The respiratory anal appendages of many Dipterous larvæ will be fully described in a subsequent Letter: I shall therefore now only further observe upon this subject, that although there is seldom any alteration in the form of these appendages &c. in the same species, the caterpillars of two moths (Cerura Vinula and Attacus Tau), however, are exceptions. The former, when young, has two hairy projecting ear-like protuberances, which it entirely loses, as I have myself observed, before it assumes the pupa; and the latter, in like manner, after its third change of skin, is deprived of its bent thorn-like points which attend it when young[368]. It is remarkable that these last larvæ, when just excluded from the egg, are also entirely destitute of these appendages; they soon, however, appear, from slight elevations which mark their situation, and rapidly acquire their usual form[369]. Changes of a similar kind, hitherto unobserved, may probably take place in other species.
iii. Figure. I am next to consider the general figure or shape of larvæ. All of them, with but few exceptions[370], agree in having a body more or less constricted at intervals into a series of rings or segments; usually in number, twelve; often nearly equal in length, but sometimes in this respect very dissimilar[371]. The general outline or shape of the body is extremely various: most frequently it approaches to cylindrical, as in most of the caterpillars of Lepidoptera, and of the Hymenopterous tribe of saw-flies (Tenthredo L.). The next most common figure is that more or less oblong or oval one, sometimes approaching to conical, found in many of the larvæ usually called grubs; such as those of the weevil (Curculio L.) and of the capricorn (Cerambyx L.), and other coleopterous tribes; of bees, and all Hymenopterous insects but the saw-flies; and also of a large number of flies (Diptera). In some the figure approaches to fusiform, as in most of the moths of the Fabrician genus Lithosia. In others, as in those of the water-beetles (Dytiscus, &c.), it approaches to an obovate shape, being widest towards the head, and terminating in a point at the anus. In others, again, it is linear; an example of which is that of Staphylinus. Some are convex, and others gibbous, above, and flat underneath; as those of Silpha, Chrysomela and many other beetles. Others are flat, both above and below, and depressed like a leaf; a remarkable instance of which has been before noticed[372]. Some are very long, as those of most Lepidoptera; others very short, as that of the ant-lion (Myrmeleon). Many other peculiarities of form in individuals might be instanced; but a dry enumeration of these would be of no great use to you. They can only be advantageously learned by the study of good figures, and by watching the actual metamorphosis of the singularly-formed larvæ that you meet with.
Instead, therefore, of any further specification of individual forms, I shall now endeavour to give you, as far as my own knowledge of them and the information I can collect from other sources will enable me, a larger and more general view of the kinds of larvæ; for analytical inquiries lose half their value and importance unless we proceed to apply them synthetically, by forming, if possible, into groups the objects with which we are individually acquainted.
Partial attempts at a synthetical arrangement with regard to the larvæ of Lepidoptera and the saw-flies (Tenthredo L.) have been made both by Reaumur and De Geer. M. Latreille also has recently given a Tableau méthodique et général of articulated animals furnished with jointed legs, considered in their first state[373]. The former of these is chiefly founded upon the number of the prolegs, and the latter upon the metamorphosis, prolegs, habits, head, and parts of the mouth, without any other notice of the configuration. Mr. Wm. MacLeay, who, though young in years, is old in science and critical acumen, has started a perfectly new hypothesis upon this subject. In the progress of his inquiries into the natural arrangement of animals, particularly of insects in the Linnean sense, he has been the first to observe, that the relation which organized objects bear to each other is of two kinds; one of real affinity, and the other only of analogy, or resemblance. This important distinction, upon which I shall enlarge in a future Letter, when I come to treat of Systems of Entomology, he has applied, in a way quite original, to larvæ in general, but more particularly to those of the Coleoptera order. For the basis of his system he assumes a relation of analogy between the larvæ of Insects that in the progress of their metamorphosis assume wings, and those that do not, which form his class Ametabola, so that the prototypes of the former shall be found amongst the latter[374]. But though Mr. MacLeay appears to consider the analogy between these two as primary, he extends it in a secondary sense to the Crustacea, at least in several instances[375]. Upon this occasion he very judiciously remarks, that "in terming larvæ Chilognathiformes or Chilopodiformes, it is not meant that they are Scolopendræ or Iuli, or even near to them in affinity; but only that they are so constructed that certain analogical circumstances attending them strongly remind us of these Ametabola[376]." This remark you will bear in mind while I am treating of this subject. It should seem from another part of the same paragraph, that the comparison which our learned Physiologist recommends, is between the young of the Ametabola and the larvæ of the corresponding groups of Coleoptera. This must be understood to refer chiefly to the young of the Chilopoda and Chilognatha, since they approach nearer to them in that state, having then only six legs; but the rest of the Ametabola should certainly be brought to this comparison in their adult state: and even the two former orders in that state more strongly resemble numerous coleopterous larvæ, than when they are young and much shorter. I before called your attention to the remarkable circumstance that contrasts very many larvæ of Hexapod insects that become winged in their perfect state with adult Myriapoda: namely, that in one the progress to this state is by losing their prolegs and shortening their body; while in the other, the reverse of this takes place, numerous prolegs and additional segments being gained before they arrive at maturity[377]. As the multiplication of organs is a sign of imperfection, it may be affirmed of the former of these tribes, that their progress is towards greater perfection; while that of the other may be called a degradation. As larvæ may be regarded as a stepping-stone by which approach is made from the apterous to the winged tribes of Insects, it seems most consistent with general analogy that each should connect with the other in that state in which the resemblance is greatest. Now the Myriapoda resemble larvæ, as we have just seen, most when in their adult state; therefore the comparison should be between larvæ and adult Myriapoda.
Mr. MacLeay divides coleopterous larvæ into five tribes thus characterized:—
1. A carnivorous hexapod larva, with an elongate linear flattened body, having a large head armed with two sharp falciform mandibles, and furnished with six granular eyes on each side. This kind he denominates Chilopodiform, as having for its type in the Ametabola, Scolopendra L. The examples he gives are Carabus and Dytiscus.
2. A herbivorous hexapod larva, with a long and almost cylindrical body, so fashioned that the posterior extremity being curved under the breast, the animal when at rest necessarily lies like an Iulus on its side. This tribe he denominates Chilognathiform, from Iulus L. His examples are, the larvæ of Petalocerous insects, as Scarabæus L., Lucanus L. &c.
3. An apod larva, having scarcely the rudiments of antennæ, but which is furnished instead of feet with fat fleshy tubercles; which, when continued along the back and belly, give the animal a facility of moving in whatever way it may be placed. These he denominates Vermiform, from certain of the Vermes intestina and Mollusca of Linné which he has associated with his Annulosa[378]. His examples are, Curculio L. and Cerambyx L.
4. A hexapod and distinctly antenniferous larva, with a subovate rather conical body, of which the second segment is longer and of a different form from the others, so as to give the appearance of a thorax. His denomination for these is Anopluriform, from Pediculus L., forming Dr. Leach's Anoplura. His examples are, Coccinella and Chrysomela L.
5. A hexapod antenniferous larva of an oblong form, having like the former vestiges of a thorax, besides two or more articulated or inarticulated setaceous or corneous appendages to the last segment of the abdomen. This tribe he calls Thysanuriform, from Lepisma and Podura L., forming M. Latreille's order Thysanura. His example is Meloe with a note of interrogation[379].