The above are the principal forms that appear to mePrimary(though some doubt may rest upon the ninth and tenth); and probably others will hereafter be discovered, since at present our knowledge of the larvæ of most of the Orders is very limited. And now having given you this generalization of them, as far as they are known to me, I shall next, in a slight survey of those of the different orders, lay before you what I have further to observe upon this subject.
Coleoptera.TheAnopluriformcoleopterous larvæ, according to Mr. W. MacLeay's view of them, include both those ofCoccinellaL.,ChrysomelaL., andCassidaL.; but this appears to me to admit of further consideration. With regard to the two former—those ofCoccinellaare carnivorous, those ofChrysomelaherbivorous; the first is also usually more flat and depressed. As to the latter,Cassida[391], it seems to me to belong to a peculiar type, distinguished not only by its radiated margin, but by the remarkable deflected anal fork on which it carries its excrement. At present I know no analogous form amongst the apterous tribes; I must therefore leave this without a denomination. Perhaps the larva ofHispaorAlurnus, when known, will throw light upon this subject. The larva ofEndomychusagrees with that ofCoccinella.
There are very few known larvæ that approach to a trueThysanuriformtype in this order: that most celebrated is the one supposed to belong toMeloe; but the claim of this to be so considered, is, as we have seen, rather dubious. Should this point at last be satisfactorily ascertained, itwill probably carry with it the whole tribe of vesicatory beetles. But even this animal in itsgeneralstructure isanopluriform: the only circumstance that gives it any analogy to theThysanurabeing its anal setæ. Mr. William MacLeay is inclined to regard some of the larvæ of theMalacodermiLatr., but which of them he does not state, as probably belonging to the tribe in question[392]. Those ofLampyrisand ofTelephorus, as described and figured by De Geer[393], appear to me intermediate between theAnopluriformandChilopodiformTypes: they have no anal setiform or styliform appendages, their mandibulæ are falcate, and their habits seem carnivorous.
Examples ofChilopodiformcoleopterous larvæ are more numerous. Of this description are those ofGyrinus,Cicindela,Carabus, andStaphylinus. That of the first, indeed, appears to be the most perfectlyScolopendriformof any yet known; yet the gills or respiratory laminæ, a pair of which issues from each abdominal segment, and two pair from the last[394], prove that there is no slight analogy between it, and indeed many other aquatic larvæ, and theStomapodaamongst theCrustacea. A remarkable instance of analogy with theDecapodaof the sameClassis presented by the larva ofDytiscus, &c. which Mr. MacLeay considers asChilopodiform, but which exhibits no other resemblance toScolopendræthanin its predaceous habits and threatening aspect. Its convex, compressed, tapering body, terminating in setæ or laminæ, is certainly much more like that of a shrimp or a prawn; to which the older Entomologists thought it was akin[395], and after which they named it. As Mr. MacLeay's object was, to take all his forms from theAmetabola, perhaps these larvæ will best fall in with hisChilopodiformtype; though in the general form of their body they most represent a section (LepismaL.) of theThysanura.
Chilognathiformforms are equally numerous in the Coleoptera with the preceding. The wire-worm, or larva ofElater Segetis, as to shape best represents the full-grownIulus[396], and those of thePetalocera(ScarabæusL.,LucanusL.) the young one.
The most abundant of all forms in this order, is, I think, theVermiform, upon which I have nothing further to remark.
With regard toCrustaceousforms inColeoptera, besides theDecapodiformjust noticed, I possess two specimens of larvæ ofSilphidæwhich seem to exhibit a considerable analogy with theIsopodous Crustacea, one ratherconvex and the other flatter, so as to give the idea of anArmadilloand of anOniscus.
Strepsiptera.LarvaVermiform.
Dermaptera.LarvaThysanuriform. TypePoduraorSminthurus.
Orthoptera.Mr. MacLeay considers the larvæ of this Order as primarilyThysanuriform[397], though he allows the resemblance between them andAmphipodato be particularly striking[398]. For my own part, their prototypes appear to me to be in theCrustacea, and their analogical relations to theThysanuramuch more distant. I trust this will appear to you the reverse of dubious in a progress through the Crustaceous Orders. I begin with theIsopoda. Take the larva of aBlatta, and place it between aLepisma, orMachilis, and anOniscus, orPorcellio; you will find that in shape and width, and the form of its anal styles, it resembles the latter much more than it does the former, with which it possesses scarcely any character in common, except its multiarticulate antennæ. It is remarkable, that amongst theBlattidæwe meet with species that represent both theOniscidæandArmadilloorGlomeris[399], the latter being more convex than the former. In their habits the Blattæ certainly agree withLepisma; and Dumeril, who thought the latter andPodurasubject to a metamorphosis, imagined they were related[400].
The Spectres of Stoll (PhasmaF.) are so strikingly analogous to another crustaceous tribe, theLæmodipoda, particularly the genusCaprella, that Montagu gave one species the Trivial name ofPhasma[401]. The jumpingAmphipodiformCrustacea are represented extremely well byGryllusL., and theStomapodiform, particularlySquilla Mantis, byMantis. The resemblance in this last instance is so very striking, that it cannot escape the eye of the least intelligent observer. Orthopterous insects may perhaps one day be discovered analogous to the two other crustaceous orders, theDecapodsandBranchiopods; but at present I know of none of that description.
Hemiptera.The larvæ of this order, which in general resemble the perfect insect, except that they have no wings, seem most commonly to belong to theAnopluriformtype[402]; but theAphides,Chermes, andThripsmay, I think, be regarded as more analogous to the generaPoduraandSminthurusin theThysanura[403]. I have some suspicion that theNepidæ,Naucoris, and the remipedes,Notonecta,Sigara, &c. may find their prototypes amongst theCrustacea; but my confined knowledge of the latter does not enable me to point to any individual genera or tribes that they may be presumed to represent.
Neuroptera.As the kinds of larvæ of the different tribes composing this order, as it now stands, are very various, it is to be expected that the analogical forms they represent are equally so. TheLibellulinaMacLeay (whose metamorphosis that gentleman has denominatedsubsemicomplete, a term warranted by their losing in their perfect state the mask before described[404]) in their oral organs, particularly by their galeatemaxillaand distinctligula[405], have some relation to theOrthoptera, the prototypes of whose larvæ we have found amongst theCrustacea: probably, therefore, those of the tribe in question lurk in that class; a suspicion that receives strong confirmation from the larva ofAgrion[406], which in its tapering body and anal natatorious laminæ represents a shrimp. The larvæ of that very peculiar and distinct tribe, theEphemerina, appear to be intermediate between theStomapodiformandThysanuriformtypes. Their natatory respiratory abdominal laminæ seem copied from the former, and their anal diverging setæ from the latter[407]. TheMyrmeleonina, as well in their general form as in their motions and habits, present a most singular analogy with the tribe of spiders, as does also in some respect that ofCicindela. With regard toPanorpa, which Mr. MacLeay remarks is related toMyrmeleon[408], and is a most ferocious insect[409], as its larva has not yet been discovered, nothing certain respecting its analogical form can be asserted; but should it, like the male fly, represent the scorpion, both orders ofArachnidawill have their representatives in the class we are considering. TheCorydalina, as far as the larva ofHemerobiusinstructs us, isChilopodiform, but with a tendency to theAraneidiformType. TheAmetabolaalso furnish the prototype of the next tribe, theTermitina, which, as is evident both fromPsocusandTermes, are perfectlyAnopluriform. TheSialina, orPlicipennesof Latreille, excludingTrichopteraKirby, appear to me to be intermediate between theChilopodiformandStomapodiformTypes, and not without some relation to theThysanuriform. Their pediform, jointed, respiratory abdominal appendages, their head and falcate mandibles, seem copied from the first tribe. The same appendagesconsidered as organs of respiration, and their taper forks, are moulded upon the plan of theStomapodiform Crustacea, and the long seta which terminates the abdomen is upon theThysanuriformplan[410].
Trichoptera.The larvæ of this order appear also to be constructed upon a double plan. The respiratory threads observable in both the upper and under sides of the abdomen connect it with theStomapoda, and its cylindrical elongate body withChilognathiformtypes in theLepidoptera[411].
Lepidoptera.The great majority of larvæ in this order areChilognathiform, but there are exceptions to this remark. Those of theGeometrærecede from this type, both in their motions and the distance and number of their legs. In both these respects they represent theLæmodipodain theCrustacea[412]. Other caterpillars areOnisciform; and a third sort seem to leave theAnnulosetype, and imitate that of theMollusca, and one is figured by Madame Merian[413]which appears to tend even to theChilopodiformtype.
Hymenoptera.In this order the larvæ of the saw-flies,TenthredoL., are in generalChilognathiform, though some areOnisciform, othersLimaciform, and those ofLydaF. (CephaleiaJur.) andSirex[414]have aVermiformtendency,and are a stepping-stone to those of the rest of the order, which are allVermiformand apods.
Diptera.The majority of this order may be set down asVermiform, though it is not improbable that some of them bear an analogy to animals that appear far removed from theAnnulosa. Thus, the larva ofStratyomis Chameleonseems to exhibit no small resemblance to some of thePolypi vaginatiin theAcritasubkingdom of Mr. W. MacLeay[415]. That ofCulexand some others is constructed on a quite different type from the rest, and seems to possess some analogy to theBranchiopod Crustacea.
Though some of these analogies are more striking than others, yet in almost all that I have stated there is that kind of resemblance that could not be the result of what is called mere chance; and Mr. MacLeay, by first pointing out this plan of the All-wise Creator, and by laying down the doctrine ofanalogiesin general, as distinguished fromaffinitiesin the animal kingdom, has furnished the believer with a new argument against those attacks of the infidel, that would render null those proofs of the wisdom and goodness of the Author of nature with which the animal and vegetable creation furnish us; by affirming most absurdly, and under the most stultifying blindness of mind, that thecreatureswere in a manner their owncreators, their wants under local circumstances stimulating them to efforts that in a long course of years produced all the different forms and organizations that are now to be found in our globe. The affinities and close connexion of beings with each other, so that the ascent from low to high is usually by the most gentle gradations, is thecircumstance on which they build this strange and impious theory. But the fact, that certain animals of one tribe were created with a view to certain animals of another, so as to present a striking aspect of correspondence, parallel almost with that of type and antitype, without any real affinity or approximation;—this triumphantly proves a Power above and without them, who has associated them not only in a complex chain of affinities, but has caused them to represent and figure each other, even when evidently far removed, so as to give a mutual correspondence and harmony to the whole, which could be produced only by a Being infinite in power and wisdom, who made all things after a general preconceived plan and system.
iv. We are now to consider theclothingwith which larvæ are furnished. Many are quite naked, and smooth or rough only with granular elevations or tubercles orderly arranged; but a very considerable number, especially of theLepidopteraorder, are clothed with hair or bristles of different kinds, in greater or less abundance, and arranged in different modes; and a proportion still smaller have their skin beset with spines or a mixture of spines and hairs. Lyonnet found that the hairs of the caterpillar of the great goat-moth (Cossus ligniperda) were hollow, though not to the apex: probably this is the case with those of other larvæ, as well as with their spines. In this instance they were set, he observes, in a corneous ring, or very short cylinder, elevated a little above the skin. The hair passes through this ring, and appears to be rooted in a soft integument, which clothes the skin within, and upon which the nerves form a reticular tissue,some of which he thinks he has even seen enter the root of the hairs, which perhaps are organs of touch[416].
Of the pilose larvæ, some, like most of those of the smaller moths (Geometra,Tortrix,Pyralis, &c.), have merely a few scattered short hairs, scarcely perceptible except through a lens: others (Odenesis potatoria,Lasiocampa Rubi) are covered with down more or less thick: in others (Eriogaster lanestris,Lasiocampa Neustria) the hair is slenderer, and more like wool; the body of two species which I purchased from the collection of Mr. Francillon is covered with woolly hairs, so long as to give them the appearance of a shock-dog; and Madam Merian has figured a similar one, which she could not bring to the perfect state[417]. The hairs of manyBombycidæ, known commonly by the name ofhairy caterpillars, asArctia erminea, &c. are stiffer, resembling bristles; sometimes, as inArctia chrysorhea, mixed with shorter ones. The hairs either spring immediately from the skin (Noctua Aceris,leporina), or, as is more general, proceed only from certain tubercular elevations, usually subhemispherical, but sometimes conical; of which a number varying from four to twelve are found on each segment of different species. They seem to issue from these tubercles, as little diverging streams from the rose of a watering-pot. In both cases, they form a coating usually so dense as to conceal the body, but sometimes more thinly set, and admitting the skin to be seen more or less between them. In a caterpillar of the beautifulArctia ocularia, the hairs are set upon tubercles alternately nearer the anterior and posterior margin of each segment, so as to form a denseband, the rest being naked; and in the lovely green and black one ofSaturnia Pavonia, each tubercle bears but six hairs, diverging like a star, the central one being the longest and capitate, so that the chief part of the body appears naked. This diverging position of the hairs is most common in the thick-clothed larvæ also, but many have them placed differently: thus, in those ofCallimorpha CajaandArctia villica[418]they are all directed towards the tail, like the quills of a porcupine: in some others the anterior ones point towards the head: in that ofEriogaster Quercushalf of the tuft of hairs of each tubercle is directed downwards, the other half upwards: in that ofArctia Salicisall the hairs point downwards, so that the belly is thickly covered, while the back is bare. Another variation is, that the hairs of half the tubercle are sometimes very long, while those of the other half are very short, and even of a different colour[419]. In the larva of Tussuck moths (Laria pudibunda,fascelina, &c.) the hairs are collected into tufts of a singular appearance, those on the intermediate segments of the back being quite level at the top, so as to resemble so many brushes; while those on the first and last segments are longer, and composed of feathered hairs converging to a point at their extremity, like a common camel-hair-pencil[420]. This last mode of arrangement prevails also in the larva ofNoctua Aceris; but in this the pencils are shorter, exactly wedge-shaped, and distinguished by another particularity, that of springing directly from theskin, and not from a tubercle. This is also the case with the large caterpillars ofOdenesis potatoria, which has a double row of short bundles of black hairs on the back, intermixed with larger ones: at each end of the body is a pencil of converging hairs, and the sides are spotted with bundles of white ones, which with longer tawny ones are bent downwards, so as to cover the sides of the creature[421]. Some have the anterior aigrettes disposed like the arms of a cross, of which the body of the caterpillar is the stem[422]. But not only is there considerable variety in the general arrangement of the hairs that clothe our little larvæ, the hairs themselves differ much in their kind and structure, of which I will now, before I proceed to consider spines, give you some account. Several of them are feathered like the plumes of a bird: this is the case with those ofMorpho Idomeneus, on each segment of the body of which are three blue tubercles, like so many little turquois beads, from each of which proceeds a long black plume[423]. Other hairs terminate in a club; those of the larva ofNoctua Alni, a specimen of which I possess taken in England, are flat and incrassated at the apex, something like the antennæ of someSphingidæ. Mad. Merian has figured the caterpillar of another moth which feeds upon the Papaw-tree (Carica Papaya) with similar hairs[424]. But the most remarkable larva for the shape of its hairs is that ofAnthrenus Musæorum, the little pest of our cabinets,which I noticed in a former letter[425]. All the hairs of its body are rough with minute points; but those of six diverging long tufts or aigrettes, laid obliquely on the anal extremity of the body, which the animal when alarmed erects as a porcupine does its quills, are of a most singular structure: every hair is composed of a series of little conical pieces, placed end to end, the point of which is directed towards the origin of each hair, which is terminated at the other extremity by a long and large conical mass, resembling somewhat the head of a pike[426].
Besides the one lately mentioned, other caterpillars are rendered striking by the brilliant colour of the tubercles from which their hairs emerge. A remarkable instance of this is the thick large caterpillar of a Bombyx, which feeds upon thePsidium pyriferum, or white Guava, figured by Madame Merian. This caterpillar, which is white, with transverse black stripes, and which has two singular long converging curved bunches of hairs near the tail, is splendidly adorned on each side with fifty red tubercles, shining like coral, from which proceed six or seven long diverging hairs. Leeuwenhoeck took these tubercles for eyes[427]. Another figured by the same lady, who mistakes it, with her usual inaccuracy, for the larva of aLygæusF., and which seems by her description to be between theonisciformandlimaciformtypes, has the apparently fleshy mamillæ that project from its sides and back crowned with little hairy red globes, which give the animal a most singular and unique appearance[428]. Having thus described some of the principal modes in whichthe All-wise Creator has decked and defended these creatures withhairs, I shall next give you a short account of thespineswith which he has armed others. The spinous larvæ are principallylepidopterous, and more particularly conspicuous in some tribes of the genusPapilioL., though some saw-flies andDipteraare also distinguished by them.Vanessa Io[429],AtalantaandUrtica,Argynnis Paphia,Urania Leilus, and many other Butterflies, &c. are clothed with long sharp points, which claim the denomination of spines, rather than that of hairs or bristles; being horny and hard, and so stiff at the point as readily to pierce the skin. Those of the last-mentioned species, Madame Merian says, are as stiff as iron-wire[430]. They are sometimes entirelysimple, and look like spikes rather than spines, as in the caterpillar ofNymphalis AmphinomeandMorpho Menelaus[431]; but ordinarily they are beset withhairs, or more commonly with shorter spines, which often give them the appearance of plumes, as inUrania Leilusjust mentioned: sometimes these lateral spines are so long as to have the appearance of a branch of a tree; this is strikingly the case with a small caterpillar which Captain Hancock brought from Brazil; its body is so thickly planted with spines of this description, that it absolutely wears the appearance of a forest or thicket in miniature. A singular circumstance attends the spines of this species: in many cases a smaller and very slender hair-like spine issues from them, resembling a sting; and this accounts for an observation of Abbott's, that many American caterpillarsstinglike a nettle, raising little white blisters on the skinwhen accidentally or slightly touched[432]. Lewin has described the caterpillar of a moth found in New Holland, which he namesBombyx vulnerans, that, like these Americans, has also the power of wounding, but in a different way. It darts out, he says, when alarmed by the approach of any thing, from as many knobs or protuberances in its back eight bunches of little stings, with which it inflicts a very painful and venomous wound[433]. The caterpillar ofPapilio ProtesilausF., if Madame Merian's account and figure of it are correct, has its body armed with hairy spines, the extreme point of which is surmounted by a star-shaped appendage[434]. Those of a few saw-flies (Tenthredo PruniL.), and another figured by Reaumur[435], are covered with a little forest of spines without lateral branches, but divided into a fork at the apex. Some spines are merely rough, with very short points, as those round the head, which give so terrific an appearance to the caterpillar of theBombyx regalis, of some proceedings of which I gave you an account in one of my former letters[436].
I must now say something upon thearrangementof these spines. Though in a few instances so thickly set as entirely to conceal the body of the animal, as in the case of the Brazil one lately mentioned, yet generally speaking, even when they are most numerous, they permit the skin to be distinctly seen. Their arrangement is various, though always orderly: in the majority theyare planted singly, but in some caterpillars in bundles. In that ofSaturnia Io, on each segment there are six bundles of longish, quill-shaped, sharp, slender, diverging spines, which also appear to sheath aculei. Madame Merian has figured this larva, or one very near it, as the grub of aEuglossa[437], with which, though she affirms she traced it to the fly, it can have no connection. With regard tonumber, some larvæ have only four spines on each segment; others five, others again six, and others seven, or even eight: they are planted on the sides and back only, never on the belly. They are often more numerous on the intermediate than on the anterior and posterior segments; but sometimes the reverse of this takes place; in that ofAttacus Erythrinæonly the head and tail are armed with spines, the rest of the body being without any[438]; and in that ofMorpho Teucerthere is only a single spine on the four intermediate segments[439]. They are usually all nearly of equal length; but in some cases those of the head and tail are much longer than the rest, and remarkably so in the caterpillar ofUrania Leilus, also beautifully plumose, and gracefully waved[440]. Those in the second and third segments are much longer than any of the rest in that ofBombyx regalis; which circumstance gives it the terrific appearance lately alluded to. In the family to whichArgynnis Paphiabelongs, the larva is adorned with two on the back of the first segment twice as long as the rest, and resembling at first sight two antennæ.
The spines, as well as the hairs of the new skin, are concealed under the old one, and not incased in its spines;but Bonnet ascertained, that if cut off very closely, the larva sometimes died in consequence, whilst no such result followed a similar operation on hairy larvæ. We learn from Reaumur[441], that some spinous larvæ of saw-flies (TenthredoL.) lose their spines at the last change of their skin; and from Madame Merian, that that ofAttacus Erythrinæbefore mentioned loses also at the same period the six tremendous black spikes that arm its black and yellow larvæ. The grubs of ants that are destined to pass the winter in the larva state are hairy, but are not so in summer[442]. The spines found in the grubs of some gad-flies (ŒstrusL.) are of a different kind from those above described, being very minute triangular flat plates, arranged in different and contrary directions[443], and serving the insect merely to change its place and fix itself[444].
Two other kinds of clothing, if so they may be called, neither coming under the description ofhairsnorspines, are found in some other larvæ, not only amongst theLepidoptera, but also in some of the other orders.Nymphalis Populiand others of the same family have larvæ furnished on the back of each segment with cylindricoconical processes of a fleshy substance, obtuse at the apex and surrounded with capitate hairs. In that ofN. Sybilla, which has on each segment two fleshy protuberances, they are bifurcate or trifurcate, and also encircled at the base with a hairy tuft[445]. Others, as those ofMelitæa Artemis,Cynthia, &c. have each segment beset on the back with from seven to nine fleshy, pubescent, wedge-shaped protuberances; two larger ones projectingover the head. Under this head, too, may be noticed, the glutinous secretion which clothes the grub ofCionus Scrophulariæ, a little weevil; and ofTenthredo CerasiL. a saw-fly, and that waxy or powdery substance which transpires through the skin of the larvæ of severalAphides,Chermes,Cocci,Hylotoma ovataF., &c. TheAphis, whose extensive ravages of our apple-trees (A. lanata) were before described to you[446], is covered and quite concealed by this kind of substance, so that the crevices in the bark which they inhabit look as if they were filled, not with animals, but with cotton. The insect, also, that forms those curious galls produced upon the spruce fir, and which imitate its cones (Chermes AbietisL.,AphisDe Geer) secretes a similar substance. In these and other cases of the same kind, this matter seems to be, if I may so speak, wire-drawn through numerous pores in certain oval plates in the skin, more depressed than the rest of the back, arranged regularly upon the segments, and exhibiting minute tuberosities. When young, these animals have more of this secretion than when more advanced: it then hangs from their anal extremity in locks[447].
But the insects most remarkable for a covering of this nature are thoseCoccidæof which Bosc has made a genus under the name ofDorthesia. De Geer is the first author that notices them, and has given a description and figure of one species under the name ofCoccus floccosus[448]. It was discovered by Modeer upon some sere fir-leaves in a thick bed of moss. Panzer has figured a second found uponGeranium sanguineum, which from the figure appears distinct from De Geer's, under the name ofCoccus dubius[449]. Fabricius regards this as synonymous with theDorthesia characiasof Bosc, inhabitingEuphorbia characiasin South Europe[450]. Olivier found a species upon the bramble[451]. I once took one, which appears to differ in some respects from the preceding species, uponMelampyrum cristatum, and our indefatigable friend Mr. Sheppard has sent me another, on what plant found I do not remember, which does not agree with any that I have mentioned. The body of the animals of this genus is covered by a number of cottony or waxy laminæ which partly cover each other, and are arranged usually in a triple series: in De Geer's figure the series appears quadruple, the lateral ones being placed obliquely. The anterior one in my specimen covered the head, and they are all canaliculate. Above the anus are four diverging ones: the whole are of the most dazzling whiteness. When these laminæ are removed, the body appears divided into segments.
With respect to those larvæ which imitate slugs by the viscid covering that besmears them and issues from their pores, we learn from Professor Peck that this exudation takes place as soon as they are hatched; that the animal retains its humidity although exposed to the fiercest heat of the sun, and that at the last moult the skin becomesquite clean, and free from all viscidity[452]. It is probable that the other limaciform larvæ are similarly circumstanced. Madame Merian has figured anonisciformone, the legs of which, she says, are covered with a viscid skin: this produced aNoctua. Those ofPapilio Anchisesalso are slimy, and adhere to each other[453].
v. Amongst other qualities which attach to larvæ, we must not omit to say something concerning theirColour. For though those which live in darkness, in the earth, in wood, in fruits, &c. are, with few exceptions[454], of an uniform whitish colour, yet such as are exposed to the influence of the light are usually adorned with a vast variety of tints, sometimes the most vivid that can be imagined. That the white colour of the former may be attributed to the absence of light is proved by an experiment of M. Dorthes, who having forced some to live under glasses, exposed to the light, found that they gradually became brown[455]. To attempt any classification of coloured larvæ would be in vain, since they are tinged with almost every possible shade that can be conceived, of many of which it would be difficult to find examples elsewhere; and infinitely diversified as to the arrangement and figure of their multiform markings and spots. A few general remarks, therefore, are all that you will expect on this head. Many are of one uniform colour; while a variety of tints, very different, and very vivid and distinct, ornament others. Sometimes they are distributed inlongitudinal rays or bands, at others in transverse ones. Sometimes they are waved or spotted, regularly or irregularly; at others they are sprinkled in dots, or minute streaks, in every possible way. Various larvæ are of the colour of the plant on which they feed, whence they are with difficulty discovered by their enemies. Thus, a large proportion ofLepidopteraare green of different shades, sometimes beautifully contrasted with black bands; a circumstance which renders the caterpillars of two of our finest insects of this order as lovely as the fly: I mean that ofPapilio MachaonandSaturnia Pavonia. Very frequently the larvæ of quite different species resemble each other so exactly, in colour as well as shape, as scarcely to be distinguishable: this sometimes takes place even where they belong to different genera, as in those ofBombyx versicolora moth, andSmerinthus Populia hawk-moth. And it sometimes happens, very fortunately for distinguishing allied species, that where the perfect insects very nearly resemble each other, the larvæ are altogether dissimilar. Thus, the female ofPieris Rapæis so much like the same sex ofPieris Brassicæ, that it might be taken for a variety of it, did not the green caterpillar of the one, and the spotted one of the other, evince the complete distinction of these butterflies.Noctua Lactuca,N. umbratica, and several other species of the same tribe, which includesN. Absinthii,Verbasci,Chamomillæ,Abrotani, are so extremely alike, that the most practised eye can scarcely discover a shade of difference between them, though their larvæ in colour and markings are constantly distinct[456]. The markings ofspecies belonging to the same family are usually different; but in some cases the latter may be prejudged from the former. The larvæ of many of the genusSphinxL., for example, have their sides marked by oblique streaks running from the back in a direction towards the head; and by this last circumstance they are distinguished from those ofBombyx versicolor,Attacus Tau, and others of the same tribe, which have also lateral oblique striæ, but running from the back towards the tail[457]. The colours of individual larvæ of the same species are usually alike, but inSphinx Elpenorand some others they vary exceedingly. Many, like those ofLasiocampa Rubi,Saturnia minor, &c., are of one colour when first disclosed, and assume others quite different in riper age. Just previously tochanging their skin, the tints of most larvæ become as dull and obscure, as they are fresh and vivid when the change has fully taken place; and in some instances the new skin is quite differently marked from the old one. This is remarkably the case with the last skin of some of the larvæ of the genusTenthredoL., which is entirely different from all the preceding ones. As people when they advance far in years usually become more simple in their dress than when they were young, so the larvæ in question change an agreeably variegated skin for one of a uniform and less brilliant colour[458]. Madame Merian has observed with respect toAttacus Erythrinæ, that its caterpillar is at first yellowish, with nine black striæ on each side: when arrived at one third of its size, they become orange; the striæ are obliterated, and in their place a round black spot appears on each of theeight intermediate segments[459]. Mr. Sheppard has remarked to me, that the skin of that ofSphinx Ligustri, after being under ground four days, was changed from a vivid green to a dull red. Very rarely, however, it becomes of a more brilliant hue just before entering the pupa state: thus, that of another hawk-moth (Smerinthus Tiliæ) changes to a bright violet; and the yellow hairs of that ofLaria pudibundathen become of a lovely rose colour. And here I may observe, that the hairs and spines also, of larvæ, vary greatly in colour. They are to be met with brown, black, red, yellow, violet, white, &c. De Geer found, that in the larva ofCimbex nitensthe two sides of the body were of a different colour, the left being of a deep green, whilst the right side and the rest of the body were paler[460]; but as he saw only a single individual, this was probably an accidental circumstance. Though the caterpillars, as I lately said, of one of the most beautiful butterflies and moths that inhabit Britain contend with the perfect insect in loveliness, yet in general no judgement can be formed of the beauty of the future fly from the colour of the larva; and the young Aurelian must not flatter himself always with the hope, because the caterpillar excites admiration by its colours and their arrangement, that the butterfly or moth it is to produce will do the same; nor ought he to despise and overlook a sombre or plain-coloured individual of the former, under the idea that it will produce one equally plain of the latter, for it often happens that the splendid caterpillar gives a plain butterfly or moth, andvice versâ. De Geer, however, gives us two instances of conformitybetween the colours of the caterpillar and those of the future moth; the one is that of the common currant-moth (Phalæna G. grossulariataL.), the caterpillar of which is white, ornamented with several black spots varying in size. At the two extremities it is yellowish, with a longitudinal ray of the same colour on each side, the head and legs being black. These colours are all to be found in the fly, the ground of its wings being white ornamented with many black spots of different sizes. Its upper wings are traversed by a yellowish band; and towards their base is a spot of the same colour. Its body is yellowish, with black spots; but the head and legs are black[461]. The other is that of a green caterpillar, which gives a green moth, figured by Reaumur (Pyralis prasinariaFab.)[462]. Sometimes, also, the sex of the future perfect insect may be predicted from the colour it exhibits in its first state: thus, the brown caterpillars ofNoctua Pronubaproduce males, and the green ones females[463]. The sexes, also, ofN. exoletaandPersicariædiffer in that state.
vi. To the full account of theFoodof insects given in a former letter[464], which had reference chiefly to their larva state, it is only necessary in this place to add a few particulars not there noticed. Many larvæ when first excluded, as those ofPieris Cratægi, &c. devour the shells of the eggs from which they have proceeded[465]; andothers (Cerura Vinula,Sphinx Euphorbiæ,Noctua Verbasci), though their usual food is of a vegetable nature, eat with great apparent satisfaction the skins which they cast from time to time, not leaving even the horny legs. This strange repast seems even a stimulating dainty, which speedily restores them to vigour, after the painful operation by which they are supplied with it. Under this head it will not be out of place to mention, that some larvæ of insects, which feed only on the juices of animals, or the nectar and ambrosia of flowers, have no anal passage, and of course no feces. This is said to be the case with the grubs of bees, wasps, the larvæ ofMyrmeleon, &c.[466]
vii. You will require no stimulus to induce you to attend to the subject I am next going to enter upon,—theMoulting, namely, of Larvæ; or their changes of skin. This, indeed, is a subject so replete with interest, and which so fully displays the power, wisdom, and goodness of the Creator, affording at the same time such large occasion for nice investigation, that a pious and inquisitive mind like yours cannot but be taken with it. In the higher orders of animals, though thehairsof quadrupeds and thefeathersof birds are in many cases annually renewed, the change, or scaling and increment of theskin, is gradual and imperceptible; no simultaneous rejectionof it, in which it is stripped off by the animal itself like a worn shirt, being observable, till you descend in the scale to the Serpent tribe[467], which at certain periods disengage themselves from their old integument, and start forth with that new and deadly beauty so finely described by the Mantuan bard:—