, and to see whether there were any others in them, I sounded with a pliable rod, and found others in one side, at a depth of five and one half inches; the other side was four inches deep, without bees. The morning was cool, so that the object in bobbing could not be to introduce fresh currents of air, but must have had some relation to those inside. The legs on such occasions are, as I have noticed, loaded with pollen." American Naturalist, vol. 1, p. 370. 134 HYMENOPTERA. beaten down and smoothed off by the bee. The other side of the partition, forming the top of the cell, is flat and rough. At the time of opening the burrow, July 8th, the cells con- tained nearly full-grown larvae, with some half developed. They were feeding on the masses of pollen, which were as large as a thick kidney-bean, and occupied nearly half the cell. Sa- pyga repanda is parasitic in the cells of Xylocopa violacea of Southern Europe. The habits and structure of the little Ceratina ally it closely with Xylocopa, as it hollows out the stems of plants, and builds in them its cylindrical cells. This bee is oblong in form, with tridentate mandibles, and a short labrum. The maxillary palpi are six-jointed, and the labial palpi are two-jointed. Ceratina dupla Say is a common small bright-green smooth-bodied species, which, in the middle of May, according to Dr. Harris' MS. notes, tunnels out the stems of the elder or blackberiy, syringa, or any other pithy shrub, excavating them often to a depth of six or seven inches, and also, according to Mr. Haldcman (Harris MS.), bores in Cocorus. She makes the walls just wide enough to admit her body, and of a depth capable of holding three or four, often five or six cells (Plate 4, Fig. 11). The finely built cells, with their delicate silken walls, are cylindrical and nearly square at each end, though the free end of the last cell is rounded off. They are four and a half tenths of an inch long, and a little over one-third as broad. The bee places them at nearly equal distances apart, the slight interval between them being filled in with dirt. Dr. T. W. Harris* states that, "May 15, 1832, one female laid its eggs in the hollow of an aster-stalk. Three perfect in- sects were disclosed from it July 28th." The observations of Mr. Angus, who saw some bees making their cells, May 18th, also confirms this account. The history of our little upholsterer is thus cleared up. Late in the spring she builds her cells, fills them with pollen, and lays one or more eggs upon each one. Thus in about two months the insect completes its transforma- tions ; within this period passing through the egg, the larval and chrysalid states, and then, as a bee, living through the win- ter. Its life thus spans one year. * According to a note hi MSS. deposited in the Library of the Boston Society of Natural History. APIARI^E. 135 The larva (Plate 4, Fig. 10) is longer, than that of Mega- chile, and compared with that of Xylocopa, the different seg- ments are much more convex, giving a serrate outline to the back of the worm. The pupa, or chrysalis, we have found in the cells the last of July. It is white, and three-tenths of an inch long. It differs from that of the Leaf-cutter bee in having four spines on the end of the body, and in having a much longer tongue and maxillae, both being almost twice as long. In none of the wild bees are the cells constructed with more nicety than those of our little Ceratina. She bores out with her jaws a long deep well just the size of her body, and then stretches a thin delicate cloth of silk, drawn tight as a drum- head, across each end of her chambers, which she then fills with a mixture of pollen and honey. Her young are not, in this supposed retreat, entirely free from danger. The most invidious foes enter and attack them. Three species of Ichneumon-flies, two of which belong to the Chalcid family, lay their eggs within the body of the larva, and emerge from the dried larva and pupa skins of the bee, often in great numbers. The smallest parasite, belonging to the genus Anthophorabia (so called from being first known as a parasite on another bee, Anthophora) , is a minute species found also abundantly in the tight cells of the Leaf-cutter bee. The species of Anthidinm, according to Smith, are gaily marked with yellow bands and spots ; the ligula is almost twice as long as the labial palpi, and acutely pointed ; the paraglossse are short, the maxillary palpi are two-jointed, and there are two subcostal cells. The males are longer than the females, with an elongated and stoutly toothed abdominal tip. The female lines her nest, situated in any hole convenient for its purpose, with down from woolly-stemmed plants. They pass the winter in the larva state, and the bees do not appear until mid-summer. The species mostly occur in the old world. In Anthophora, which approaches nearer to Bombus in its plump and hairy body than the two preceding genera, the lig- ula is twice as long as the labial maxillse, ending in a bristle- like point ; the basal joint of the hind tarsus is thickly hirsute, while the middle tarsus of the males is generally elongated. The species are gregarious, their numerous cells, while indepen. 136 HYMENOPTERA. dent, are crowded together in grassy banks. Species of Melecta are parasitic on them, ovipositing in their cells. The larvae are infected by the Chalcid flies, Anthophorabia and Monodontomerus, and by a peculiar species of Mite, Hete- ropus ventricosus, described by Newport. Say has described Antltophora abrupta and A. taurea from Indiana. In Eucera the antennae are very long, while the body is still plump and hairy : our more common form in the Middle States is Eucera maculata St. Fargeau. The species are likewise gregarious, and, according to Smith, their habits are precisely the same as those of Anthophora. In Megachile, the Leaf-cutter Bee, the head is broad, the body stout, oblong, the ligula is about one-half longer than the labial palpi, being quite stout, while the paraglossae are short and pointed ; the maxillae are long and sabre-shaped, while their palpi are short and two-jointed. There are two subcostal cells in the fore wing. It is a thick-bodied bee, with a large square head, stout scissor-like jaws, and with a thick mass of dense hairs on the under side of the tail for the pur- pose of carrying pollen, since it is not provided with a pollen basket as in the Honey and Humble-bees. The larva is broader and flatter than that of Bombus, the raised pleura! region is a little more prominent, and the raised, thickened tergal portion of each ring is more prominent than in Bombus. The Megachile lays its eggs in burrows in the stems of the elder (Plate 4, Fig. 9), which we have received from Mr. James Angus ; we have also found them in the hollows of the locust tree. Mr. F. W. Putnam thus speaks of the economy of M. centuncularis, our most common species. "M}' attention was first called, on the 26th of June, to a female busily en- gaged in bringing pieces of leaf to her cells, which she was build- ing under a board, on the roof of the piazza, directly under my window. Nearly the whole morning was occupied by the bee in bringing pieces of leaf from a rose-bush growing about ten yards from her cells, returning at intervals of a half minute to a minute with the pieces which she carried in such a manner as not to impede her walking when she alighted near her hole. [We give a figure of the Leaf-cutter bee in the act of cutting out a circular piece of a rose-leaf (Plate 4, Fig. 8). She 137 alights upon the leaf, and in a few seconds swiftly runs her scissors-like jaws around through the leaf, bearing off the piece iu her hind legs.] About noon she had probably com- pleted the cell, upon which she had been engaged, as, during the afternoon, she was occupied in bringing pollen, preparatory to laying her single egg in the cell. For about twenty days the bee continued at work, building new cells and supplying them with pollen. . . . On the 28th of July, upon removing the board, it was found that the bee had made thirty cells, arranged in nine rows of unequal . length, some being slightly curved to adapt them to the space under the board. The longest row contained six cells, and was two and three-quarters inches in length; the whole leaf-structure being equal to a length of fifteen inches. Upon making an estimate of the pieces of leaf in this structure, it was ascertained that there must have been at least a thousand pieces used. In addition to the labor of making the cells, this bee, unassisted in all her duties, had to collect the requisite amount of pollen (and honey?) for each cell, and lay her eggs therein, when com- pleted. Upon carefully cutting out a portion of one of the cells, a full-grown larva was seen engaged in spinning a slight silken cocoon about the walls of its prison, which were quite hard and smooth on the inside, probably owing to the move- ments of the larva, and the consequent pressing of the sticky particles to the walls. In a short time the opening made was closed over by a very thin silken web. The cells, measured on the inside of the hard walls, were .35 of an inch in length, and .15 in diameter. The natural attitude of the larva is some- what curved in its cell, but if straightened, it just equals the inside length of the cell. On the 31st of July, two female bees came out, having cut their way through the sides of their cells." In three other cells "several hundred minute Ichneu- mons [Anthophorabia megachilis] were seen, which came forth as soon as the cells Avere opened." (Com. Essex lust., vol. iv, p. 105, 18G4.) Megachile integer Say MS., according to Dr. Harris (MS. notes), forms its nest of leaves the first of August. This spe- cies is tAvice as large, but closely resembles Megachile brevis of Say. The front of the head is covered with dense ochreous 138 HYMENOPTERA. hairs, becoming shorter and black on the vertex. The nest, preserved in the Harris collection, now in the Museum of the Boston Society of Natural History, is made of rose-leaves, and is scarcely distinguishable from that of M. centuncularis. Osmia, the Mason Bee, is another genus of Carpenter or Upholsterer bees. The species are generally bluish, with greenish reflections, with smooth shiny bodies, and the species are of smaller size than in Megachile. The tongue in this genus is three times us long as the labinm, tapering from the base to the acute apex, and clothed with short hair. Mr. F. Smith states that the larva of the English species hatch in eight days after the eggs are laid, feeds ten to twelve da} r s, when it becomes full-grown, then spins a thin silken c ;vering, and remains in an inactive state until the following spring, when it completes its transformations. The habits of the little Mason-bees are quite varied. They construct their cells in the stems of plants and in rotten posts and trees, or, like Andrena, they burrow in snnnj" banks. An European species selects snail-shells for its nest, wherein it builds its earthen cells, while other species nidificate under stones. Curtis found two hundred and thirty cocoons of a British species (Osmia paretina), placed on the under side of a flat stone, of which one-third were empty. Of the remainder, the most appeared between March and June, males appearing first ; thirty-five more bees were developed the following spring. Thus there were three successive broods for three succeeding years, so that these bees lived three j-ears before arriving at maturity. Mr. G. R. TVaterhouse, in the Transactions of the Entomo- logical Society of London, for 1864 (3d series, vol. 2, p. 121), states that the cells of Osm/n leucomelana "are formed of mud, and each cell is built separately. The female bee. having de- posited a small pellet of mud in a sheltered spot between some tufts of grass, immediately commences to excavate a small cavity in its upper surface, scraping the mud away from the centre towards the margin by means of her jaws. A small shallow mud-cup is thus produced. It is rough and uneven on the outer surface, but beautifully smooth on the inner. On witnessing thus much of the work performed, I was struck with APIARI^E. 139 three points. First, the rapidity with which the insect worked ; secondly, the tenacity with which she kept her original position whilst excavating ; and thirdly, her constantly going over work which had apparently been completed. . . . The lid is excavated and rendered concave on its outer or upper surface, and is convex and rough on its inner surface ; and, in fact, is a simple repetition of the first-formed portion of the cell, a part of a hollow sphere." The largest species of Osmia known to us is a very dark-blue species which seems to be undescribed. We will call it the wood-boring Osmia (Osmia lignivora). It is larger than the Osmia lignaria of Sa}-, being just half an inch long. The head is much shorter, and less square than in Say's spe- cies. The front of the head below the antennte is clothed with dark hairs, but above and on the thorax with yellowish ochreous hairs. The body is deep blackish blue, with greenish reflec- tions. We are indebted to a lady for specimens of the bees with their cells, which had been excavated in the interior of a maple tree several inches from the bark. The bee had industri- ously tunnelled out this elaborate burrow (Plate 4, Fig. 12), and, in this respect, resembles the habits of the Carpenter-bee (Xyloco2M) more closely than any other species of its genus. The tunnel was over three inches long, and about three- tenths of an inch wide. It contracted a little in Avidth between the cell, showing that the bee worked intelligently, and wasted no more of her energies than was absolutely necessary. The burrow contained five cells, each half an inch long, being rather short and broad, with the hinder end rounded, while the opposite end, next to the. one adjoining, is cut off squarely. The cell is somewhat jug-shaped, owing to a slight constriction just behind the mouth. The material of which the cell is com- posed is stout, silken, parchment-like, and very smooth within. The interstices between the cells are filled with rather coarse chippings made by the bee. The bee cut its way out of the cells in March, and lived for a month afterwards on a diet of honey and water. It eagerly lapped up the drops of water supplied by its keeper, to whom it soon grew accustomed, and whom it seemed to recognize. The female of Osmia lignaria Say MS., according to Dr. 140 HYMENOPTERA. Harris' MS. notes, was found in the perfect state in cocoons within earthen cells under stones, April loth. The cell she con- structs is half an inch long, oval, cylindrical, and contracted slightly into a sort of neck just before the opening for the exit of the bee. From Mr. James Angus I have received the pellets of pollen, about the size of a pea, in which it deposits its eggs ; the larvse were about one-third grown in August. This species is larger than Osmia simillima of Smith, while the male antennae are much paler, being fuscous. The front of the head is covered with long dense yellow ochreous hairs. The vertex is not of so dark a green as in 0. simillima, and is covered with coarse punctures. The thorax is heavily clothed with yellow ochreous, thick, hairs. The abdomen is yellowish, and much more hairy. The legs are stout, fuscous, with yel- lowish hairs. Length, .35 inch. Our smallest and most abundant species is the little green Osmia simillima of Smith. It builds its little oval, somewhat urn-shaped cells, against the roof of the large deserted galls of the oak-gall fly (Diplolepis coufluentus), placing them, in this instance, eleven in number, in two irregular rows, from which the mature bees issue through a hole in the gall (Plate 4,* Fig. 14. From specimens communicated by Mr. F. G. Sanborn). The earthen cells, containing tke 'tough dense cocoons, were arranged irregularly so as to fit the concave vault of the larger gall, which was about two inches in diameter. On emerging from the cell the Osmia cuts out with its powerful jaws an ovate lid, nearly as large as one side of the cell. Both sexes may be found in April and May in the flowers of the willow * EXPLANATION OF PLATE 4. Fig. 1, a cell of the Humble-bee; natural size, with the pollen mass upon the top. Tig. 2, end view of the same mass, showing the three eggs laid in three divisions of the cavity. Fig. 3, Xylocopa Virginian, the Carpenter Bee. Fig. 4, the larva of Xijloropa Virfjinica; natural size. Fig. 5, the nest containing the cells of the same, with the partitions and pollen masses, on which the young larva is seen in the act of feeding; natural size. Fig. 6, young larva of Anthrax simiosa; side view. Fig. 7, pupa of Anthrax sinuosa, side view; natural size. Fig. 8, the Leaf-cutter Bee (Meyaclille), on a rose leaf, in the act of cutting out a circular piece. Fig. 9, cells of Megachile, in the elder; natural size. Fig. K), larva of c, mtinn ilt/pla, the little green Upholsterer Bee; enlarged. Fig. 11, cells of the same in the stem of the elder; natural size. Fig. 12, cells of Osmia Ur/nivora, new species, the wood-devouring Mason-bee, exca- vated in the maple ; natural size. Fig. 13, cells of Omnia simitlima, the common green Mason-bee, built in the deserted gall of the Oak-gall Fly. Fig. 14, a single earthen cell of the same; natural size. Fig. 15, pollen mass, or bee-bread of Osmia lignaria ; natural size. It is made up of distinct pellets of pollen, which are probably stuck together with saliva. Flute 4. 141 and fruit trees which blossom later. The antennas are black, and the green body is covered with fine white hairs, becoming yellowish above. In the Harris collection are the cells and specimens of Osmia pacifica Say, the peaceful Osmia, which, according to the man- uscript notes of Dr. Harris, is found in the perfect state in earthen cells (Plate 5, Fig. 2) beneath stones. The cell is oval cylindrical, a little contracted as usual with those of all the spe- cies of the genus, thus forming an urn-shaped cell. It is half an inch long, and nearly three-tenths of an inch wide, while the cocoon, which is rather thin, is three-tenths of an inch long. The following genera, called Cuckoo Bees, are parasitic on other bees, laying their eggs in the cells, or nests, of their host. In Codioxys the body is stout, and the bee closely mimics its host, Megachile. The ligula is very long, being almost three times the length of the labium, and the paraglossse are wholly wanting ; the maxillary palpi are short, three-jointed, and the abdominal tip of the male is variously toothed. Codioxys octo- dentata Say, is abundant late in the summer about flowers. An allied genus, Melecta, is parasitic on Anthophora, and Epeolus is parasitic on Colletes. The species of Nomada are very numerous ; in all, the tongue is long and acute, with paraglossa? about one-fourth as long as the tongue ; the maxillary pair of palpi are six-jointed ; and there are three subcostal cells. The species in their slen- der, smooth, gaily colored body resemble the wasps. These Cuckoo-bees lay their eggs in the nests of Andrena and Ha- lictus, and, according to English authors, Panurgus and Eucera, where they may be found in all stages of development corre- sponding to those of their hosts. The females do not sting severely. The species emit sweet, balmy, or balsamical odors. Shuckard states that these bees should be killed with burning sulphur to preserve their bright colors. The larvie differ greatly from those of their hosts, Andrena, the head being much smaller, the body being smoother and rounder, and belonging to a more degraded, lower type. The whole body is more attenuated towards both extremities. The pupa differs from those of any other genus of this family known to us, except Audrena, by having three conspicuous 142 HYMENOPTERA. spines on the upper and posterior edge of the orbit, which are also found in the pupa of Stigmus, a Oabronid genus, and which evidently aid in locomotion. Thus the same law of degrada- tion obtains in these highly organized bee-parasites as in the lower parasitic species, though in a much less marked degree. From specimens found in the nests of Andrena and Halictus, collected at Salem by Mr. J. H. Emerton, and now in the Mu- seum of the Essex Institute, we have been enabled in great part to clear up the history of this bee. We have found in the nests of Andrena vicina both sexes of Nomada imbricata Smith, and several females of Nomada pulchella of Smith ; and in the cells of Halictus parallelus Say, specimens of Nomada imbri- cata. Both full-grown larvae and pupae of different ages, up to the adult Nomada, ready to take leave of its host, were found in the cells of the Andrena vicina. It seems, there- fore, that the newly hatched young of Xomada must feed on the pollen mass destined for the Andrena. But there seems to be enough for both genera to feed upon, as the young of both host and parasite were found living harmoniously to- gether, and the hosts and their parasites are disclosed both at the same time. Does not this mild sort of parasitism in No- mada throw much light on the probable habits of Apathus, the Humble-bee parasite? It is more than probable that the Apa- thus larvae simply eat the food of the Bombus larvae, and do not attack the larvae of their hosts. Both Nomada and Apathus in their adult stages live harmoniously with their hosts, and are seen gathering food from the same flowers, and flying about the same nest. In the second subfamily, Andrenetce, the ligula, or tongue, is for the most part short and broad, and the maxillary palpi have four joints of equal size. In Spliecodes the body is smooth and wasp-like, and in its habit of running and flying in dry sandy places, it resembles Sphex, whence its generic name. The abdomen is generally light red, farther aiding in the resemblance to the Sphegidce . The ligula is short, lancet-shaped, fringed with setae ; the para- glossae are not so long as the tongue, Avhile the labial palpi are shorter than the paraglossoe, and the maxillae are broad, lan- ceolate, with six-jointed palpi. The antennae of the males are APIAEI.E. 143 short and sometimes moniliform. Sphecodes dichroa Harris is our most common species. Mr. F. Smith, from direct observa- tion, states that this genus builds cells, though earlier authors have stated that it is parasitic on Halictus and Andrena. Prosopis is generally yellow on the face, and is "less pubes- cent than any of the bees." The tongue is broad, subemar- ginate, the paraglossre reach a little beyond the tongue ; the- labial palpi are as long as the tongue, while there are two sub- costal cells in the fore wings. Smith states that the genus 13 not parasitical as formerly supposed, as he has "repeatedly bred them" from cells laid in a regular order in the hollow of bramble stems. Mr. S. Saunders has also raised them in Alba- nia where "they construct their cells in bramble sticks (which they bore in the same manner as Colletes) with a thin transpa- rent membrane, calculated for holding semi-liquid hone}', which they store up for their young. The species are much attacked by Stylops." Like Sphecodes and Ceratina, this genus, accord- ing to Smith, is unprovided with pollenigerous organs. We have several species in this country of which P. affuiis Smith, and P. elliptica Kirby, are found northward. The habits of our species are not known. Augochlora comprises beautiful shining metallic green spe- cies, very commonly met with. The thorax is globose, and the anterior wings have one marginal and three submarginal cells ; the first submarginal cell as long as the second and third united. Augochlora pants Smith is a small, green, rather common species. Mr. J. H. Emerton has found its nests in Sa- lem, near those of Andrena. The mouth of the hole opened under a stone, and was built up so as to form a tube of sand (Plate 5, Fig. 1). The burrow on the 28th of June was four inches deep. Andrena is a genus of great extent, and the species are often difficult to distinguish. The lanceolate tongue is moderately long, and the paraglossse are half as long as the tongue itself, while the six-jointed maxillary palpi are longer than the maxillae themselves. The wings have three subcostal cells, with the rudiments of a fourth one ; the second is squarish, and the third receives a recurrent nervure near the middle. The pos- terior legs " have a long curled lock upon the trochanter be- 144 HYMENOPTERA. neath, and the anterior upper surface of the femora is clothed with long loose hair, which equally surrounds the whole of the tibiie." (Shuckard.) The abdomen is banded more or less conspicuously with reddish. The larva (Fig. 79) is stout and thick, with a head of moder- ate size, and the month-parts are a little shorter than usual, the maxilliB and labinm especially. The segments of the body are much more convex (angularly so) than usual, giving a tuberculate outline to the body. It is stouter than that of Halictus, the wings are less convex than in that genus ; while the maxilla; are much stouter and blunter. The pupa is distinguished from the other genera by much the same characters as the imago, except that there rig. 70. are wo tubercles on the vertex near the ocelli. From a comparison of all its stages, this genus stands inter- mediate between those placed above, and Halictus, which, in all its characters, is a more degraded form. The males often differ widely from the other sex, in their broad heads and widely spreading bidentate mandibles. Mr. Emerton has observed the habits of our most common species, Andrena vicina Smith, which builds its nest in grassy fields. The burrow is sunken perpendicularly, with short pas- sages leading to the cells, which are slightly inclined downwards and outwards from the main gallery. The walls of the gallery are rough, but the cells are lined with a mucus-like secretion, which, on hardening, looks like the glazing of earthen-ware. In Fig. 80 Mr. Emerton gives us a profile view of natural size of the nest showing the main burrow and the cells leading from it ; the oldest cell, containing the pupa (a) is situated nearest the surface, while those containing larvje (b) lie between the pupa and the cell (e) containing the pollen mass and egg resting upon it. The most recent cell (/) is the deepest down, and contains a freshly deposited pollen mass. At c is the begin- ning of a cell ; g is the level of the ground. The bees were seen at work on the 4th of May, at Salem, Mass., digging their holes, one of which was already six inches deep ; and by the 15th, hundreds of holes were observed. On the 28th of May, in unearthing six holes, eight cells were found to contain pol- APIARI^E. 145 len, and two of them a small larva. On the 29th of June six full-grown larvae were exhumed, and one about half-grown. About the first of August the larva transforms to a .pupa, and during the last week of this mouth the mature bees appear. In Halictus, which is a genus of great extent, the head is trans- verse, and flattish ; the mouth- parts are of moderate length, the tongue being very acute, with acute paraglossae half the length of the tongue, while the labial palpi are not quite so long as the paraglossae. There are three subcostal cells in the wings, with the rudiments of a fourth often present, and the second cell is squarish. The abdomen is ob- long ovate, with a longitudinal linear furrow on the tip in the female. In the males the body is longer ajiid the antennae more filiform and slender than usual in this family. The larvae are longer, -and with more acutely convex segments than in Andrena. The pupae differ much as the adult bees from Andrena, especially in the shorter mouth-parts. Halictus parallelus Say excavates cells almost exactly like those of Andrena ; but since the bee is smaller, the holes are smaller, though as deep. Mr. Emerton found one nest, in a path, a foot in depth. Another nest, discovered September 9th, was about six inches deep. The cells are in form like those of Andrena, and like them are glazed within. The egg is rather slender and much curved ; in form it is long, cylindrical, ob- tuse at one end, and much smaller at the other. The larva 10 Fig. so. 146 HYMENOPTERA. (Fig. 81) is longer and slenderer, being quite different from the rather broad and flattened larva of Andrena. The body is rather thick behind, but in front tapers slowly towards the head, which is. of moderate size. Its body is somewhat tuberculated, the tubercles aid- ing the grub in moving about its cell. Its length O O O C^ is .40 of an inch. On the pupa are four quite dis- tinct conical tubercles forming a transverse line just in front of the ocelli ; and there are also two larger, longer tubercles, on the outer side of each of which an ocellus is situated. Figure 82 represents the pupa seen from beneath. Search was made for the nests on July 16th, when the ground w r as very hard for six inches in depth, below which the soil was soft and fine, and over twenty cells were dug out. '"The upper cells contained nearly mature pupne, and the lower ones larvae of various sizes, the smallest being hardly distinguishable by the naked eye. Each of these small larvae was in a cell by itself, and situated upon a lump of pollen, which was of the size and shape of a pea, and was found to lessen in size as the larva grew larger. These young were probably the offspring of several females, as four mature bees were found in the hole." (Emerton.) The larva of an English species hatches in ten days after the eggs are laid. Another brood of bees appeared the middle of September, as on the ninth of that month (1864) Mr. Emerton found sev- eral holes of the same species of bee made in a hard gravel road near the turnpike. When opened, they were found to contain several bees with their young. September 2, 1867, the same kind of bee was found in holes, and just ready to leave the cell. Like Bombus, the females are supposed 4o hybernate, the males not appearing until late in the season. Like Andrena, these bees suffer from the attacks of Stylops, and according to Shuckarcl, an Ichneumon preys upon them, while certain spe- cies of Cerceris, Philantlms, and Crabro carry them off to store their nests with. Fig. 82. VESPARI^E. 147 In Cottetes the females, as Shuckard observes, resemble the workers of the Honey-bee, while there is considerable disparity between the sexes, the males being much smaller, the tongue and maxillae very short ; and the four-jointed labial palpi much shorter than the paraglossae. There are three subcostal cells, with the rudiments of a fourth. These bees form large colo- nies, burrowing in the earth eight or ten inches deep, lining their cells "at the farther end with a very thin transparent mem- branaceous coating, resembling goldbeaters' skin." The}' thus furnish six or eight cartridge-like cells, covering each with a cap, "like the parchment on a drum-head." Smith, from whom we have been quoting, states that Miltogramma, punctata, which is a Tachina-like fly, and the Cuckoo-bee, Epeolus variegatus^ have, in Europe, been reared from their cocoons. VESPARI^E Latreille, Wasps. In this family, which comprises about 900 species, the body is more attenuated, more cylindri- cal, with a harder and smoother tegument than in the Apiariw . In the species with densely populated colonies, such as Vespa and Polistes, there are workers which are often very numerous, while in Eumenes and Odynerus, etc., there are only males and females. The antennae are elbowed, the mandibles are large, stout ; the maxilhe and labium of varying length ; the maxil- lary palpi are six-jointed ; while on the labial palpi, which are four-jointed, there are well-developed paraglossae. The pro- thorax is prolonged on each side to the insertion of the wings which are long and narrow, and once folded longitudinally when at rest ; the fore pair have two or three subcostal cells ; the hind shanks and tibiae are smooth. The eggs, when first laid, are globular, soon becoming oval. The larvae of this family are soft, fleshy, with larger heads in proportion to the rest of the body, than in the Apiarice; the antennal tubercle, or rudimentary antennae, are more dis- tinct, and the mandibles are larger. The surface of the body is smoother in Vespa and Polistes, but more tuberculated in the solitary genera, Cktynerus and allies, while the end of the bod} r is more acute. As in the Apiarice the higher genera are social, building papery nests, while the lower are solitary and build cells of mud or sand in protected places. 148 HYMENOPTERA. In Vespa, the Paper Wasp, the ligula is squarish, with the paraglossae nearly as long as the tongue, the outer maxillary lobes rounded oval, half as long as the palpi, and the labial maxillae are scarcely longer than the tongue. The abdomen is broad at base, acutely conical. The nests are either with or without a papery covering, supported by a short pedicel. Such females as have hybernated, begin to make their cells in the early part of summer. Smith states that the soli- tary female wasp " begins by making three saucer-shaped re- ceptacles, in each of which she deposits an egg ; she then proceeds to form other similar -shaped receptacles, until the eggs first deposited are hatched and the 3'oung grubs require a share of her attention. From the circular bases she now be- gins to raise her hexagonal cells, not building them up at once, but from time to time raising them as the young grubs grow. (Proc. Ent. Soc., London, 1S58, p. 35.) "VVaterhouse states that the cells formed by the solitary fe- male early in the season appear " to be built entirely of glisten- ing, whitish, silk-like threads which I have little doubt are a secretion from the insect, all the threads being firmly attached together as if the}- had originally been of a glutinous nature." The cells formed later in the season by the workers, differ in consisting of masticated rotten wood. "Almost simultane- ously with the commencement of the cells, it appears that the nest-covering is commenced. At first it has the appearance of a miniature umbrella, serving to shelter the rudimentary cells." Plate 5, Fig. 3, shows a group of cells surrounded l^ one layer of paper, and the beginning of another. As the nest grows larger the cells are ar- ranged in galleries, supported by pedicels, and the number of layers in the outside covering greatly increases in number. While our common and largest species, Vesjm mactilnta Linn. (Fig. 83), and the yellow wasp, Fig- 83. V. arenaria Fabr., build papery nests consisting of several galleries, with the mouth of the cells directed downwards, the East Indian species, V. orientalis, VESPARLE. 149 builds its cells of clay, and, according to Waterhouse, "the work is exceedingly beautiful and true." Another species, according to Smith, makes its nest of sandy loam, the exterior being so hard that a saw used in opening one of its sides was blunted. The larva of Vespa arenaria is long and cylindrical, not so much curved as in Polistes. Its position in its cell corre- sponds to its form, as the cell is longer and narrower than that of Polistes. Each segment of the body is posterioi'ly some- what thickened, as is the lateral (plenral) ridge of the body. The tip of the abdomen is rather blunt, the last steruite be- ing large and transverse. The pupa is provided with a single tubercle on the vertex, where there are two in the Crabron- idce and Spliegidce. By the time the nest of V. arenaria is large enough to contain ten full-grown larvae, and has about fourteen cells in all, being about an inch in diameter, the occupants of the two or three central cells will have changed to pupae, and one wasp will have been excluded. In a nest of the same species two inches in diameter, there were a second brood of larvae. The outer row of cells were occupied by pupa?, while the central ones, emptied of the first brood, were filled with a second brood of larvae. Evidently as soon as an imago leaves its cell, the female 'deposits an egg therein, as very minute larvae were found occupying cells next to those containing large full-grown larvae. In comparing a number of pupae from a large nest, they will be found to be in all stages of perfection, from the larva which has ceased feeding, and is preparing to transform, to the imago, still veiled by its thin subimago pellicle. It is dif- ficult to draw lines between these stages. Also when com- pared closely side by side, it is difficult, if not impossible to find any two pupae just alike, the development proceeding very un- equally. Thus the limbs may be more perfect than the antennae, or certain parts may be less perfect in some than in others, while the limbs may be more highly colored like the imago. Like the bees, Vespa suffers from numerous parasites, includ- ing Rhipiphorous paradoxus, which is a beetle allied to Stylops, and Lebia (Drornius) linearis. The larva of Volucella is said 150 HYMENOPTERA. to feed on the Vespa-larvae, and Mr. Stone says that Anthomyia incana is also parasitic in Wasps' nests, while two species of Ichneumons, one of which is Anomalon vesparum, also in- fest the larvae. No parasites have been as yet detected in this country. The Hornet, F. crabro Linn., has, according to Mr. Angus, become domesticated about NCAV York. This and the smaller wasps are sometimes injurious by eating into ripe fruit, but the injury is more than counterblanced by the number of flies and other insects they feed their young with. Indeed, as Saussure states, the species of Vespa are more omnivorous in their .tastes than any other wasps. They live by rapine and pillage, and have obtained a worse repute than other insects more injurious. In spring and early summer they feed on the sweets of flowers ; but later in the season attack strawber- ries, plums, grapes, and other fruits, and often enter houses and there help themselves to the dishes on the table. They will eat raw meat, and then aid the butcher by devouring the flies that lay their eggs on his meats. The}' will sometimes destroy Honey- bees, attacking them on their return from the fields laden with pollen ; they throw themselves upon their luckless victims, and tear the abdomen from the rest of the bod}% and suck their blood, devouring only the abdomen. They fall upon flies and butterflies, and, biting off their wings, feet, and head, devour the trunk. In attacking insects they use only their powerful jaws, and not the sting, differing in this respect from the fossorial wasps. Saussure states that though wasps do not generally lay up food, yet at certain periods they do fill the cells with honey. The females feed their young with food chewed up and re- duced to a pulp. Saussure questions whether the larvae of one sex are not fed on animal and the other on vegetable food, since Iluber had shown "what a great influence the kind of o food exerts on the sex of Bees." But it is now known that the sexes of some, and probably all insects are determined before the larvae is hatched. I have seen the rudiments of the ovi- positor in the half-grown larva? of the Humble-bee, and it is most probable that those rudiments began to develop during embryonic life. It is far more probable that the sexual differ- ences are determined at the time of conception. VESPARI.E. 151 Westwood states that the larvae, which live head-downward from the reversed position of the comb, retain their position in the cell, while j-oung, by a glutinous secretion, and afterwards *"by the swollen front of the bod}' which fills the open part of the cell." "The female cells are mostly placed apart from those of the males and neuters, those of the males being often mixed, but in a small number, in the neuter combs. The egg state lasts eight days, the larva state thirteen or fourteen, and that of the pupa about ten. After the imago has been produced, one of the old workers cleans out the cell, and fits it for the reception of a fresh inhabitant. The upper tier of cells, being first built, serves for the habitation of the workers ; the females, being produced at the end of the summer, occupy the lowest tiers." When about to transform the larvsa spin a thin cover- ing, thus closing over the cell. In Polistes the paraglossae are slender, and a little longer than the long, or as in one instance noticed by us in P. Cana- densis, barrel-shaped ligula, which is split at the end ; the palpi are stouter, while the whole body is much longer than in Vespa ; the abdomen is subpedimculate, ami the thorax is rather ob- long than spherical, as in Vespa. The larva differs from that of Vespa in its much larger head, and shorter, more ovoid form of the body, which is dilated in front so as to retain the insect in its cell, while the tip is more acute ; the antennal tubercles are closer together ; the clypeus is more regularly triangular and more distinct, while the labrum is much larger and excessively swollen, as are the mouth-parts generally. The mandibles are bidentate, where in Vespa they are tridentate. The pupa differs from that of Vespa, besides the usual generic characters, in having the tubercle on the head smaller. The nests of Polistes (Plate 5, Fig. 4, nest of P. anmdaris Fabr., from Saussure) are not covered in by a papery wall as in Vespa, but may be found attached to bushes, with the mouth of the cells pointed downwards. While at Burksville Junction, Va., in the last week of April, I had an opportunity of watch- ing three species beginning their cells on the same clump of bushes. They all worked in the same method, and the cells only differed slightly in size. The cells were formed mostly of 152 HYMENOPTERA. crude silk, and the threads could be seen crossing each other, the same structure being observed at the top and bottom of each cell. In the three-celled nest of Polistes (Plate 5, Fig. 5, 5 a) first noticed April 29th, there were but two eggs deposited, the third cell being without an egg, and a little smaller, and the rim not so high as in the other two. The outer edge did not seam to be perfectly circular, though stated by AVater- house to be so in the incipient cells, for in some cases we de- tected two slight angles, thus making three sides, which, however, would be easily overlooked on casual observation ; as there are only two sides within, the cell, from being at its earliest inception hemispherical, or "saucer-shaped," becomes five, and subsequently six-sided, and thus from being cir- cular, it is converted by the wasps into a hexagonal cell. In some cells, perhaps a majority, both in this and the other spe- cies, the newly made rim of the small cells is thinner than the parts below, and slightly bent inwards ; thus being quite the re- verse of the thickened rim of the cells of the Hive Bee. It would seem that the Avasp plasters on more silk, especially on the angles, building them out, and making them more promi- nent, in order to complete, when other cells are added, their hexagonal form. The three cells are of much the same size and height when the third egg is laid, as we observed in another nest, that of Pollstes Canadensis (Linn.), built at the Defences of AVashington, near Munson's Hill, June 9th. Again, when one or two more cells have been added to the nest, and there are four or five in all (Plate 5, Fig. 6 ; 6 a, top view, in which there are four cells), two of them are nearly twice as large as the others, while the fifth has been just begun, and is eggless. The form of the two which run up much higher than the others is the same as that of the smaller and shorter ones, i.e. they are on one side nearly semicircular, and on the other, partly hexagonal, and the angular sides show a tendency to be even more circular than when the others are built around them, for the little architect seems to bring out the angles more prominently when carrying up the walls of the other cells. Thus she builds, as if by design, one and the same cell both by the "circular" and "hexagonal" methods, afterwards adopt- VESPARI.E. 153 ing only the latter, and if she devotes her attentions specially to plastering the corners alone, with the design of making the cell six-sided, then we mnst allow, contrary to Mr. Water- house's views, that the wasp builds the hexagon by choice, and not as the mere result of her blindly "working in segments of circles ;" for if our point be proved, and the most careful obser- vation of the wasp while at work is needed to prove it, then it may be shown that the wasp is a free agent, and can abandon one method of working at a certain stage of her work, and adopt a different mode of operating. The eggs are oval, pointed at the end, and glued to the in- side of the cell. They are situated midway from the top and bottom of the incipient cell, and placed ou the innermost sides, so that in a group of several cells the eggs are close together, only separated by the thin cellular walls. In a completed cell the egg is placed very near the bottom. For several days a Polistes Canadensis was engaged in build- ing its nest in my tent in camp near Washington. When first noticed on June 9th, there were three cells, two of which con- tained eggs; and it was not for two days, the llth, that the third cell was completed, and a third egg deposited in it. The wasp paid especial attention to strengthening the pedicel, going over it repeatedly for an hour or two with its tongue, as if lay- ing on more silken matter, and then proved the work by its swiftly vibrating antennre. It would often fly out of the tent, and on its return anxiously examine each cell, thrusting its head deep down into each one. It gradually became accustomed to my presence, but eventually abandoned the nest, without adding more cells. The others, while at work on the bushes, abscond- ed at my approach, and seemed very wary and distrustful, as if djsirous of concealing their abodes. Mr. Smith has found Trifj iitalys bi^ntstulatus to be a parasite on Polistes lanio Fabr. (P. Canadensis Linn.), from St. Salvador, S. A. Saussnre arranges the higher Vespidns into two parallel series. Vespa is offset by Chartergus and Nectarina ; lower down we find Tatua and Sjniffiea, while Polistes is offset by Polybia. These five genera are tropical, and in their habits, the general appearance of their nests, and in the number of individuals represent Vespa and Polistes of the temperate zone. The 154: HYMENOPTERA. genus Nectar I na is a short plump wasp, somewhat like Odyne- rus iu shape ; its distinguishing mark is the concealment of the postscutellirai by the scutellum. Nectarina mellifica Sa}-, of Mexico, builds a large nest externally like that of a wasp, but it is more irregular, and the papery covering consists of but one layer. The interior of the nest is very different, the galleries of cells, instead of being parallel, being arranged in concentric spheres. Charteryus has the tip of the clypeus slighted excavated, and an oval sessile abdomen. C. chartarius Olivier makes an ex- ceedingly thick tough nest, attached by a broad base to the bough of a tree, about twice as long as thick, and ending in a cone, pierced in the centre by the entrance which passes through the middle to the basal gallery ; the other galleries are formed by a continuation of the sides of the nest, and arrayed in a conical plane. In Tatua, the abdomen is pedicelled, but the petiole is not enlarged, and the abdomen itself is very regularly conical. T. morio Cuvier, from Cayenne, forms a nest like that of Charter- gus ; but the galleries form a flat floor, and each gallery has an entrance from tli3 outside of the nest, where in the latter there is one common entrance. Plate 5, Fig. 9, shows how the bases of the cells are laid out on the edge of a gallery. In Synoeca the peculiarly shaped abdomen is cordate and compressed. The curious nest of S. cyanea Fabr. is formed of a single layer of cells fixed against the trunk of a tree, and covered in with a dense covering made from the bark of dead trees. Some nests of Synoeca are three feet long. In the very extensive genus Polybia, which resembles Polistes in its general shape, the abdo- men is pedicelled, and the mandibles are four-toothed. The nests are somewhat like those of Chartergus, but much smaller. Sev- eral species occur in Mexico, and in Brazil the number of species is very great. In Apoica the abdomen is very long, and the third segment is as long as the second. Plate 5, Fig. 11. represents the nest of Apoica pallida Olivier, from Cayenne. It is unprotected, with a conical base, and with a single row of cells. In Icaria we have an approach to Polistes in the slender series of cells composing the nest, forming two or three rows VESPART^E. 155 only. Plate 5, Fig. 7, represents the nest of /. guttatipennis Saussure, from Senegal ; 8, ground plan of a similar nest. These wasps are mostly distinguished from Polybia by the petiole ending in a globular mass. Plate 5, Fig. 10, represents the elegant nest of Mischocyttarus kibiattts Fabr., from Cay- enne and Brazil, which consists of a few cells supported by a long pedicel. The wasp itself much resembles Polistes, but the petiole is very much longer. The remaining genera noticed here are solitary, building separate cells, and with only males and females. There are three subcostal cells in the fore wings, and the maxillte and labium are much elongated. In Eumenes the abdomen has a long pedicel, being sessile in Odynerus. While authors place Eumenes higher than Ody- nerus, we would consider the latter as a higher, more cepha- lized form, since the abdomen is less elongated, and the head is larger. In Odynerus the ligula is long, deeply forked at the slender extremity, while the slender paraglossee are shorter, ending in a two-toothed claw-like tip ; the maxillaa are slender, and the palpi have an elongated basal joint ; the clypeus is nearly circular, toothed on the front edge. The larva differs from those of the higher Vespa rice , in its more elongated head, the square clypeus, the unusually deep fissure of the bilobate la- brum, and in the larger tubercles of the body, as the larva is more active, turning and twisting in its cell, while feeding on its living food ; and in this respect it is more closely allied to the young Crabronidcv . In the pupa of 0. albftyhaleratus, the tip is more incurved than in the pupa of Vespa, so that the hind legs (tarsi) reach to the tip, and the abdomen is rounded ovate, Avhile in Vespa it is oblong. The cells (Plate 4, Figs. 13, 14) of Odynerus aJbopJtaleratus Sauss. have been detected like those of Osmia in a deserted gall of Diplolepis confluens, where several were found in a row, arranged around one side of the gall, side by side, with the holes pointing towards the centre of the gall. The cells are half an inch long, and one-half as wide, being formed of small pellets of mud, giving a corrugated, granulated appearance to the outside, while the inside is linsd with silk. 156 HYMENOPTERA. We have received from Mr. Angus deserted cells of Cera- tina in a syringa stem, in which we detected a pupa of an Odynerus, perhaps 0. leucomelas ; the cell was a little shorter than that of the Ceratina it had occupied. The cocoon of the Odynerus was of silk, and almost undistinguishable from the old cocoon of Ceratina. The wasp had dispensed with the necessity of making a mud cell. If future research shows that either this or any other species makes a mud cell or not at will, it shows the intelligence of these little "free-agents;" and that a blind adherence to fixed mechanical laws does not obtain in these insects. The larvae of Odynerus and Eumenes are carnivorous. I found several cells of 0. albophaleratus, June 22d, in the deserted nest of a Clisiocampa, which were stored with micro- lepidopterous larvae and pup*, still alive, having been para- lyzed by the sting of the wasp. The larvre of the wasp was short and thick, being, when contracted, not more than twice as long as broad ; the rings of the body are moderately convex, and the pleural region is faintly marked. Prof. A. E. Vcrriil has discovered the cells of an Odynerus at New Haven, forming a sandy mass (Plate 5, Fig. 12) attached to the stem of a plant. In Eumenes the lingua is very long, being narrower and more deeply divided than in Odynerus ; the second subcostal space of the wings is long and narrow, while in Odynerus it is triangular. The genus is easily recognized by the very long pedicel of the abdomen. Eumenes fraterna Say constructs a thin cell (Plate 5,* Fig. 15) of pellets of mud, and as large * EXPLANATION OF PLATE 5. Fie:. 1. Mouth of the tunnel of Aur/oclilorapurus; from Emerton. Fig. 2. Cells of Osmla pacifica ; communicated by Mr. Sanborn. Fig. o. Vertical section of nest of 1'espa with a group of primitive cells surrounded by one layer of paper, and part of another; from Saussure. Fig. 4. Nest of Po- listes annularis ; from Saussure. Fig. 5. Three primitive cells of Polistes; 5, top view of the same, one being eggless. The sides adjoining are angular. Figs. G and 6, a cell farther advanced, consisting of four cells, each containing an egg, and with the edges of the cells built up higher and more decidedly six-sided; original. Fig. 7. Cells of Icaria guttatipennis, showing that each cell is built up independently in regular hexagons. Fig. 8. Ground plan of a similar nest. Fig. 9. Ground plan, of cells of Tatua mono; from Smith. Fig. 1C. Nest of Afischoryttctrus labiatus ; from Saussure. Fig. 11. Nest of Apoica pallida ; from Saussure. Fig. 12= Nest of Odynerus birenimnculatus . Fig. 13. Nest of Odynerus albophaleratiis ; original. Fig. 14. Mud cell of Pelopceus flavipes ; original. Fig. lo. A row of spherical cells of Eumenes fraterna, with the female; from Harris. jriate a. A RHTT I TF, flTTTR'R OF W A S P S. CRABRONID^E. 157 as a cheriy. It is attached by a short stout pedicel to bushes, and the cavity is filled with the larvae of small moths. Raplnglossa odyneroides, from Epirus, described by S. S. Satmders, makes elongated cells in galleries in briars, storing them with the larvae of what he supposed to be weevils. The dark brown dense tough cocoon of a Chrysis was also found in the cells. In Masaris, which connects the Ve spar ice with the succeed- ing family, the wings are not completely folded when at rest ; there are but two subcostal cells ; the maxilla? are rudimen- tary ; and the antennae are clavate and eight-jointed. Masaris vespoides Cresson, inhabits Colorado Territory. CRABRONIDJE Latreille. Sandrwasps, Wood-icasps. In the more typical genera the head is remarkably large, cuboidal, while the clypeus is very short, and covered for the most part with a dense silvery or golden pile. The antennae are genicu- late, the long second joint being received, when at rest, in a deep frontal vertical groove ; the mandibles are large, and of even width throughout, and the mouth-parts are rather short, especially the lingua, which is often, however, well developed. There is only one subcostal cell, except in the PhilantJiince. The thorax is sub-spherical, and the abdomen is either short and stout, or more or less pedicellate. The forefeet are adapted for digging and tunnelling, the forelegs in the females being broad and flat, and in the males, which are supposed to do no work, they are sometimes, as in Thyreopus, armed with vexhillate expansions. The larva is rather short and thick, a little flattened on the under side, but much rounded above ; the segments are convex above, the thoracic segments differing from the abdominal seg- ments in not being thickened posteriori}* on each ring. They spin either a very slight cocoon, or a thin dense brown oval cylindrical case, generally reddish brown in color. The pupae have much the same character as the imago, with prominent acute tubercles above the ocelli. The members of this family afford, so far as we are ac- quainted with their habits, most interesting examples of the interdependence of structure and the habits of insects. Most 1,38 HYMENOPTERA. of the species are wood-wasps, making their cells in cy- lindrical holes in rotten wood, or enlarging nail-holes in posts, as is the case with C'rabro singularis, according to the observations of Mr. C. A. Shurtleff, thus adapting them to the requirements of their young. Other genera (Rhopalum pedicel- latum, Stigmus fraternus, and Crabro stirpicola) avail them- selves of those plants whose stem has a pith which they can readily excavate and relit for their habitations. The females provision their nests with caterpillars, aphida?, spiders, and other insects. This family is most difficult to classifj- ; it consists rather of groups of genera, some higher and some lower, though as a general rule those genera with pedunculate abdomens are the lowest in the series. In illustration, we regard Stigmus, with its elongated decephalized bod}', as inferior to Blepharipus, which again is subordinate to the more cephalized Crabro, where the body is shorter, the abdomen sessile, the anterior part of the body more developed headwards, while its nests are constructed more elaborately. The genus Psen, for the same reason, is lower than Cerceris, of which it seems a de- graded form. Some of the most useful characters in separating the genera of this family are to be found in the form of the clypeus, its sculpturing and relative amount of pubescence or hirsuties ; in the form and sculpturing of the propodeum (Newman), or tho- racico-abdominal ring of Newport ; while the tip of the abdo- men presents excellent generic and also specific characters, depending 6n its grooved or flattened shape. The species of this family are mostly found in the north temperate zone, being very abundant in North America and in Europe. The Pempnredoninse occur far north in abundance, while Cerceris occurs farthest towards the tropics. The subfamily Philanthince includes the three genera, PJti- lanthus, Eucerceris, and Cerceris. In Philanthus (Fig. 84, wing), the head is short, transversely suboval, the clypeus longer than broad, with the first joint of the abdomen nearly as broad when seen from above as the succeeding one. Our more com- mon form southward is Philanthus vertilabris Say (Fig. 85). In Europe P. apivorus provisions its nest with hone}^-bees. CRABRONIDvE. 159 84. Fijr. 83. Fig. SKa. Fig. 85. Cresson remarks that Eucerceris (Fig. 86, fore wing of male ; a, female) differs from Cerceris in the venation, which differs greatly in the two sexes. E. zonatus Say occurs in the west. The species of Cerceris (Fig. 87, wing) have transversely oblong heads, the front of the head is flattened and destitute of hairs, and the rings of the abdomen are contracted, the middle part being un- usually convex and coarsely punctured, while the basal ring is nearly one-half nar- rower than the succeeding ones. Cerceris deserta Say is our most com- mon form. In Europe some species are Fig. 87. known to store their nests with bees, and the larva: of Cur- cnlionidce and Buprestidce. Dufour unearthed in a sin- gle field thirty nests of C. bupresticida which were filled with ten species of Buprestis, comprising four hundred individuals, and none of any other genus. Cerceris tuberculata provisions its nest with Leucosomus ophthalmicus ; and C. tricincta with Clythra. In the subfamily CrabronincK, there is a great disparity in the sexes, the form of the females being the most persistent. In the male the head is smaller, narrow behind, with shorter mandibles, and a narrower clypeus ; the body is also much slenderer, especially the abdomen, and the legs are simple in Crabro, but in Thyreopus variously modified by expansions of the joints, especially the tibia. The species of Crabro (Fig. 88) are readily distinguished by the large cubical head, and the sharp mucronate abdo- minal tip of the female. The more typical form of this very extensive genus is Crabro sex-macidatus Say, so-called from the six yellow spots on the subpedunculate abdomen. According to Dr. T. W. Harris (MS. notes), this wasp was seen by Rev. Mr. Leonard, of Dublin, N. H., burrowing in decayed wood, June 10th. 160 HYMENOPTERA. Crabro singularis Smith, was discovered by Mr. C. A. Shurtleff boring in a post. In Thyreopus, the body is slender, and the forelegs are curiously dilated in the males, often forming a broad expansion, and so dotted as to present a sieve-like appearance, while the head is much shorter, being more transverse. T. latipes Smith is known by the broad, long, acute, mucronate, shield-like ex- pansion of the fore tibia, which is striped with black at the base. The species of Bhopahtm are usually blackish, without the gay colors prevalent in the genera before mentioned ; the legs are simple, and the abdomen is long and slender, with a long peduncle. The body of the larva is short and thick, tapering rapidly towards each extremity ; the segments are convex, those of the thorax especially being smooth, broad, and regu- larly convex, while the abdominal rings are provided with prominent tubercles. The tip of the body is quite extensible, and when protruded is subacute, terminating in a small knob- like body, formed by the last ring. The larvae of this genus differ from those of the Vesparice arid yip m?-/ce known to us by having a few hairs scattered over the body. In the pupa the antennae, in their natural position, do not quite reach to the second pair of trochanters, and reach only to the tip of the maxillary palpi. The tip of the abdomen is very acute and elongated unusually far beyond the ovipositor. On the head, between the ocelli and antenna 1 , are two very prominent, acute tubercles, and the abdominal segments are dentate on the hind edge. Thus both the larva and pupa would seem, b} T their anatomy, to be unusually active in their loose, illy-constructed cells, which do not confine their food so closely as in the other wasps, as the insects on which they prob- ably feed have a greater range in their rather roomy cells. April 18th we opened several stems grown in the open air, and found both larvae and pupae ; the latter in different stages of development. The cells were placed in the closely packed dust made by the larva of an JEgeria, or directly bored in the pith of the plants. There were six such cells, each with its inhabitant, within a space an inch in length, some laying cross- wise, others along the middle. The larva? spin but a very CRABRONIDJE. 161 slight cocoon, not at all comparable with that of Crabro ; the walls of the cell being simply lined with silken threads. Under other circumstances, i. e. where the cells are more exposed, it is not unlikely that a more elaborate cocoon may be spun. Mr. James Angus has bred numerous specimens of Rlwpa- lum pedicellatum Pack., from stems of the Rose, Corcorus, Ja- ponica, and Spiraea, grown in hot-houses at West Farms, N. Y. The larva is a quarter of an inch long. The following genera belong to the subfamily Pemphre- donince : The genus Stigmus, as its name indicates, may at once be known by the very large pterostigma, as well as the unusually small size of the species. The body of the larva is moderately long and slender, cylindrical, tapering slowly towards both ex- tremities. The rings are short, very convex, subacutely so, and the larva is of a beautiful roseate color. Stfgmus frater- nus Say burrows in the stems of the Syringa, of which speci- mens have been received from Mr. Angus with the lame and pupse. In Cemonus the front narrows rapidly towards the insertion of the mandibles, and there is a short triangular enclosure on the propodeum, while the abdomen is shorter and thicker than in Pempliredon, a closely allied genus ; the pedicel is also longer. The larvse of Cemonus inornattis Harris live in irregu- lar burrows in the elder, like those of Rhopalum from which they have been reared by Mr. Angus. They are known by the broad flattened head and body, serrate side and tergum of the body, and large, conspicuously bidentate mandibles, as well as by the peculiarly flattened abdominal tip. In Passalwcus the labrum is very prominent, while the man- dibles are very large, widening towards the tip, and in the com- mon P. mandibularis Cresson they are white, and thus very conspicuous. This species burrows in company with the other wood-wasps mentioned above in the stems of the elder and syringa. The cells are lined with silk. The wasps appear early in June. Their nests are tenanted by Chalcids. The female stores her cells with Aphides, as AVC have found them, abundantly in stems of plants received from Mr. Angus. The genus Psen seems to be a degraded Cerceris, but the 11 1G2 HYMENOPTET7.A. abdomen is peclicellecl, and differs from Mimesa, a still more slender-bodied genus, in having the tip of the abdomen more or less grooved, while in Mimesa it is flat and not grooved at all. Fsen leucopus Say has a dense silvery pile on the front of the head, with black antennae, and the pedicel is rather short. NYSSOXIDJE Leach. In this family the head is transversely longer and less cubical than in the preceding group ; the ver- tex is higher and more convex, while the front is narrow, the clypeus long and narrow, the eyes long and narrow, and the antennae are more clavate than in the Crabronidce, and the propodeum is sometimes armed with acute spines, while the enclosed space is smoothly polished or striated. The wings arc long and narrow, and the abdomen is sessile in the typical genera, where it is obconic, but clavate when pedicellate. In Trypoxylon the body is long, with a pedicellate clavate abdomen. In Europe "Mr. Johnson has detected it frequent- ing the holes of a post pre-occupied by a species of Odynerus, and into which it conveyed a small round ball, or pellet, con- taining about fifty individuals of a species of Aphis ; this the Odynerus, upon her return, invariably turned out, flying out with it, held by her legs, to the distance of about a foot from the aperture of her cell, where she hovered a moment, and then let it fall ; and this was constantly the case till the Trypoxylon had sufficient time to mortar up the orifice of the hole, and the Odynerus was then entirely excluded ; for although she would return to the spot repeatedly, she never endeavored to force the entrance, but flew off to seek another hole elsewhere." T. politum Say has purplish wings, and no enclosure on the propodeum. T. frigiclum Smith lives in the stems of Syringa, from which it has been reared by Mr. Angus. The thin, delicate cocoon is long and slender, enlarging slightly towards the anterior end. The genus Mellinus (belonging to the third subfamily, Mel- Jinince^ is known by its broad front, and slender antennae, and its pedunculate abdomen, while in Alyson, a slender- bodied genus, it is sessile. Mellinus bimaculatus Say has a black head, w r ith pale tipped antennae, and two ovate yellow spots on the abdomen. Alt/son oppo situs is black, with two NYSSONID^E. 163 yellow spots on the abdomen, which has the basal ring yel- lowish red in the female. The fourth subfamily is the Nyssonince, so named from Nys- son, a typical genus. The genus Gorytes is truty a mimetic form, closely simulat- ing the genus Odynerus, one of the Vesparicv. The front of the head is narrow, while the clypcus is larger than usual. The species are numerous, occurring late in the summer on the flowers of Spiraea. Gorytes flavicornis Harris is polished russet brown, with narrow yellow rings on the abdomen, the propo- deum is smooth and polished, and the basal ring of the abdomen is black. A species has been observed in Europe protruding her sting into the frothy secretion of Tettigoniie living on grass, and carrying off the insect to provision its nest with. Oxybelus is a short, stout, black genus, with whitish abdomi- nal spots, and stout spines on the thorax, while the sessile abdomen is distinctly conical. "Its prey consists of Diptera, which it has a peculiar mode of carrying by the hind legs the while it either opens the aperture of its burrow or else forms a new one with its anterior pair. Its flight is low, and in skips ; it is very active." (West wood.) Oxybelus emarginatus Say has two oval membranous appen- dages to the metathorax, and is -a common black species found abundantly on the flowers of the Virginia Creeper. In Nysson the body is a little longer, narrow compared with that of Oxybelus, while the terminal joint of the antennae is thickened, flattened, and excavated beneath. Nysson lateralis Say is dull black, with six light spots on the abdomen. The species of Stizus are of large size and easily recognized l\y their hirsute body, stout legs, triangular silvery clypens, and the high transverse vertex of the head. The propodeum has a faintly marked triangular enclosure. The species are very rapacious, paralyzing grasshoppers and other large insects with their formidable sting, and carrying them off to provision their nests. Professor S. Tenney has sent us a specimen of the Dog-day Cicada (C. canicularis) which Stizus sf)eciosus had thus stung. Mr. Atkinson has observed the same fact, and has found the deep burrows of this species, the hole being three- fourths of an inch in diameter. He has observed it feeding on sap running from a tree. 1G4 HYMENOPTEEA. The species of Larra are smaller, and differ from those of Stizus in the long, narrow, very prominent labrum, the shorter clypeus, broader front and longer abdomen, the tip of which is without the broad subtriangular area which is present in Stizus and the other genera of this family. Larra unicincta Say is black- ish, with a single reddish band on the second abdominal ring. BEMBECID^: Latreille. We have but two genera, Bembex and Monedida, which have large heads and flattened bodies, bearing a strong resemblance to Syrphus flies from their similar coloration. The labrum is very large and long, triangular, like a beak. The species are very active, flying rapidly about flowers with a loud hum. "The female Bembex burrows in sand to a considerable depth, burying various species of Dip- tera (S} r rphidie, Muscid;e, etc.), and depositing her eggs at the same time in company with them, upon which the larvae, when hatched, subsist. When a sufficient 'store has been collected, the parent closes the mouth of the cell with earth." u An anonymous correspondent in the Entomological Magazine, states that B. rostrata constructs its nests in the soft light sea-sands in the Ionian Islands, and appears to catch its prey (consisting of such flies as frequent the sand ; amongst others, a bottle- green fly) whilst on the wing. He describes the mode in which the female, with astonishing swiftness, scratches its hole with its forelegs like a dog. Bembex tarsata, according to Latreille, provisions its nests with BombyUi." (Westwood.) Dufour states that two Diptera, Panopea carnea and Toxophora fasciata, the latter allied to Systrophus, are parasites on Bem- bex. Mr. F. G. Sanborn has noticed the exceedingly swift flight of our common Bembex fasciata Fabr. on sandy beaches where it is found most abundantly. Monedida differs from Bembex in its slenderer body, more clavate antennae, and its shorter, very obtuse labrum. The body is smoother, and most generally more highly colored and more gaily spotted than in Bembex. Monedida Carolina Fabr. and M. 4-fasciata Say are common southwards of New England. LARRID/E Leach. Mr. F. Smith defines this family as having "mandibles notched exteriorly near the base ; the labrum con- L ARRIVE. 165 cealed, with a single spine at the apex of the intermediate tibire ; the abdomen is ovoid-conical." The genus Astata is a large hairy form, with long antennte and palpi and an elongated prothorax. Its spiny legs show its near relationship to the SpJiegidce.' Astata unicolor Say repre- sents the genus in this country. Tachytes is also of larger size than the following genus. It is covered with long dense golden short hairs, with a trap- ezoidal front. Tachytes aurulentus Fabr. is rare ; it frequents the flowers of the Asclepias, as we have found pollen masses at- tached to the spines of its legs. We figure (89) a tarsus of a wasp belonging probably to this genus, received from Mr. V. T. Chambers, showing the pollen masses of Asclepias at- tached to the spines. The genus Larrada "contains those species which have the marginal cell truncated at the apex and appendiculated, and three stibmarginal cells, the first as long as the two following ; .... the metathorax [propodeum] truncated posteriorly, elongate, the sides being generally parallel ; the mandibles are large and arcuate, with a tooth on their exterior towards the base ; abdomen ovate-conical, acuminate at the apex." Larrada argentata Beauv. is covered with silvery pile. It is a slender form, with short, nearly unarmed legs. A Brazilian species of Larrada, according to Mr. H. W. Bates, builds a nest composed apparently of the scrapings of the woolly texture of plants ; it is attached to a leaf, having a close resemblance to a piece of German tinder, or a piece of sponge. The cocoons were dark brown, and of a brittle consist- ency. The reporter, Mr. F. Smith, adds : "I am not aware of any similar habit of building an external nest having been pre- viously recorded ; our British species of the closely allied genus Tachytes, are burrowers in the ground, particularly in sandy situations ; their anterior tarsi are strongly ciliated, the claws bifid and admirably adapted for burrowing. On examin- ing the insect which constructed the nest now exhibited, I find the legs differently armed ; the anterior pair are not ciliated, 1GG HYMENOPTERA. and the claws are simple and slender, clearly indicative of a peculiar habit differing from its congeners, and how admirably is this illustrated in the nest before us?" SPHEGID./E Latreille. Smith defines this family as having 1 ' the posterior margin of the prothorax not prolonged back- wards to the insertion of the wings, and anteriorly produced into a neck, with the abdomen petiolated." The very fossorial legs are long and spiny, the posterior pair being of unusual length. The mandibles are large, curved, narrow, and acute, the base not being toothed externally, and the antennae are long and filiform. The species are often gaily colored, being ornamented with black and red, brown and red, or are entirely black, or blue. They love the sunshine, are very active, rest- less in their movements, and have a powerful sting. The sting of these and other wasps which store up insects for their young, penetrates the nervous centres and paralyzes the victim without depriving it of life, so that it lives many days. A store of living food is thus laid up for the young wasp. After being stung the caterpillars will transform into chrys- alids, though too weak to change to moths. Mr. Gueinzius, who resides in South Africa, observes that "large spiders and caterpillars became immediately motionless on being stung, and I cannot help thinking that the poisonous acid of Ilymen- optera has an antiseptic and preserving property ; for cater- pillars and locusts retain their colors weeks after being stung, and this, too, in a moist situation under a burning sun." These insects either make their nests in the sand, or, like the succeeding family, are "mud-daubers," building their cells of mud and plastering them on walls, etc. The tropical genus Ampulex is more closely allied to the preceding famity than the other genera. The species are brassy green. Dr. G. A. Perkins has described in the Ameri- can Naturalist, vol. 1, p. 293, the habits of a wasp, probably the Ampulex Sibirica Fabr., which inhabits Sierra Leone, and oviposits in the body of the cockroach. The dead bodies of the cockroaches are often found with the empty cocoon of the wasp occupying the cavity of the abdomen. A species of this genus, abundant at Zanzibar at certain sea- SPHEGIDYE. 1G7 sons, was frequently observed by Mr. C. Cooke to attack the cockroach. The cockroach, as if cowed at its presence, im- mediately yields without a struggle. The Ampulex stings and paralyses its victim, and then flies away with it. Chlorion is closely allied, containing blue and metallic green species, often Avith golden yellow wings. Chlorion cyaneum Dahlb., a blue species, is found in the Southern States. The genus Priononyx "differs from the genus Sphex in hav- ing the claws quadridentate beneath at their base ; the neura- tion of the wings and the form of the abdomen are the same as in Hai'pactopus" which is found only in the tropics and Aus- tralia. Priononyx Thomce is found from South Carolina to Brazil, including the AVest Indies. The genus Spliex is quite an extensive one. The head is as Avide as the thorax ; the antennae are filiform, mandibles large and acute, bidentate within, the teeth notched at their base, forming a rudimentary tooth, the apical tooth being acute. The thorax is elongate-ovate, truncated behind, with a trans- verse collar (prothorax). The fore wings have one marginal and three submarginal cells ; the marginal cell elongate, rounded at its apex ; the first submarginal cell as long as the two following. The abdomen is pedun- culated, conic-ally ovate, and the an- terior tarsi are cili- ated in the females. Sphex iclmeumo- nea Linn. (Figure 90) is a large rust- red species, with a dense golden pu- Fig. 90. bescence. It is common from Massachusetts southwards. In the last week of July, and during August and early in Sep- tember, we noticed nearly a dozen of these wasps busily en- gaged in digging their holes in a gravelly walk. In previous seasons they were more numerous, burrowing into grassy 1G8 HYMEXOPTEEA. banks near the walk. The holes were four to six inches deep. In beginning its hole the wasp dragged away with its teeth a, stone one half as large as itself to a distance of eight inches from the hole, while it pushed away others with its head. In beginning its burrow it used its large and powerful jaws almost entirely, digging to the depth of an inch in five minutes, com- pleting its hole in about half an hour. After having inserted its head into the hole, where it loosened the earth with its jaws and threw it out of the hole with its jaws and fore legs, it would retreat backwards and push the dirt still farther back from the mouth of the cell with its hind legs. In cases where the farther progress of the work was stopped by a stone too large for the wasp to remove or dig around, it would abandon it and begin a new hole. Just as soon as it reached the required depth the wasp flew a few feet to the adjoining bank and falling upon an Orchelimum vulgare or O. gracile, stung and paralyzed it instantly, bore it to its nest, and was out of sight for a moment, and while in the bottom of its hole must have deposited its egg in its victim. Reappearing it be- gan to draw the sand back into the hole, scratching it in quite briskly by means of its spiny fore tarsi, while standing on its two hind pairs of legs. It thus threw in half an inch of dirt upon the grasshopper and then flew off. In this way one Sphex will make two or three such holes in an afternoon. The walk was hard and composed of a coarse sea-gravel, and the rapidity with which the wasp worked her way in with tooth and nail was marvellous. Sphex tibialis St. Fargeau is a black, stout, thick insect. Mr. J. Angus has reared this species, sending me the larva; in a cavity previously tunnelled by Xylocopa Virginica in a pine board. The hole was six inches long, and the oval cylin- drical cocoons were packed loosely, either side by side, where there was room, or one a little in advance of the other. The interstices between them were filled with bits of rope, which had perhaps been bitten up into pieces by the wasp itself ; while the end of the cell was filled for a distance of two inches with a coarse sedge arranged in layers, as if rammed in like gun- wad- ding. The cocoons are eighty to ninety hundredths of an inch long, oval lanceolate, somewhat like those of Pompilus. They SPHEGID^E. 1G9 consist of two layers, the outer very thin, the inner tough, parchment-like. The larvae hybernate and turn to pupae in the spring, appearing in the summer and also in the autumn. The larva is cylindrical, with the pleural ridge prominent, and with no traces of feet ; the head, which is small and not prominent, and rather narrow compared with that of Pelopoeus, is bent inwards on the breast so that the mouth reaches to the sternum of the fourth abdominal ring. The posterior half of each ring is much thickened, giving a crenulated outline to the tergum. The abdominal tip is obtuse. Spliex Lanierii Guerin, according to Smith (Proceedings of the Entomological Society of London, Feb. 7, 1859), con- structs its nest of a cottony substance, filling a tunnel formed by a large curved leaf. The species of the genus are sup- posed to burrow in the ground, and the two cases above cited show an interesting divergence from this habit. Mr. Smith adds, that in "the Sphex which constructs the nest in the rolled leaf, the anterior tarsi are found to be very slightly ciliated, and the tibiae almost destitute of spines, thus affording another instance proving that difference of structure is indica- tive of difference of habit." The genus Pelopwus is of a slighter form than in Sphex, the body being longer and slenderer ; the clypeus is as broad as long, triangular above, in front convex, or produced and end- ing in two teeth. The outer costal cell is lanceolate oval, the second subcostal cell subtrapezoidal, being widest above ; it is also somewhat longer than broad. The first median cell is very long and narrow, much more so than usual. The pedicel of the abdomen is long, the first joint in the male being often as long as the remainder of the abdomen. The larva of P. cceruleus Linn, is much like that of Sphex, having a cylindrical body with the rings thickened posteriorly. It differs from that of Pompilus in its longer and narrower head, the short broadly trapezoidal clypeus, and the distinctly marked exserted labrum. The mandibles are long and tridentate. The pupa (of P. flavipes) differs from that of the Ve spar ice, in having the head more raised from the breast ; the palpi are not partially concealed, as they may be easily seen for their whole length. The long curved mandibles cover the base of the 170 HYMENOPTERA. maxillae and lingua, and the antennae reach to the posterior coxae. The maxillae are slender, not reaching to the tip of the labium. The female usually provisions her cells (Plate 5, Fig. 14) with spiders. The cells are constructed of layers of mud of unequal length, and formed of little pellets placed in two rows, and di- verging from the middle. They are a little over an inch long, and from a half to three-quarters of an inch wide, and are some- what three-sided, the inner side next the object, either stone- walls or rafters, to which it is attached, being flat. As the earthen cells sufficiently protect the delicate larvae within, the cocoons are veiy thin, and brown in color. The cells of Pelopwus Jlavqyes from Brownville, Texas, col- lected by an United States officer and presented to the Boston Society of Natural History, contained both spiders and numer- ous pupae of a fly, Sarcophaga nudipennis Loew (MS) which is somewhat allied to Tachina. These last hatched out in mid- summer a few days before the specimens of Pelopaeus. It is most probable that they were parasitic on the latter. These specimens of P. flavipes were more highly oiliamented with yel- low than in those found northwards in the Atlantic States, the metathorax being crossed by a broad j-ellow band. The genus Ammopldla is a long slender form, with a petio- late abdomen, the tip of which is often red. The petiole of the abdomen is two-jointed, and very long and slender, being longer than the fusiform part. In the males the petiole is in some species much shorter. The wings are small, with the apex more obtuse than usual ; the second subcostal cell is pentag- onal, and the third is broadly triangular. Westwood states that "the species inhabit sandy districts, in which A. sabulosa forms its burrow, using its jaws in bur- rowing ; and when they are loaded, it ascends backwards to the mouth, turns quickly around, flies to about a foot's distance, gives a sudden turn, throwing the sand in a complete shower to about six inches' distance, and again alights at the mouth of its burrow." "Latreille states that this species provisions its cells with caterpillars, but Mr. Shuckard states that he has observed the female dragging a veiy large inflated spider up the nearly per- pendicular side of a sand-bank, at least twenty feet high, and 171 that whilst burrowing it makes a loud whirring buzz ; and, in the Transactions of the Entomological Society of London, he states that he has detected both A. sabulosa and A. Inrsuta dragging along large spiders. Mr. Curtis observed it bury the caterpillars of a Noctua and Geometra. St. Fargeau, how- ever, states that A. sabulosa collects caterpillars of large size, especially those of Noctnae, with a surprising perseverance, whereas A. arenaria, forming a distinct section in the genus, collects spiders." (Westwood.) Ammopliila cementaria Smith, and A. urnaria King, are the more common species in this country ; they are red and white, while A. luctuosa Smith is a black, shorter, stouter, more hirsute species. They may all be seen flying about hot sandy places, and alighting near wells and standing water to drink. O ^ * ' POMPILID^E Leach. In this family the body is oblong, the sides often compressed, and the head shorter, when, seen from above, being more trans- versely ovate than in the preceding family. The antennas are long, not geniculate, and in the males are stouter and with shorter joints than in the females. The eyes are narrow oval, and the maxillary palpi are six, and the labial palpi four-jointed. The prothorax is ex- tended on the sides back to the base of the wings, Fl - 91- which latter are large and broad, the fore pair having three subcostal cells. The legs are very long and slender, with thick slender spines. The Pompilidce, of which about seven hun- dred species are known, have a wide geographical range, from the temperate zone to the tropics. Like the Sphegidm, they oviposit in the body of other insects, storing their nests, usually built in the sand, with spiders and caterpillars. The head of Pompilus (Fig. 91) is a little longer, seen from 172 HYMENOPTEEA. above, than in the other genera ; the front of the head is about a third longer than broad. The antennae are long and fili- form and sometimes crenulate, as in Figure 91 a, in the males ; the mandibles are stout, broad, sabre-shaped, being much curved, with low flattened teeth, and the maxillary palpi are longer than the labial palpi. The wings are rather broad, with the three subcostal cells O ' tying in a straight row. The abdomen is slightty com- pressed, and equals in length the remainder of the body. The sting is very large and formidable, and ex- cessively painful, benumbing the parts it enters. They 91 a ' are exceedingly active, running and flying over sandy places like winged spiders. There are about five hundred species of this genus described. They are usually shining black or deep bluish black, with Fig. 92. smoky or reddish wings, and sometimes a reddish abdominal band. This genus is interesting, as affording in its form a mean between the globular thorax and short body of the Apia rice and the elongated body of the Iclineumonidce. The Pompilus formosus Say (Fig. 92), called in Texas the Tarantula-killer, attacks that immense spider the Mygale Hentzii, and, according to Dr. G. Lincecum (American Naturalist, May, POAIPILID2E. 1 7^ J. t fj Fig. 93. Fig. 94. 1867), paralyzes it with its formidable sting, and inserting an egg in its body, places it in its nest, dug to the depth of five inches. There is but a single brood, produced in June, which is killed off by the frosts of November. This species feeds in summer "upon the honey and pollen of the flowers of the Elder, and of Vitis ampelopsis, the Virginia Creeper ; but its favorite nourishment is taken from the blossoms of Asclepias quadri folium" (Lincecum.) P. cylindricus Cresson (Fig. 93, wing) is one of our smallest species, being from three to five lines long. It occurs in the South and West. P. arctus Cresson (Fig. 94, wing) in- habits Colorado Territory. P. Marice Cresson (Fig. 95, ? enlarged) is a beautiful and rare species found in Pennsylvania. The genus Priocnemis is characterized by the two hind pair of tibire being serrated ($, Fig. 96, a, wing ; &, pos- terior leg ; c, anterior leg), and by the want of spines on the an- terior legs. P. unifasciatus Say is a wide-spread species and readily recognized by the deep black color of the body, the yellow an- tennae and the large yellow spot at the tip of each anterior wing. The 'genus Agenia (Fig. 97, a, wing ; &, posterior leg) differs in having smooth legs. A. brevis Cres- Fig. 96. son (Fig. 98, wing) is a little spe- cies found in Georgia. A. congruus Cresson (Fig. 99, wing) was captured in West Virginia ; and A. acceptus Cresson (Fig. 100, wing) in Georgia. The genus Notocyphus (Fig. 101, ?, wing) is found in Brazil and Mexico. Planiceps (Fig. 102, Fig. 95. 174 HYMENOPTEEA. b -, Fig. 97. cies, which builds its nest in fields. J. Fig. 99. Fig. 101. wing) contains a few species, of which P. niger Cresson, an entirely black species, is found in Connecticut. Aporus (Fig. 103, wing) contains a single American species, A. fasciatus Smith, taken in North Carolina. From Mr. F. G. Sanborn we have re- ceived the larva and cocoon of Pompilus funereus St. Farg., a small black spe- The larva is short and broad, with the lateral region rather prominent, and the tip of the abdomen rather acute. It differs from Pelopaeus in its stouter, rather flat- tened body, and thickened segments, though as our specimen is preserved in alcohol these characters may have be- come exaggerated. It more nearly re- sembles Pelopaeus in its transverse clypeus, thin bilobate labrum, and the stout mandibles, which are, however, much stouter than in Pelopaeus, while the whole head is shorter, broader, and rounder. It is probable that this pecu- liar form of the head (which as in Sphex is bent beneath the breast), together Fig. 103. with the broad transverse clypeus, and broad, short, bilobate, thin, transparent labrum, and especially the unidentate short broad mandibles are family characters, sep- arating the larvae of this group from those of the Sphegidce . The cocoon is ovate, long, and slender, much smaller at one end than the other, not being so regularly fusiform as in Sphex. Ceropales differs from the foregoing gen- era in its broad head, its much shorter ab- domen ; and also in the eyes being a little excavated, in the depressed labium, the narrow front, which dilates above and below the middle, and in the greatly elongated hind legs, gen- erally banded with red or whitish. Ceropales bipunctata Say is generally distributed throughout the United States. It Fig. 102. Fig. 98. SCOLIAD.E. 175 - 104 ' is easily recognized by the black body and legs, and red pos- terior femora, and is six lines long. C. RoUnsonii Cresson (Fig. 104, J) is an elegant species found in West Virginia. An allied genus is Mygnimia (Fig. 105, wing) containing M. Mex- icana Cresson and M. us- tulataDaklb., two Mexican species. In the genus Pepsis (Fig. 106, wing) the max- illary and labial palpi are of equal length. The spe- cies are large, some of them being among the lar- gest of Hymenoptera, and are generally indigo-blue in color. Pepsis lieros Dahlbom is found in Cuba ; it is two inches long. P. cyanea Linn., which is blackish-blue, with blue abdomen and wings, the latter reddish at the apex, has been described by Beauvois from the United States, while P. elegans St. Farg. also occurs in the Southern States. P. formosa Say affords another example of a species Fig. 106. common to both sides of the Rocky Mountains, as it has been found both in Texas and Cal- ifornia. It is black, with bluish or greenish reflections, with bright fiery red wings, and is thirteen to eighteen lines long. SCOLIAD^E Leach. This family forms a group very easily distinguished from the Bembecidce or Chrysididce, as well as the Pompilidce, by the broad front, the small indented eyes, and the great sexual differences in the antennae, those of the male being long and slowly thickened towards the tip, while in Fig. 105. 17G HYMENOPTERA. the female they are short, thick, and elbowed on the second joint. The clypeus is large, irregularly quadrilateral, becom- ing shorter in the lower genera, and the labrum is small, scarcely exserted, while the mandibles are, in the female es- pecially, large and broad. The prothorax is very square in front. In the fore-wings are three subcostal spaces. The abdomen in the typical genus (Scolia) is broad and flat, longer than the rest of the body. The abdomen of Mutilla approaches that of the Chrysididce in having the second ring much en- larged over the others. The males usually have the anal stylets very prominent, while the sting of the female is very powerful. The body and legs are generally very hirsute, and the first tarsal joint is as long as the tibiae.. The genus Sapyga is easily recognized by its smooth slender body, being ornamented .with yellow, with transverse bands on the abdomen. The head is long, very convex in front, and the anteunre are clavate ; the prothorax is very broad, giving an oblong appearance to the thorax. The legs are slender and smooth. It is said to be parasitic, laying its eggs in the -cells of Osmia. Sapyga Martinii of Smith is found northward. The species of Scolia are often of great size, being black and very hirsute, with the labium composed of three linear di- visions ; the abdomen alone being banded or spotted with yellow on the sides. They are found in the hottest places about strongly scented flowers. In Europe, Scolia bicincta "makes its burrows in sand-banks, to the depth of sixteen inches, with a very wide mouth ;" and it is probable that the nest is stored with grasshoppers. .Scolia quaclrimaculata Fabr. is found in the Middle and Southern States. The larva of Scolia flavifrons was found by Passerini to live in the body of the lamellicorn beetle, Oryctes nasicornis. In Madagascar, Scolia oryctoplmga lives on Oryctes simia, according to Coquerel. Professor Sumichrast states that at Tehuacan (Department of Puebla) the Scolia Azteca Sauss. is very common ; and is particularly abundant in the leather tanneries, which leads him to think that the females of this species also deposit their eggs under the epidermis of the larva which abounds in the tan. Tiphia is black throughout and rather hirsute. The antennae MUTILLAEI^E. 177 are shorter than in Scolia or Myzine ; the clypeus is also shorter, while the prothorax is longer. In the fore-wings the outer cos- tal cell is short, broad, angulated, oval ; and of the two sub- costal cells, the outer one is broad and triangular, twice as long as broad, while the first median cell is regularly short rhom- boidal, much more so than in the other genera. The females, according to Westwood, "make perpendicular burrows in sandy situations, for the reception of their eggs ; but the pr-ecise food stored up for the larvae has not been ob- served." TipMa inornata Say is a common species with us, and flies low over sandy places early in the season. The short oval head, the large eyes, short meso-scutum, large meso-scutellum, and the flattened, rather smooth body, characterize the genus Myzine. The females are very different from the males, the two sexes being for a long time considered as separate genera. The female, especially, differs in the great length of the square prothorax, which is very broad and convex in front. In the male the eyes are lunate, while in the female they are small, entire, and remote. In its general form the fe- males much resemble Scolia, while the males are long and nar- row, with broad yellow bands, especially on the abdomen, and a large exserted sting-like organ. Myzine sexcincta Fabr. is seen from New England southwards, flying low over hot sandy places. The genus Elis is closely allied. Sumichrast (American Nat- uralist, vol. 2), surmises that Elis costalis St. Farg. lives on certain Scarabeeides, which undergo their metamorphosis in the formicary of CEcodoma in Mexico. MUTILLARI.E Latreille. This interesting family is character- ized by the females alone being wingless, though Morawitz says that wingless males occur in two species ; and by the absence, generally, of the three ocelli. In Mutilla and Myrmosa the thorax is still high, compressed, and oblong cuboidal, and ex- cept in the closely united tergal pieces the females do not greatly recede from the type of the winged males. The species are very equal in size, are black, or black and red, and either smooth or hirsute. The antennae arc inserted low down on the front, the clypeus being very short and broadly ovate (especially in Myrmosa), 12 178 HYMENOPTEKA. or it is indented, as in Mutilla. The tongue is shorter than usual. The sides of the thorax contract in width, both before and be- hind. The meso-scutum is squarer than usual, while the meso- scutellum is mu'ch narrower and longer, and the propodeum is squarely truncated behind, thus presenting a full convex surface. The abdomen is not much longer than the rest of the body, be- ing shorter than usual. In all these characters this family shows its affinities to the Ants. The wings are very dissimilar in the different genera. In Myrmosa the neuratioii closely approaches that of Sapyga, while in the larger, more acute primaries of Mutilla, and especially in the short outer costal cell, and short open pterostigma, the latter genus differs from the others. The male of /Sderoderma closely mimics the Procto- trypidce, the veins of the wings being absent, -while the form of the head and abdomen also reminds us of some genera in that family. The wingless female is very different, having more of the form of Mutilla, with a large oblong head and long acutely conical abdomen. The species are minute and rarely met with. S. contracta Westwood is found in "Carolina." In the female Methoca the eyes are very long, and the seg- ments of the abdomen are widely separated, much as in the ants. Methoca Canadensis Smith is shin- ing black, and slightly villose. The species of Myrmosa may be known by the very short clypeus, the broad ver- tex, and the rings of the abdomen of the male being unusually contracted. The abdomen of the female is cylindrical, about twice as long as broad, and thickest on the second ring. The rings are densely hirsute on the hinder edge. Myrmosa unicolor Say (Figs. 107, male ; 108, female) is widely di.3tributed. We have taken this species in Maine, while sex- ually united, early in June. The wingless female is like an ant, and is pale reddish on the thorax and basal ring of the abdomen, and the antennae and feet are concolorous, while the head and remaining abdominal rings are much darker. It is .20 inch long. The male is .28 inch long and entirely black. Fig. 107. I ' Fiff. 108. FOKMICARI2E. 179 The genus Mutilla is a very extensive one, and enjoys a wide geographical range. It is throughout stouter than Myrmosa, the head is more cubical, and the thorax and abdomen is shorter, the tip of the latter being somewhat truncated. The wingless female closely resembles, both in its form and motions, a worker ant. The body is coarsely granulated and either naked or densely hirsute, and of a scarlet, black, or pale red, or brown-black color. The females are found running in hot sandy places, and hide themselves quickly when disturbed, while the males frequent flowers. Mutilla ocddentalis is a large species. It is of a beautiful scarlet color and is armed with a very powerful sting. According to Profes- sor A. E. Verrill this species was found by him, at New Haven, to construct deep holes in a hard beaten path, storing its nest with insects. This species is also said by rig. 100. Kirby to be very active, "taking flies by surprise." (West- wood.) Mr. Verrill noticed that this insect makes a slight creaking noise. The larvae of M. Europcea are said to live parasitically in Humble-bees' nests. Jlutilla fcrrugata Fabr. (Fig. 109) is found frequently in. New England. FORMICARI.E Latreille. The family of ants would seem naturally to belong with the truly fossorial Hymenoptera, both from their habits and structure. Both males and females are winged, but the males are much smaller than the females, while the wingless workers are smaller than the males. In these wingless forms the segments of the thorax become more or less separated, making the body much longer and slenderer, and less compact than in the winged nor- mal sexual forms, the prothorax being more developed than in the males and females. The workers often consist of two forms : one with a large cubical head, or worker major, some- times called a soldier, and the usual small-headed form, or worker minor. The head is generally triangular. The eyes are large in the males, smaller in the workers, and in those of some genera (Ponera, Typhlopone, etc.) they are absent ; while in the 180 HYMENOPTERA. workers the ocelli are often wanting, though present in the winged individuals of both sexes. The antennae are long, slender and elbowed. The mandibles are stout, and toothed, though in those species that do not themselves labor, but en- slave the workers of other species, they are unarmed and slender. The maxillary palpi are from one to six-jointed, and the labial palpi two to four-jointed. The fore-wings usually have but a single complete subcostal (cubital) cell. The sting is often present, showing that in this respect as well as their fossorial habits the ants are truly aculeate Ilymenoptera. The larva is short, cylindrical, with the end of the body obtuse. The rings of the body are moderately convex. The head is rather, small and bent upon the breast. The larvae are fed by the workers with food elaborated in their stomachs. The larvae of the stingiess genera usually spin a delicate silken cocoon, while those of the aculeate genera do not. Both Latreille and Westwood, however, state that sometimes, as in Formica /c, of Europe, the pupae are naked, and at other times enclosed in a cocoon. The colonies of the different species vary greatly in size. In the nests of Formica sanguinea the number of individuals is very great. The history of a formicarium, or ant's nest is as follows : The workers only (but sometimes the winged ants) hibernate, and are found early in spring, taking care of the eggs and larvae produced by the autumnal brood of females. In the course of the summer the adult forms are developed, swarming on a hot sultry day. The little yellow ants, abundant in paths and about houses in Xew England, generally swarm on the af- ternoon of some hot day in the first week of September, when the air is filled towards sunset with myriads of them. The females, after their marriage flight in the air, may then be seen entering the ground to lay their eggs for new colonies, or, as "Westwood states, they are often seized by the workers and retained in the old colonies. Having no more use for their wings they pluck them oft', and may be seen running about wingless. According to Gould, an early English observer, the eggs destined to hatch the future females, males and workers, are deposited at three different periods. The nests of some species of Formica are six feet in diameter FOKMICARLE. 181 and contain many thousand individuals. Ants also build nests of clay or mud, and inhabit hollow trees. They enjoy feeding upon the sweets of flowers and the hone} r of the Plant- lice, which they domesticate in their nests. Several species of beetles, including some of the Stapliylinidce , take up their abode in ants' nests. Ants are useful as scavengers, feeding on decaying animal matter. A good method of obtaining the skeletons of the smaller animals, is to place them on a densely populated ant-hill. The habits of the ants, their economy and slave-making habits, are described in the works of Huber, La- treille, and Kirby and Spence. Upwards of a thousand species of ants have already been described ; those of this country have still to be monographed. The first group of this extensive family consists of Dorylus and its allies, and Formica and the neighboring genera, all of which are distinguished by having onby the first abdominal seg- ment contracted, while in the second group (Myrmicarice) , the two basal rings are contracted into knot-like segments. The genus Dorylus was, by Latreille, Klug, and others, in- cluded in the Mutillari ce . The head is very short, the ocelli are large and globular. The thorax and abdomen are elongated, the last is cylindrical, with a small, round, basal joint. The legs are short, with broad compressed femora and feather-like tarsi. In the wings the outer subcostal cells are wanting. The females are not yet known. Mr. F. Smith says that Dorylus was found by Hon. "W. Elliot to live in the man- ner of ants, under the stone foundation of a house in India. The society was very numerous. The difference in size of the male and worker is very remarkable. The males are of large size and are found in tropical Asia and Africa. Typhlopone is an allied genus. T. pallipes Haldeman is found in Pennsylvania. To the genus Anomma belong the Driver-ants of Western Africa. They march in vast armies, driving everything before them, so formidable are they from their numbers and bite, though they are of small size. They cross streams, bridging them by their interlocked bodies. Only the workers are known. Two species only, A. Burmeisteri Shuckard, and A. arcens Westwood, are described from near Cape Palm as, TVest Africa. 182 I1YMENOPTEKA. The genus Ponera is found distributed throughout the tropics. The females and workers are armed with spines ; the abdomen is elongated, the segments more or less diminished O i O in size, the first comparatively large and often cubical. The legs are slender. P. ferniginea Smith is a Mexican species. The allied genus Odontomachus springs like some leaping spiders. It uses for this purpose its unusually long mandibles, which are bent at right angles. 0. darns Roger lives in Texas. Formica includes the typical species of ants. Over two hun- dred species of this genus have been already described. The body is unarmed. The abdomen is short, oval or spherical, the scale-like first segment being lenticular in form, with a sharp upper edge. The subcostal cell of the fore-wings ends in a point. Formica sanguinea Latr. is one of our most abundant species, making hillocks of sand or clay, according to the nature of the ground. From the formicary walks, and underground galleries, radiate in all directions. This species has been ob- served making forays upon each others colonies. We have found a variety of this species in Labrador, where it is com- mon. It does not throw up hillocks, but tunnels the earth. This species has been observed in Europe by P. Iluber, to go on slave expeditions. They attack a " negro-colon}- " be- longing to a smaller black species, pillaging the nest, and carry- ing off merely the larvae and pupae. The victors educate them in their own nests, and on arriving at maturity the negroes take the entire care of the colon}'. Pohjergus rvfcscens is also a slave- making ant, and "Latreille very justly observes that it is physi- cally impossible for the rufescent auts (Polyergus rufesccns), on account of the form of their jaws, and the accessory parts of their mouth, either to prepare habitations for their family, to procure food, or to feed them." Formica sanguinea sallies forth in immensely long columns to attack the negro ant. Hu- ber states that only five or six of these forays are made within a period of a month, at other seasons they remain at peace. Iluber found that the slave-making Polycrgus nifescens when left to themselves perish from pure laziness. They are waited upon and fed by their slaves, and when they are taken away, their masters perish miserably. Sometimes they are known to labor, and were once observed to carry their slaves to a spot chosen FORMIC ARI^E. 183 for a nest. The F. sanguinea is not so helpless, "they assist their negroes in the construction of their nests, they collect their sweet fluid from the Aphides ; and one of their most usual occupations is to lie in wait for a small species of ant on which they feed ; and when their nest is menaced by an enemy they show their value for these faith- ful servants, by carrying them down into the lowest apartments, as to a place of the greatest security." (Kirby.) Pupae of both of the slave- making species were placed in the same formicary by Huber, where they rig. no. were reared by the "negroes," and on arriving at maturity " lived together under the same roof in the most perfect amity," as we quote from Kirby. Darwin states that in England, F. sanguinea does not enslave Other species. In this country Mr. J. A. Allen has described in the Proceedings of the Essex Institute, vol. 5, 1866, a foray of a colony of F. sanguinea upon a coloiry of a black species of Formica, for the purpose of making slaves of them. Formica Pensylvanica, our largest species, is found in oaks and decay- ing trees, while F. herculanea Latr. burrows in the earth, its hole opening beneath stones and sticks. Gould, who wrote in 1747, states that there are two sizes of workers of the common European Formica rufa, and flava; one set of individuals exceeding the other by about one-third. Kirby states that in his specimens "the large workers of For- mica rufa are nearly three times, and of F. jlava, twice the size of the small ones." Mr. E. Norton describes F. fulvacea (Fig. 110, worker minor), and also Tapinoma tomentosa (Fig. Ill, worker major; antennae broken off), from Mexico. The tropical genus Polyrliacliis includes, according to Smith, all those species that closely resemble Formica, but which in. 184 HYMENOPTERA. 112. have the thorax and node of the peduncle armed with spines or hooks. They construct small semicircular nests, of a kind of net-work, on the leaves of trees and shrubs. Their communities are small, sel- dom exceeding twenty individuals. Mr. Norton describes P. arboricola (Fig. 112, worker major) from Mexico. An allied genus is Ectatomma (Fig. 113, worker major of E. ferruglnea Norton, from Mexico). Mr. F. Smith has described a new genus, CEcopliylla, which is allied to Formica. They are green ants, found building in trees in the tropics of the old world. The nest of (E. smaragdina Smith is "formed by drawing together a number of green leaves, which they unite with a fine web. Some nests are a foot in diameter. They swarm, says Mr. Wallace, in hilly for- ests in New Guinea. Their sting is not very severe. This genus forms a link between Formica and Myrmica ; it agrees with the former in hav- ing a single node to the pe- duncle, and with the latter in having the ocelli obsolete in the workers, and in being fur- nished with a sting." The curious Iloney-ant of Texas and Mexico, Myrmeco- cystus Mexicanus Westwood, has two kinds of "workers of very distinct forms, one of the usual shape," according to Smith, "and performing the Fi o- 113 - active duties of the formica- rium ; the other and larger worker is inactive and does not quit the nest, its sole purpose, apparently, being to elaborate a kind of honey, which they are said to discharge into prepared recep- tacles, which constitutes the food of the entire population of the community. In the hone^v-secreting workers the abdomen is distended into a large globose bladder-like form. From this honey an agreeable drink is made by the Mexicans." FORMICARY. 185 /Tr The second Subfamily, Myrmicarice, includes those species in which the two first abdominal segments are contracted and lenticular. In Myrmica the females and workers are armed with spines, and the ocelli are absent in the workers. The species are very small, and mostly bright colored. Myrmica molesta Say is found in houses all over the world. G. Liucecum describes the habits of the Agricultural Ant of Texas, Myrmica, molefaciens. It lives in populous communi- ties. "They build paved cities, construct roads, and sustain a large military force." In a year and a half from the time the colony begins, the ants previously living concealed beneath the surface, appear above and "clear awajr the grass, herbage, and other litter, to the distance of three or four feet around the entrance to their city, and construct a pavement, .... con- sisting of a pretty hard crust about half an inch thick," formed of coarse sand and grit. These pavements would be inun- dated in the rainy season, hence, " at least six months pre- vious to the coining of the rain," the} r begin to build mounds rising a foot or more from the centre of the pavement. Within these mounds are neatly constructed cells into which the "eggs, young ones, and their stores of grain, are carried in time of rainy seasons." No green herb is allowed to grow on the pavement except a grain-bearing grass, Aristida stricta. This grain, when ripe, is harvested, and the chaff removed, while the clean grain is carefully stored away in dry cells. Lincecum avers that the ants even sow this grain. They also store up the "grain from several other species of grass, as well as seeds from many kinds of herbaceous plants." Pheidole is distinguished by having workers with enormous heads. P. notabilis Smith, from the Island of Bachian, Indian Archipelago, is noted for the enormously enlarged, cubical head of the worker major, which is at least six times the size of the abdomen, while in the worker minor, the head is of the ordinary size. An Indian species, P. providens Westwood, according to Col. S3'kes, "collects so large a store of grass seeds as to last from January and Februaiy, the time of their ripening, till October." The genus Atta is also well-armed, while the workers have a very large, deeply incised and heart-shaped head, without 186 HYMENOPTERA. ocelli, and the second abdominal knot-like ring is very trans- verse. A. dypeata Smith is a Mexican species. In Eciton the man- dibles nearly equal the length of the in- sect itself. This ge- nus is the most ferocious of all the ants, entering the nest of species of Formica and tearing them, limb from limb, and then carrying off the remains to their own houses. Eciton Mexicana Roger (Fig. 114, worker major, o, front view of head, show- rig, in. ing the immense sickle-like mandibles, and only the two basal joints of the antennas ; Fig. 115, worker minor, with a front view of the head, showing the mandi- bles of the usual size). This species, with Eciton iSitmichmsti Norton, (Fig. 11G, worker minor) has been found by Professor Sumichrast at Cordova and Orizaba, Mexico. The males of Eciton are not yet known. Smith supposes that Labidus (a genus allied to Dorylus) is the male form, and Sumi- chrast thinks this conjec- tivre is "sustained by the Fig. 115. fact that it is in the season when the sorties of the Eciton are the more frequent that the Labidus also show themselves." FORMIC ABLE. 187 An allied genus is Pseudomyrma, P. bicolor Guerin (Fig. 117) is found in Central America. P.Jlavidula Smith, found in Central and South America, in Mexico lives, according to Sumichrast, within the spines which arm the stems of certain species of Mimosa. These spines, fixed in pairs upon the branches, are pierced near the end by a hole (Fig. 118), which serves for the entrance and exit of the ants. The genus (Ecodoma differs from Atta in having the thorax armed with spines. (E. Fig. no. Mexlcana Smith (Figs. 119, female; 120, worker major) is abundant on the Gulf Coast of Mexico. In many places, ac- cording to Sumichrast, the natives eat the females after hav- ing detached the thorax. The intelligence of these ants is wonderful. They are seen in immense num- bers transporting leaves. Sumichrast states that "the ground at the foot of the tree, where a troop of these ' arrieras,' or workers, is assembled for despoil- ing it of its leaves, is ordinarily strewn with frag- ments cut off with the greatest precision. And if the Fig. 117. tree is not too lofty, one can satisfy himself that a party of foragers, which have climbed the tree, occupies itself wholly in the labor of cutting them off, while at the foot of the tree are the carriers which make the journeys between the tree and the nest. This manage- ment, which indicates among these a ._ insects a rare degree of intelligence, is, perhaps, not a constant and in- variable practice, but it is an incon- testable fact, and one which can be constantly proved." '.' It is specially in the argillaceous countries that the GEcodonias build their enormous formicaries, so that one perceives them from afar by the Fig. us. projection which they form above the level of the soil, as well as by the absence of vegetation in their immediate neighborhood. These nests occupy a surface of many square 188 HYMENOPTEEA. metres,* and their depth varies from one to two metres. Very many openings, of a diameter of about one to three in- ches, are contrived from the exterior, and conduct to the inner cavities which serve as storehouses for the eggs and larvae. The central part of the nest forms a sort of funnel, designed for the drainage of water, from which, in a country where the periodical rains are often abundant, they could hardly es- cape without be- ing entirely sub- merged, if they did not provide for it some out- let. "The system which reigns in Fig. 119. the interior of these formicaries is extreme. The collection of vegetable debris brought in by the workers is at times considerable ; but it. is deposited there in such a manner as not to cause any inconvenience to the inhabitants, nor impede their circulation. It is mostly leaves which are brought in from without, and it is the almost exclusive choice of this kind of vegetation which makes the (Ecodoma a veritable scourge to agriculture. At each step, and in almost every place in the elevated woods, as on the plains ; in desert places as well as in the neighborhood of habitations, one meets numerous columns of these insects, occupied with an admirable zeal in the transportation of leaves. It seems even that the great law of the divi- sion of labor is not ignored by these little creatures, judging from the observations which I have often had occasion to make." (Sumiehrast.) "The CE. ccphcdotcs" saj-s II. "W. Bates, "from its immense numbers, eternal industry, and its plundering propensities, be- comes one of the most important animals of Brazil. Its immense hosts are unceasingly occupied in defoliating trees, and those most relished by them are precisely the useful kinds. They * A metre is about thirty-nine (39.37) inches. N 120. FORMICAELE. 189 have regular divisions of laborers, numbers mounting the trees and cutting off the leaves in irregularly rounded pieces the size of a shilling, another relay carrying them off as they fall." "The heavily laden fellows, as they came trooping in, all de- posited their load in a heap close to the mound. About the mound itself were a vast number of workers of a smaller size. The very large-headed ones were not engaged in leaf-cutting, nor seen in the processions, but were only to be seen on dis- turbing the nest." Bates also says, "I found, after removing a little of the surface, three burrows, each about an inch in diameter ; half a foot downward, all three united in one tubular burrow about four inches in diameter. To the bottom of this I could not reach when I probed with a stick to the depth of four or five feet. This tube was perfectly smooth and covered with a vast number of workers of much smaller size than those oc- cupied in conveying the leaves ; they were unmixed with any of a larger size. Afterwards, on probing lower into the bur- row, up came, one by one, several gigantic fellows, out of all proportion, larger than the largest of those outside, and which I could not have supposed to belong to the same species. Be- sides the greatly enlarged size of the head, etc., they have an ocellus in the middle of the forehead ; this latter feature, added to their startling appearance from the cavernous depths of the formicarium, gave them quite a Cyclopean character." Of another species, the (Ec. sexdentata, Mr. Smith quotes from Rev. Hamlet Clark, that at Constancia, Brazil, the pro- prietor of a plantation used every means to exterminate it and failed. " Sometimes in a single night it will strip an orange or lemon tree of its leaves ; a ditch of water around his garden, which quite keeps out all other ants, is of no use. This spe- cies carries a mine under its bed without any difficulty. In- deed, I have been assured again and again, by sensible men, that it has undermined, in its progress through the country, the great river Paraiba. At any rate, without anything like a nat- ural or artificial bridge, it appears on the other side and con- tinues its course." This testimony is confirmed by Mr. Lincecum (Proceedings of Academy of Natural Sciences, Philadelphia, 1867, p. 24) in an interesting account of the (Ec. Texana, which he has observed for eighteen yeai's. Pie states 190 HYMENOPTERA. that they often carry their subterranean roads for several hun- dred yards in grassy districts, where the grass would prove an impediment to their progress. On one occasion, to secure ac- cess to a gentleman's garden, where they were cutting the vegetables to pieces, they tunnelled beneath a creek, which was at that place fifteen or twenty feet deep, and from bank to bank about thirty feet. He also observes that the smaller workers which remain around the nest do not seem to join in cutting or carrying the leaves, but are occupied with bringing out the sand, and generall}- work in a lazy way, very differently from the quick, active leaf-cutters. Also, that the pieces of leaves are usually dried outside before being carried in, and that if wet by a sudden shower are left to decay without. He also thinks that their lives are dependent upon access to water, and that they always choose places where it is accessible by digging wells. In one case, a well was dug by Mr. Pearson for his own use, and water found at the depth of thirty feet. The ant-well which he followed was twelve inches in diameter." Fig. 121. (Norton, American Naturalist, vol. 2.) The genus Cryptocerus is remarkable for its flattened head, with the sides expanded into flattened marginal plates, con- cealing, or partly hiding the eyes. C. multispinosus Norton (Fig. 121) is the most common species about Cordova, Mexico, where they live, according to Surnichrast, within the trunks of trees. CIIRYSIDID^E Latreille. In this small group the thirteen- jointed autennre are elbowed, the eyes, are oval and the ocelli distinct. The maxillary palpi are five, aftd the labial palpi three-jointed. There are about four hundred species known. These insects are very different from the ants in their oblong compact form, their nearly sessile, oblong abdomen, having only three to five rings visible, the remaining ones being drawn with- in, forming a. long, large, jointed sting-like ovipositor, which can be thrust out like a telescope. The abdomen beneath is concave, and the insect can roll itself into a ball on being dis- turbed. They are green or black. The sting has no poison- bag, and in this respect, besides more fundamental characters, CHRYSIDID^. 191 the Cluysis family approaches the Ichneumons.. They best merit the name of "Cuckoo-flies," as they fly and run briskly in hot sunshine, on posts and trees, darting their ovipositor into holes in search of the nests of other Hymeiioptera, in which to lay their eggs. Their larvae are the first to hatch and devour the food stored up by other fossorial bees and wasps. "St. Fargeau, however, who has more carefully examined the econ- omy of these insects, states that the eggs of the Chrysis do not hatch until the legitimate inhabitant has attained the greater part of its growth as a larva, when the larva of the Chrysis fastens on its back, sucks it, and in a very short time attains its full size, destroying its victim. It does not form a cocoon, but remains a long time in the pupa state." (Westwood.) " In the Entomological Magazine has been noticed the dis- covery of Iledychrum bidentulum, which appears to be parasitic upon Pscn caliginosus ; the latter insect had formed its cells in the straws of a thatched arbor, as many as ten or twelve cells being placed in some of the straws. Some of the straws, per- haps about one in ten, contained one or rarely two, of the Iledychrum, placed indiscriminately amongst the others. Walkenaer, in his Memoirs upon Ilalictus, informs us that Hedychrum lucidulum waits at the mouth of the burrows of these bees, in order to deposit its eggs therein ; and that when its design is perceived by the bees, they congregate together and drive it away. St. Fargeau states that the females of Hedychrum sometimes deposit their eggs in galls, while H. regium oviposits in the nest of Megachile muraria ; and he mentions an instance in which the bee, returning to its nearly finished cell, laden with pollen paste, found the Hedychrum in its nest, which it attacked with its jaws ; the parasite im- mediately, however, rolled itself into a ball, so that the Mega- chile was unable to hurt it ; it, however, bit off its four wings which were exposed, rolled it to the ground and then deposited its load in the cell and flew away, whereupon the Hedychrum, now being wingless, had the persevering instinct to crawl up the wall to the nest, and there quietly deposit its egg, which it placed between the pollen paste and the wall of the cell, which prevented the Megachile from seeing it." (Westwood.) In Cleptes the underside of the abdomen is not hollowed out ; 192 HYMENOPTERA. it is acutely oval, and with five rings in the male. Ckptes semiaurata Latr. is found in Central Europe. We have no na- tive species. In Chrysis and the other genera, Stilbum, Parno- pes, and Iledychrum, the abdomen is hollowed beneath, and the tip is broad and square. Chrysis hilaris Dahlb. (Fig. 122) is a short, thick, bluish green species, .32 inch in length. It is not uncommon in New England. In Hedyclirum the maxillary palpi and ligula are rather short, the last cordate ; the mandibles are three-toothed within. The abdomen is broad and short, almost spherical, the second seg- ment being the largest. H. dimidiatum Say is found in the Middle States. The European Stilbum splendidum, Fabr. according to Du- four, lives in the cells of Pelopasus spirifex. It makes oblong cocoons of a deep brown, with rounded ends ; they are of great tenacitj', being mixed with a gummy matter. Mr. Guenzius states that in Port Natal "a species of Stilbum lays its eggs on the collected caterpillars stored Fig. 122. up by Eumenes tinctor, which con- structs a nest of mud and attaches it to reeds, etc., not in a single, but a large mass, in which cells are excavated, similar to the nest of Chalicodoma micraria ? * First, it uses its ovi- positor as a gimlet, and when its point has a little penetrated, then as a saw or rasp ; it likewise feels with its ovipositor, and, finding an unfinished or an empty cell it withdraws it immedi- ately, without laying an egg." ICHNEUMOXIIXE Latreille. The Ichneumon-flies are readily recognized by the usually long and slender body, the long ex- serted ovipositor, which is often very long, and protected by a sheath formed of four stylets of the same length as the true ovipositor. The head is usually rather square, with long, slender, many-jointed antenna? which are not usually elbowed. The maxillary palpi are five to six-jointed, while the labial *A query (?) after the name of a species indicates a doubt whether the insect really belongs to that species; so with a ? after the name of a genus. A ? before both the genus and species expresses a doubt whether that be the insect at all. ICHXEUMOXIDJE. 193 palpi are three to four-jointed. The abdomen is inserted im- mediately over the hind pair of trochanters, and usually consists of seven visible segments. The fore- wings have one to three subcostal (cubital) cells. The larva is a soft, fleshy, cylindrical, footless grub, the rings of the body being moderately convex, and the head rather smaller than in the foregoing families. The eggs are laid by the parent either upon the outside or within the caterpillar, or other larva, on which its young is to feed. When hatched it devours the fatty portions of its victim which dies gradually of exhaustion. The ovipositor of some species is very long, and is fitted for boring through very dense substances ; thus Mr. Bond, of England, observes that Wiyssa persuasoria actually bores through solid wood to deposit its eggs in the larvae of Sirex ; the ovipositor is worked into the wood like an awl. When about to enter the pupa state the larva spins a cocoon, consisting in the larger species of an inner douse case, and a looser, thinner, outer covering, and escapes as a fly through the skin of the caterpillar. The cocoons of the smaller genera, such as Cryptus and Microgaster, may be found packed closely in considerable numbers, side by side, or sometimes placed up- right within the body of caterpillars. The Ichneumon-flies are thus very serviceable to the agricul- turist, as they must annually destroy immense numbers of cat- erpillars. In Europe over 2,000 species of this family have been described, and it is probable that we have an equal num- ber of species in America ; Gerstaecker estimates that there are 4,000 to 5,000 known species. The Ichneumons also prey on certain Coleoptera and Hymen- optera, and even on larvre of Pliryganidce, which live in the water. In Europe, Pimpla Fairmairii is parasitic on a spider, Clubione holosericea, according to Laboulbene. Boheman states that P. ovivora lives on a spider, and species of Pimpla and Hemiteles were also found in a nest of spiders, according to Gravenhorst. Bouche says that Pimpla rufata devours, during winter and spring, the eggs of Aranea diadema, and Eatzburg gives a list of fourteen species of Ichneumons parasitic on spiders, belonging to the genera Pimpla, Pezomachus, Ptero- malus, Cryptus, Hemiteles, Microgaster, and Mesochorus. Mr. 13 194 HYMENOFTERA. Eraertoii informs me that he has reared a Pezomachus from the egg-sac of Attus, whose eggs it undoubtedly devours. They are not even free from attacks of members- of their own family, as some smaller species are well known to prey on the larger. Being cut off from communication with the external world, the Ichneumon larva breathes by means of the two principal trachea 1 , which terminate in the end of the body, and are placed, according to Ger- staecker, in com- munication with a stigma of its host. From the com- plete assimilation of the liquid food, Fig. 123. the intestine ends in a cul de sac, as we have seen it in the larvae of Humble-bees and of Stylops, and as probably occurs in most other larvse of similar habits, such as young gall-flies, weevils, etc., which live in cells and do not eat solid food. The first subfamily, the Evaniidce, are insects of singular and very diverse form, in which the antenrse are either straight or elbowed, and thirteen to fourteen- joiuted ; the fore-wings have one to three subcostal (cubital) cells, and the hind wings are almost without veins. In Evania and Foemis the abdomen has a very slender pedicel, originating next the base of the metanotum. The former genus has a remarkably short triangular compressed abdomen in the female, but ovate in the male. The species are parasitic on Blatta and allies. Evania kevigata Olivier (Fig. 123, $ and pupa) is a black species, and is para- sitic on the cockroach, Periplaneta, from the eggs of which we have taken the pupa and adult. The eggs of the cockroach are just large enough to accommodate a_smgle Evania. This species ICHNEUMONID^E . 195 125. is widely distributed, and in Cuba, according to Cresson, it devours the eggs of Periplaneta Americana. The genus Autocodes of Cresson, "forms a very close con- necting-link between the minute Ichneumons and the Evauiae." A. nigriventris Cresson (Fig. 124, a; &, metathorax ; c, inser- tion of the abdomen) lives in Cuba. Foenus is quite a different genus, as the abdomen is very long and slender. Foenus jaculator Linn, is known in Europe to frequent the nests of Crabronidce, ovipositing in the larvae. Pelecimis is a fa- miliar insect, the im- mensely elongated, linear abdomen of the female easily distinguishing it. The male is extremely rare ; its abdomen is short and clavate. It strikingly resembles Trypoxylon, though the abdomen is considerably larger. Pelecimis poly- cerator Drury (Fig. 125, $ and ?) is widely distributed throughout this country. The genuine Ichneumonidce have long, straight, multiarticu- late antennae. The first subcostal (cubital) cell of the fore- wings is united with the median cell lying next to it, while the second is very small or wholly wanting. There are two recurrent veins. Mr. Cresson has described the genus Eipliosonm (Fig. 12G), Fig. 126. which he states may be known by the long, slender, compressed abdomen, and the long posterior legs, with their femora toothed beneath the tips. E. annu- latum Cresson, a Cuban species, is, according to Poey, "para- sitic upon a larva of Pyralis." (Cresson.) In Opliion the antennae are as long as the body, the abdo- men is compressed, and the species are honey-yellow in color. 0. macrurum Linn. (Fig. 127) attacks the American Silk- worm, Telea Polyphemus. Anomalon is a larger insect and usually black. A. vesparum is, in Europe, parasitic on Vespa. 19G HYMENOPTEKA. The genus Rhyssa contains our largest species, and frequents the holes of boring insects in the trunks of trees, inserting its remarkably long ovipositor in the body of the larvae deeply embedded in the trunk of the tree. Harris states that Rhyssu (Pimpla) atrata and lunator (Fig. 128, male) of Fabricius, " may frequently be seen thrusting their slender borers, measur- ing from three to four in- ches in length, into the trunks of trees inhabited by the grubs of the Tre- mex, and by other wood- eating insects ; and, like the female Tremex, they sometimes become fastened to the trees, and die without being able to draw their borers out again." The abdomen of the male is very slender. Pimpla has the ovipositor half as long as the abdomen. P. pedalis Cresson is a parasite on Clisiocampa. The genus Trorjus leads to Ichneumon. The antenna? are shorter than the body ; the abdomen is slightly petiolate, fusi- form, and the second subcostal cell is quadrangular. Troy us exesorius Brulle is tawny red, and is a para- site of Papilio Asterias. The genus Ichneumon (Fig. 129) is one of great extent, probably containing over three hundred spe- cies. The abdomen is long and slender, lanceolate ovate, slightly petiolate. The second subcostal cell is five-sided, and the ovipositor is either concealed or slightly exserted. Ichneumon suturalis Say is a very common form, and has been reared in abundance from the larva of the Army-worm, Leu- cania unipuncta. The body is pale rust-red, with black sutures on the thorax. Another common species, also parasitic on the ICHNEUMONIDJE . 197 Army-worm, is the Ichneumon paratus, which is blackish, banded and spotted with yellow. The singular genus Grotea, established by Mr. Cresson, has along and narrow thorax (Fig. 130 a), and a very long and petiolated abdomen (c). We have taken G. anguina Cresson, the only species known, from the cells of Crabro in raspberry stems received from Mr. Angus. Cryptus is a genus of slender form, with a long, cylindrical abdo. men, which is petiolate. In the fe- male it is oval with an exserted ovipositor. Cresson figures a wing (Fig. 131) of C.? ornatipennis, a Cuban species, which has the wings differently veined from the other species. Westwood remarks that in Europe a species of this genus preys on the larvae of the P tin idee. Pezomachus is usually wingless, and might at first sight read- ily be mistaken for an ant. The body is small, the oval abdo- men petiolate, and the wings, when pres- ent, are very small. The species are very V Fig. 129. Fix. 131. numerous. Gerstaecker suggests that some may be wingless females, belong- ing to winged males of allied genera. The third subfamily is the Braconidce, containing those genera having long multiarticulate antennae, and with the first subcostal cell separate from the first median, lying just behind it. The second subcostal cell is usually large, and there is only one recurrent vein. The genus Bracon is distinguished by the deeply excavated clypeus. The first sub- costal cell is completely formed behind, wanting the recurrent nerve ; the second cell is long, and four-sided. More than five hundred species, mostly of bright, gay colors, are already known. The genus Rhopalosoma of Cres- son connects Bracon and other minute genera (Braconidae) with the true Ichneumons. R. Poeyi Cresson (Fig. 132) is a 130. 198 HYMEXOPTERA. pale honey-yellow species, with a long club-shaped abdomen. It lives in Cuba. Rogas is a genus differing from Bracon in having the three first abdominal rings long, forming a slender petiole. In Microgaster, a genus containing numerous species, the antennae are eighteeii-jointed, and the abdomen is shorter than usual, and clavate. There are two or three subcostal cells, the second very small. Mi- crogaster nephoptericis (Plate 3, figs. 3, 3 ) is parasitic on Nephopteryx Edmandsii, found in the cells of the Humble-bee. Aphidius, the parasite of the Plant-lice, is a most valuable ally of man. It is known by its small size, and by having the second and third segments of the abdomen moving free on each other. There are three cubital cells, though the wings are sometimes wanting. Aphidius (Praon) avena- pliis of Fitch, the Oat-louse Aphidius, is black with honey- yellow legs, and is one-tenth of an inch long. Aphidius (Toxares) triticuphis Fitch, the Wheat-louse Aphidius, is black, shining, with thread-like antennte composed of twenty-five joints. Its length is .08 inch. Frequently the large size of the parasite causes the body of the dead Aphis to swell out into a globular form. PROCTOTRYPID^E (Proctotrupu) Latreille. Egg-parasites. In this family are placed very minute species of parasitic Ich- neumon-like Hymeuopters which have rather long and slender bodies, with straight or elbowed antenna? of various lengths, often haired on the joints, usuallj 1 ' ten to fifteen, sometimes only eight in number, while the wings are covered with minute hairs and most of the iiervures are absent. The maxillary palpi are three to six, the labial palpi usually three-jointed. The abdo- men has from five to seven joints, and the tarsi are mostly five- jointed, rarely four -jointed. These insects are often so minute that they can scarcely be distinguished by the naked eye unless it is specially trained ; they are black or brown, and very active in their habits. They may be swept off grass and herbage, from aquatic plants, or from hot sand-banks. They PEOCTOTRYPIDJE. 19 141 jointed, with a long branch attached to the third, fourth, and fifth joints. The abdomen is flattened, sessile. E. basalts Say was described from Indiana. We figure a Chalcid (Fig. 141, (?), allied to Eulopus, which preys upon the American Tent Caterpillar. A species of Bla'stophaga (B. grossorum Grav.) is interest- ing as it is the means of assisting in the fertilization of the Fig Fig. 140. 208 HYMEXOPTERA. blossoms, which act, as applied to this instance of the fertiliza- tion of flowering plants by insects, has been called by Mr. Westwood "capriflcation." CYXIPIDJE Westwood. (Diploleparioz Latreille.) Gall-flies. In this most interesting family we have a singular combination of zoological and biological characters. The gall-flies are closely allied to the parasitic Chalcids, but in their habits arc plant- parasites, as they live in a gall or tumor formed by the ab- normal growth of the vegetable cells, due to the irritation first excited when the egg is laid in the bark, or substance of the leaf, as the case may be. The generation of the summer broods is also anomalous, but the parthenogenesis that occurs in these forms, by which immense numbers of females are produced, is necessary for the work they perform in the economy of nature. When we see a single oak hung with countless galls, the work of a single species, and learu how numerous are its natural v. C 5*3 Fig. 142 enemies, it becomes evident that the demand for a great nu- merical increase must be met by extraordinary means, like the generation of the summer broods of the Plant-lice. The gall-flies are readily recognized b}* their resemblance to certain Chalcids, but the abdomen is much compressed, and usually very short, while the second, or the second and third seg- ments, are greatly developed, the remaining ones being imbri- cated or covered one by the other, leaving the hind edges exposed. Concealed within these, is the long, partially coiled, very slender ovipositor, which arises near the base of the abdo- men.* Among other distinguishing characters, are the straight *Fig. 142. I, abdomen of C>/iu'j>s q-iierrna-fiHcuJata Osten Sacken, with the ovipos- itor exserted ; II, the same with the ovipositor retracted; III. the abdomen of the female of Figites (Diplolepis) n-lineatns Say; IV, the same showing the ventral portion, in nature covered by the tergal portion of the abdomen ; V, end view of the CYXIPIDJE. 209 (not being elbowed) thirteen to sixteen-jointed antenna?, the labial palpi being from two to four-jointed, and the maxil- lary palpi from four to six-jointed. The maxillary lobes are broad and membranous, while the ligula is fleshy, and either rounded or square at the end. There is a complete costal cell, while the subcostal cells are incomplete. The egg is of large size, and increases in size as the embryo becomes more devel- oped. The larva is a short, thick, fleshy, footless grub, with the segments of the body rather convex. When hatched they immediately attack the interior of the gall, which has already formed around them. Many species transform within the gall, while others enter the earth and there become pupae. It is well known that of many gall-flies the males have never been discovered. "Ilartig says that he examined at least 15,000 specimens of the genns Cynips, as limited -by him, with- out ever discovering a male. To the same purpose he collected about 28,000 galls of Cynips divisa, and reared 9,000 to 10,000 Cynips from them ; all were females. Of C. folii, likewise, he had thousands of specimens of the female sex without a single .male." (Osten Sacken.) Siebold supposes in such cases that there is a true parthenogenesis, which accounts for the immense number of females. Mr. B. D. "Walsh has discovered (American Entomologist, ii, p. 330) that Cynips quercus-aciculata O. Sack., which pro- duces a large gall in the autumn upon the black oak, in the spring of the year succeeding lays eggs which produce galls disclosing Cynips quercus-spongifica O. Sack. He proved this by colonizing certain trees with a number of individuals of C. quercus-aciculata, and finding the next spring that the eggs laid by them produced C. quercus-spongifiea. The autumn brood of Cynips consists entirely of agamous females, while the vernal brood consists of both males and females, and Mr. Walsh declares after several experiments that " the agamous autumnal female form of this Cynips ((7. q. aciculata) sooner or later reproduces the bisexual vernal for.n, and is thus " :i mere dimorphous female form" of C. q. spongijica. abdomen of Cynips, showing the relations of segments 7-8, the sternal portion of the eighth segment being obsolete; up, the single pair of abdominal spiracles; VI, terminal ventral piece, from which the sheaths (s s} and the ovipositor () take their origin : it is strongly attached at m to the tergites of the sixth and seventh rings; o, ovipositor; s, s its sheaths; a, an appendage to r, the terminal sternite. From Walsh. U 210 ELYMENOPTEKA. In this connection he refers to the discovery of Claus, in 18G7, of several males of Psyche helix, which had been sup- posed to be parthenogenous, thousands of specimens having been bred by Siebold, all of which were females. Baron Osten Sacken (in the Proceedings of the Entomol- ogical Society of Philadelphia, vol. 1, p. 50) says that "a strong proof in confirmation of my assertion is, that in those genera, the males of which are known, both sexes are obtained from galls in almost equal numbers ; even the males, not unfreqnently, predominate in number (sec Hartig, 1. c. iv, 399). Now the gall-flies, reared by me from the oak-apple, were all females. Dr. Fitch, also, had only females ; and Mr. B. D. Walsh, at Rock Island, Illinois, reared (from oak-apples of a different kind) from thirty-five to forty females, without a single male. This leads to the conclusion that the Cynipes of the oak-apples belong to the genera hitherto supposed to be agamous." For an account of the habits and many other interesting points in the biology of these interesting insects, we further quote Baron Osten Sacken. ' ' Most of the gall-flies always attack the same kind of oak ; thus, the gall of (7. seminator Harris, is always found on the white oak ; C. tiibicola Osten Sacken on the post oak, etc. Still, some galls of the same form occur on different oaks ; a gall closely resembling that of C. quercus- rjloljulus Fitch, of the white oak, occurs also on the post oak, and the swamp chestnut oak ; a gall very similar to the com- mon oak-apple of the red oak occurs on the black-jack oak, etc. Are such galls identical, that is, are they produced by a gall-fly of the same kind? I have not been able to investigate this question sufficiently. Again, if the same gall-fly attacks dif- ferent oaks, may it not, in some cases, produce a slightly differ- ent gall ? It will be seen below, that C. quercus-futilis, from a leaf-gall on the white oak, is very like (7. quercus-papillata from a leaf-gall on the swamp-chestnut oak. I could not perceive any difference, except a very slight one in the coloring of the feet. Both gall-flies may belong to the same species, and although the galls are somewhat different, they are in some respects analogous, and might be the produce of the same gall- fly on two different trees. CYNIPID.E. 211 "Some gall-flies appear very early in the season; Cynips quercus-palustris for instance, emerges from its gall before the end of May ; these galls are the earliest of the season ; they grow out of the buds and appear full grown before the leaves are developed. May not this gall-fly have a second generation, and if it has, may not the gall of this second generation be different from the first produced, as it would be under different circumstances, in a more advanced season, perhaps on leaves instead of buds, etc? "A remarkable fact is the extreme resemblance of some of the parasitical gall-flies with the true gall-fly of the same gall. Thus, Cynips quercus-futilis, O. Sacken, is strikingly like Aulax? futilis, the parasite of its gall. The common gall on the black- berry stems produces two gall-flies which can hardly be told apart at first glance, although they belong to different genera." (Proceedings of the Entomological Society of Philadelphia.) Hartig has divided this family into three sections : First, Cynips and its allies, the true gall-flies (Pseuides) in which the second (counting the slender pedicel as the first) segment of the abdomen is longer than half its length, and the subcostal area is narrow, the basal areolet (cell) being opposite the base of the former. Cynips conflnens Harris forms the oak-apple commonly met with on the scrub-oak. There is a spring and summer brood. These galls, sometimes two inches in diameter, are green and pulpy at first, but when ripe have a hard shell with a spongy interior, in the centre of which, lodged in a woody kernel, which serves as a cocoon, the larva transforms, escaping through a hole, which it gnaws through both the kernel and shell. AVe have found the fly ready to escape in June, and Dr. Harris has found it in October. Two galls are represented on Plate 4, fig. 13 ; the larger of which has been tenanted, after the gall-flies had escaped, by an CkVpierus. Cynips gallce-tinc- torice Olivier produces the galls of commerce, brought from Asia Minor. Biorhiza (Apophyllus Hartig) is a wingless genus, and lives beneath the earth in galls formed at the roots of oak trees. Biorhiza nigra Fitch is black throughout, including the antennae and feet, and is but .08 inch long. 212 HYMENOPTEEA. Galls are often found on the blackberry, tenanted by another genus, Diastroplius, which has usually fifteen-jointed antennae in the male, and one joint less in the female. On opening a gall containing this fly, we often find an inquiline gall-fly, Aulax, " showing the most striking resemblance in size, color- ing and sculpture, to the Diastrophus, their companion. The one is the very counterpart of the other, hardly showing any differences, except the strictly generic characters." (Osten Sacken.) These galls are also infested by Chalcid parasites, Callimome (two species), Ormyrus, and Eurytorna. Osten Sacken enumerates "eight cynipidous galls on the dif- ferent kinds of roses of this country." The flies all belong to the genus Rliodites, which is distinguished by the under side of the last abdominal segment being drawn out into a long point, while the antennas are fourteen-joiuted in both sexes. R. rosce produces the bede- IJIKIT gall ("from the Hebrew bedeguach, said to mean rose- apple"). It was formerly used as a medicine. The galls form a moss-like mass, encircling the rose branch. Rliodites dichlocerus of Harris (Fig. 143), produces hard, woody, irregular swellings of the branches. We now come to the second section, the Guest gall-flies (In- quiliufe), which are unable to produce galls themselves, as they do not secrete the gall-producing poison, though possessing a well developed ovipositor. Hence, like the Nomada, etc., among bees, they are Cuckoo-flies, laying their eggs in galls already formed. This group may generally, according to Mr. Walsh, be dis- tinguished from the preceding by the sheaths of the ovipositor always projecting, more or less, beyond the "dorsal valve," which is a small, hairy tubercle at the top of the seventh ab- dominal segment. This dorsal valve also projects greatly. In almost all the species, the ovipositor projects from between the tips of the sheaths. Among the Inquiline genera are Synoplirus, Amblynotus, Synerges, and Aulax, which are guests of various species of Cjaripides. In Ficjites and allies (Figitidie), the third section of the TENTHREDINID^E . 213 family, the second segment is shorter than half the length of the abdomen, being much longer and less high and compressed than in the Cynipides, and the ovipositor is retracted within the abdomen. These insects are true internal parasites, re- sembling the Chalcids. Ibalia is a parasite on a wood-beetle. This genus has, by Walsh, been placed in the Cynipides. Figites has feather-like antennae in the male ; it is a parasite .on the larvae of Sarcophaga. The genus Allotria is a para- site on Aphis. Walsh states that two genera, which he has identified as Kleidotoma and Eucoila are true Figitidce, and "have the wings fringed like a Mymar, and the former has them emargi- nate at tip with the radial area in my species distinctly open, and the latter simple at tip with the radial area in my species marginally closed by a coarse brown, vein." Eucoila is sup- posed to be parasitic on some insect attacking the turnip. TENTHREDINID^E Leach. The Saw-flies connect the Hymen- optera with the Lepidoptera. In the perfect state they con- form to the Hymenop- k a terous tj'pe, but as t larvae they would often be mistaken for Lepi- rf - c ..- dopterous larvae, and |;; in their habits closely "H" resemble many cater- pillars. The three divisions of the body, usually so trenchantly marked in the higher Hymenoptera, are here less distinct, since the abdomen is sessile, its basal ring being broad and applied closely to the thorax, while the succeeding rings are very equal in size. The head is broad and the thorax wide, closely resembling that of the Lepidoptera. The wings (Fig. 144, fore-wing) are larger in proportion to the rest of the body than usual ; they are more net- veined, the cells being more numerous and extending to the outer margin.* *In treating of this family we avail ourselves largely of the important work on the American species, publishing at the time of writing, by Mr. E. Norton, in the Transactions of the American Entomological Society, vols. 1, 2. We therefore 214 HYMEXOPTERA . All these characters show that the saw-fi}" is a degraded Hyuienopter. The antennae are not elbowed ; are rather short and simple, clavate, but in rare instances fissured or feathered. The ab- domen consists, usually, of eight external segments, the two last being aborted on the under side, owing to the great develop- ment of the ovipositor. The ovipositor or "saw" (compare Fig. 24) consists of two lamellae, the lower edge of which is. toothed and fits in a groove in the under side of the upper one, which is toothed above, both protected by the usual sheath-like stylets. On pressing, says Lacaze-Duthiers, the end of the abdomen, we see the saw depressed, leave the direction of the axis of the body, and become perpendicular. By this movement the saw, which both cuts and pierces, makes a gash in the soft part of the leaf where it deposits its eggs. The eggs are laid more commonly near the ribs of the leaf, in a series of slits, each slit containing but a single egg. "Some species, 011 the other hand, introduce their eggs by means of their saws into the edges of leaves (Nematus conju- gatus Dahlb.), and others beneath the longitudinal ribs of the leaves. A few, indeed, merely fasten their eggs upon the outer surface of the leaves (Nematus grossularice, etc.), attaching them together like a string of beads (Reaumur, vol. v, plate 10, fig. 8), whilst a few place them in a mass on the surface of the leaf (ibid, plate 11, figs. 8, 9)." (Westwood.) The irritation set up })y the saws in the wounded leaf, causes a flow of sap which is stated by Westwood to be imbibed by the egg, so that it swells gradually to twice its original size. It is known that the eggs of ants increase in size as the embryo develops, and we would copy his diagram (Fig. 144), showing the venation of the wing (compare Fig. 29 and our nomenclature), with the explanation of parts given by him. a, stigma; b, costa or costal margin; c, apical margin; d, costal and post- costal veins; e, externomedial ; ./", #, anal; /*, posterior margin; ', marginal vein; j, submargin.il vein; /, first, second, and third (transverse) submargmal nervures; /, recurrent nervures (discoidal); m, dm-oklal vein; n, first and second inner api- cal or submarginal nervures. Bullw or clear spots, on the veins or nervures, with bullar or clear lines crossing them. 1,2, marginal or radial cells ; 3, 4, 5, 6, submar- ginal or cubital cells; 7, 8, 9, discoidal cells; 10, costal cell; 11, 12, brachial or me- dial cells; 1.3, 14, inner and outer apical cells. (Hinder cells, Hartig. Cellule du limbe, St. Farg.) Xo. 11 is sometimes the medial, and Nos. 12 and 13 the submedial cells; Nos. 9 and 14 the apical cells; Nos. 7 and 13 discoidal; Nos. 10, 11, 12, 15, the first, second, third and fourth brachial celts"; ~~Kt, lanceolate cell. 1, open; 2, con- tracted; 3, petiolate; 4, subcontracted; 5, with obluifre cross nervure; 6, with straight cross nervure. TENTHREDINID^E. 215 question whether the increase in size of the eggs of the Saw- fly is not rather due to the same cause. The punctures in the plant often lead, in some genera, to the production of galls, in which the larvae live, thus showing the near relationship of this family to the gall-flies (Cynipidoe) . The larvae strongly resemble caterpillars, but there are six to eight pairs of abdominal legs, whereas the caterpillar has but five pairs. Many species curl the hind body up spirally when feeding or at rest. They are usually green, with lines and markings of various colors. They usually moult four times, the last change being the most marked. Most of the larvae secrete silk and spin a tough cocoon, in which they hiber- nate in the larva, and often in the pupa state. The pupa has free limbs, as in the other families. The eggs are usually de- posited in the leaves of plants, but in a few cases, according to Norton, in slender or hollow stems. While some are slug- shaped, like the Pear-slug, others like Lyda inanita, mentioned by Westwood, live on rose bushes, and construct a "portable case, formed of bits of rose-leaves arranged in a spiral coil;" and other species are leaf-rollers, like the Tortricids. The larva of Cephus does injury to grain, in Europe, by boring within the stems of wheat. A remarkable instance of the care of the saw-fly for her young, is recorded by Mr. II. II. Lewis, who observed in Australia, the female of Pcrga Lcv:isii deposit its eggs in a slit next the midribs of an Eucalyptus leaf. They were placed transversely in a double series. "On this leaf the mother sits till the exclusion of the larviB ; and as soon as these are hatched, the parent follows them, sitting with out- stretched legs over her brood, protecting them from the attacks of parasites and other enemies with admirable perseverance." (Westwood.) The species are mostly limited to the temperate zone, but few being found in the tropics. The perfect insects mostly occur in the early summer, and are found 011 the leaves of the trees they infest, or feeding on flowers, especially those of the umbelliferous plants. The genus Cimbex contains our largest species, the antennae ending in a knob. C. Americana Leach is widely distributed, and varies greatly in color. The large whitish larva, with a 216 HYMENOPTEKA. blackish dorsal stripe, may be found rolled up in a spiral on the leaves of the elm, birch, linden and willow trees. When disturbed it ejects a fluid from pores situated above the spira- cles. It constructs a large tough parchment-like cocoon, and the fly appears in the early summer. The genus Trichiosoma is recognized by its hairy body, and the antennae have five joints preceding the three-jointed club. T. triangulum Kirb3 r is found in British America and Colorado, and a variety, T. bicolor Harris, on Mount Washington ; it is black, except the tip of the abdomen, with the fourth and fifth joints of the antennae piceous, and the thorax is covered with ash-colored hair. In Abia the antennas are seven-jointed, with the club obtuse ; the bocby is villose, the abdomen having a metallic silken hue. The Abia caprifolii Norton (Fig. 145, larva) is very destruc- tive to the Tartarian Honeysuckle, sometimes stripping the bush of its leaves during successive sea- sons in Maine and Massachusetts. It hatches out and begins its ravages very soon after the leaves are out, eating cir- cular holes in them. It lies curled up on the leaf and when disturbed emits drops of a watery fluid from the pores in the sides of the body, and then falls to .the ground. During the early part of August it spins a pale yellowish silken cocoon, but does not change to a pupa, Mr. Riley states, until the following- Fig. 145. spring. He describes the larva as being common about Chicago ; that it is "bluish green on the back, and yellow on the sides, which are pale near the spiracles, and covered with small black dots. Between every segment is a small, transverse, yellow band, with a black spot in the middle and at each end. Head free, of a brownish black above and color of the body beneath." The fly is described by Norton as being black, with faint greenish reflections on the abdomen ; there are two white bands at the base of the metathorax, and the wings are banded. It is .30 inch long and the wings ex- pand .70 inch. The larvae can easily be destroyed from their TENTHEEDINID^E. 217 habit of falling to the ground when the bush is shaken, where they can be crushed by the foot. Dr. Fitch has reared Abia cerasi from one or two cocoons found on the wild cheriy, the fly appearing in New York during March. Hylotoma is a much smaller genus ; the basal joint of the antenna is oval, while the second is small and round, and the terminal joint is very long. The larva is twenty-footed, and when eating curves the end of the body into the form of an S. The pupa is protected by a gauzy, doubly enveloping cocoon. H. McLeayi Leach is wholly black, sometimes with a tinge of blue. It is found throughout the Northern States. The genus Pristipliora, closely allied to Nematus, is known by its nine-jointed antennae, and the single costal cell ; the first submarginal (subcostal) cell having two recurrent veinlets. P. identidem Norton has been discovered by Mr. W. C. Fish to be destructive to the cranberry on Cape Cod. He has reared the insect, and sent me the following notes on its habits, while the adult fly has been identified by Mr. Norton, to whom I submitted specimens. The larvae were detected in the first week of June, eating the leaves ; "they were light or pale yel- lowish green when first hatched," and grew darker with age. The head of the young was dark, but in the full-grown worm lighter. When full-grown they were about .30 of an inch in length, and had two lighter whitish green stripes running along the back from head to tail. They had spun their cocoons by the 20th of June in the rubbish at the bottom of the rearing bot- tles. On the 29th of June the} r came out in the perfect state. We would add to this description that the body, in two alco- holic specimens of the larvae, was long, cylindrical, and smooth, Avith seven pairs of abdominal feet. The head is full, rounded and blackish, but after the last moult pale honey-yellow. The male is shining black, and Mr. Norton informs me that it is his P. idiota. P. grossularice Walsh is a widely diffused species in the Northern and Western States, and injures the currant and gooseberry. The female fly is shining black, while the head is dull yellow, and the legs are honey-yellow, with the tips of the six tarsi, and sometimes the extreme tips of the hinder tibiae and of the tarsal joints pale duskjr for a quarter of their length. The wings are partially hyaline, with black veins, a 218 HYMENOPTERA. hone3 r -3"ellow costa, and a dusky stigma, edged with honey- yellow. The male differs a little in having black coxae. Mr. Walsh states that the larva is a pale grass-green worm, half an inch long, with a black head, which becomes green after the last moult, but with a lateral brown stripe meeting with the opposite one on the top of the head, where it is more or less confluent ; and a central brown-black spot on its face. It appears the last of June and early in July, and a second brood in August. They spin their cocoons on the bushes on which they feed, and the fly appears in two or three weeks, the specimens reared by him flying on the 26th of August. P. sycoplianta Walsh is an "inquiline," or guest gall-saw-fly, inhabiting a Cecidomyian gall on a willow. The genus Euura comprises several gall-making species. It differs from the preceding genus in the second, instead of the first, submarginal cell having two recurrent venules. Mr. Walsh has raised E. orbitalis Norton (E. genuina Walsh) from galls found on Salix humilis. This gall is a bud which is found enlarged two or three times its natural size, before it unfolds in spring. The larva is twenty-footed, is from .13 to .19 of an inch long, of a greenish white color, and the head is dusliy. It bores out of its gall in autumn, descending an inch into the ground, where it spins a thin, silken, whitish cocoon. The gall of E, salicis-ovum Walsh is found on Salix cordata. The female is shining yellow, while the ground color of the male is greenish white. The gall of this species is an oval roundish, sessile, one-chambered, green or brownish swell- ing, .30 to .50 of an inch long, placed lengthwise on the side of small twigs. The larva is pale yellowish, and the fly appears in April. The fly is, according to Walsh, " absolutely undistin- guishable b} r any reliable character from the guest gall-saw-fry, Euura perturbans Walsh," which inhabits dipterous galls made by Cecidomyian flies on the willow and grape (Wal&h). If these two "species" do not differ from each other, either in the larva or adult state, "by any reliable characters," then one must question whether the variation in habits is sufficient to separate them as species, and whether E. salicis-ovum does not, some- times, instead of forming a new gall, lay its eggs in a gall read}*- inade by a dipterous gall-fly. We have seen that Odynerus TENTHKEDINIDJE. 219 albophaleratus, which usually makes a, mud cell situated in the most diverse places, iu one case at least, makes no cell at all, but uses the tunnel bored out by a Ceratina ! and yet we should not split this species into two, on account of this difference in its habits. We had written this before meeting with Mr. Norton's remark that "it is difficult to give a hearty assent to Mr. Walsh's inquilines or guest-flies, without further inves- tigation." (Transactions of the American Entomological So'ciety, vol. i, p. 194.) In Nematus the nine-jointed antennae have the third joint longest. There is one costal and four subcostal cells, the second cell receiving two recurrent veinlets ; the basal half of the lanceolate cell is closed ; the hind wings have two mid- dle cells, and the tibiae are simple. The larvse are hairy with warts behind the abdominal feet. The}- have twenty feet, the fourth and eleventh segments (count- ing the head as one) being footless. They are either solitary, feeding upon the leaves of plants, or social and generally found on pine trees, while some species live in the galls of plants. The pupa, according to Hartig, is enclosed in an egg-shaped cocoon, like that of Lophyrus, but less firm, though with more outside silk. It is generally made in the earth, or in leaves which fall to the ground. N. vertebratus Say is green, with the antennse and dorsal spots blackish, the thorax being trilineate. There are fifty species in this country, of, which the most injurious one, the Gooseberry saw-fly, has been brought from Europe. This is the JY". ventricosus King which was undoubtedly imported into this country about the year 1860, spreading mostly from Rochester, N. Y., where there are extensive nurseries. It does more injury to the currant and gooseberry than any other native insect, except the currant moth (Abraxas ribearia). Professor Winchell, who has studied this insect in Ann Arbor, Michigan, where it has been very destructive, observed the female on the 16th of June, while depositing her cylindrical, whitish and transparent eggs, in regular rows along the under side of the veins of the leaves, at the rate of about one in forty-five seconds. The embryo escapes from the egg in four days. It feeds, moults and burrows into the ground within a period of eight days. It remains thirteen da}-s in the ground, being 220 HYMENOPTERA . most of the time in the pupa state, while the fly lives nine days. The first brood of worms appeared May 21, the second brood June 25. Winchell describes the larva as being pale-green, with the head, tail and feet, black, with numerous black spots regularly arranged around the body, from which arise two or more hairs. Figure 14G, 1, shows the eggs deposited along the under side of the midribs of the leaf; 2, the holes bored by the very young larvae, and 3, those eaten by the larger worms. In transporting gooseberiy and currant bushes, Walsh recom- mends that the roots be carefully cleansed of dirt, so that the cocoons may not be car- ried about from one gar- den to another. The leaves of the bushes should be examined during the last week of May, and as only a few leaves are affected at first, these .can be de- tected by the presence of the eggs and the little round holes in them, and should be plucked off and burnt. The female saw- fly is bright honey-yellow, with the head black, but yellow below the insertion of the antennae. The male differs in its black thorax, and the antennae are paler reddish than in the female.* The genus Empliytus has nine-jointed antennae ; the third *Mr. Norton has communicated the following description of the larva of another saw-fly of this genus which infests the weeping-willow. " Nematus trilineatus Norton. The larvae of this were flrst seen upon the weep- ing-willows about August 1st, in immense numbers, almost wholly stripping large trees of their leaves. They begin upon the edge of the leaf and eat all of it except the inner midrib. They are very sensitive to disturbances, very lively, and are generally found with the hinder part of their bodies bent up over the back. They are twenty-footed, of a bright green color, palest at head and tail, with five rows of black dots down the back, the outer row upon each side irregular and with inter- vals. On each side above the feet is another row of larger black dots, and the tlu-ee anterior pair of feet are black at the base, middle and tip. " A great number of the saw-flies Avere found flying about the trees, August 10th, in the proportion of about ten males to one female. The males being almost wholly black upon the thorax." TENTHEEDINIDJE. 221 and fourth joints of equal length ; the wings have two subcos- tal and three median cells, the first as long as the second, gen- erally longer ; the first receiving one recurrent vein, the second two. We have found the larva of E. maculatus Norton on the cultivated strawberry, to which, in the Western States, it some- times does considerable damage, but it can be quite readily exterminated by hand-picking. Mr. Riley has carefully ob- served the habits of this insect, and we condense the follow- ing remarks from his account in the Prairie Farmer : Early in May, in Northern Illinois, the female saw-fly deposits her eggs in the stem of the plant. They are white and .03 of an inch long, and may be readily perceived upon splitting the stalk ; though the outside orifice, at which they were intro- duced, is scarcely perceptible, their presence causes a swelling in the stalk. By the mid- dle of May the worms will have eaten innumerable small holes in the leaves. They are dirty yellow and Fi >- 147 - gray green, and at rest curl the abdomen up spirally. They moult four times, and are, when full-fed, about three-fourths of an inch in length. They make a loose, earthen cocoon in the ground, and change to perfect flies by the end of June and the beginning of July. A second brood of worms appears, and in the early part of August descend into the ground and remain in the larva state until the middle of the succeeding O April, when they finish their transformations. The fly is pitchy black, with two rows of dull, dirty white, transverse spots upon the abdomen. The nine-jointed antenna? are black, and the legs are brown, and almost white at the joints. Fig. 147 rep- resents the Strawberry Emphytus in all its stages of growth. 1, 2, ventral and side-view of the pupa; 3, the fly enlarged; 222 HYMEXOPTEEA. 5, the same, natural size ; 8, an antenna enlarged ; 4, the larva while feeding ; 6, the same, at rest ; 7, the cocoon ; 9, an egg enlarged. Of the genns Dolerus, known by the second submargiual cell receiving two recnrreuts, D. arvensis Say, is a common blue- black species found in April and May on willows. The genus Selandria is the most injurious genus of the family. It embraces the Pear and Rose-slugs, the Vine-slug and the Raspberry slug. The flies are small, black, with short and stout nine-jointed an- tennae, and broad thin wings. "The larvae are twenty and twenty-two-footed, present- ing great differences in appearance and habit, being slim}', hairy or woolly, feeding in companies or alone, eating the whole leaf as they go, or, removing only the cuticle of the leaf, and forming sometimes one and some- times two broods in a year. /Selandria vitis, the Vine-slug, is twenty-footed ; it has a smooth skin, and the body is somewhat thick- ened in the middle but slender towards the tail. "While growing, the color is green above, with black dots across each ring, and yellow beneath, with head and tail black. They live upon the vine and are very destruc- tive, feeding early in August in companies, on the lower side of the leaf, and eating it all as the}' go from the edge inwards. There are two broods in a season. The fly is shining black, with red shoulders, and the front wings are clouded." (Norton.) S. ruin Harris feeds on the raspberry, appearing in May. The larva is green, not slimy, and feeds in the night, or early in the morning. S. tilice feeds on the linden. The Pear-slug, S. cerasi Peck (Fig. 148, larvae feeding on a leaf of the pear, and showing the surface eaten off in patches ; o, enlarged ; b, fly) , is twenty-footed ; it narrows rapidly behind the swollen thorax, and is covered with a sticky olive-colored slime. It feeds on the upper side of the leaves of both the wild and culti