THE LEPIDOPTERISTS' NEWS

Transcription

1 THE LEPIDOPTERISTS' NEWS Volume Numbers 1-3 THE STRUCTURE OF THE LARVAL PROLEGS OF T HE LEPIDOPTERA AND THEIR VALUE IN THE CLASSIFICATION OF THE MAJOR GROUPS by H. E. HINTON ~:~ :\;.'~J '.. '" \.; In 1946 I proposed a new subordinal classification of the Lepidoptera. This classification differed from that of Borner (1939) in two.lmportant particulars: ( 1) the Micropterygidae were placed in a separate order, the Zeugloptera, as first suggested by Chapman (1917), and (2) the suborder Dacnonypha was erected to contain the Eriocraniidae and related families with a decticous pupa (Hinton, 1946a). Each year since then a number of classifications of the Lepidoptera have appeared. Most of these, it must be admitted, are new arrangements produced by reshuffling already known facts. The classification of Kiriakoff (1948), however, deserves especial attention, as he has discovered a number of new facts about the structures of the tympanum. In the not too distant future I hope to reply in detail to the various critics of my classification, particularly as regards the position of the Micropterygidae. In the space now at my disposal I can do no more than reply rather briefly to those who believe that I paid insufficient attention to the structure of the larval prolegs. For instance, Kiriakoff (1948, p. 133) says of the structure of the larval prolegs, "Borner 1939 has rejected it as of no phylogenetic value; nor is it mentioned in Hinton's provisional scheme." The great majority of the Lepidoptera are placed by Kiriakoff in two groups, the Stemmatoncopoda and the Harmoncopoda. To the first group belong all these species that have a complete or nearly complete circle of crochets on the ventral prolegs and to the second those that have a single longitudinal row of crochets. Thus Kiriakoff's division of the majority of the Lepidoptera does not differ from the older classifications of Karsch, Heymons, and others. It is interesting to note that in the recent classification by Bourgogne.,,('1951) no importance is attached to the structure of the prolegs, although Bourgogne still accepts Tilly~rd's two suborders, the Homoneura and the Heteroneura, a division of the Lepidoptera considered to be unnatural by both Borner (1939) and myself (1946). If it be admitted that the primary aim of classification be to show, so far as is,possible, the natural or genetic relations of the groups within the order, it fonows as a necessary corollary that each group shall be monophyletic in terms of any other. If, therefore, it can be shown that the prolegs of the harmonocopodous type have been independently evolved by two or more families, the basis for the retention of the Stemmatocopoda and Harmonocopoda will have been destroyed, since it will be then apparent that the Harmonocopoda merely include those families in which the structure of the larval prolegs happens to. I

2 Plate I HINTON: LARVAL PROLEGS Lep. News, Vol Figs. 1-2 Figs. 3-4 Adela sp. (1) Right side of fourth abdominal segment of full grown larva. (21 Posterior view of left proleg of sixth segment. (mv) Level of mid-ventral line. (s) Level of spiracle. The slender diagonal muscle shown is the occlusor of the spiracle. Polia nebulosa Hufn. (3) Cross section through the fifth abdominal segment to show proleg, which is considerably retracted. ( 4) Inner side of left proleg of fifth abdominal segment as seen when procracted.

3 1952 The Lepidopterists' N ews 3 converge, and, moreover, will contain families more related to some of those included in the StemmatOcopoda than to others included in the Harmonocopoda. For instance, in Kiriakoff's system the Hesperiidae are placed in the Stemmato copoda but the remaining families of the Papilionoidea are included 111 the Harmonocopoda. The morphology of the larval prolegs bas received very little attention from a comparative point of view, although the external form of a considerable range of species is described by Goossens (1887). Schultze (1920) has described in some detail the variations in the number and form of the crochets of several species. The many other studies that have been made are restricted ro single genera or families; and apart from isolated accounts of single species and a few very general studies, such as that of Snodgras:; (1935), nothing is known of their internal structure. The ventral prolegs (3-6) are hollow cylindrical outgrowths of the body wall. In most Microlepidoptera (Monotrysia and Ditrysia), and in the early instars of all and the final instar of some Papilionoidea, they have a complete or nearly complete apical circle of strongly sclerotised curved hooks known as crochets. The apical area bounded by and bearing the crochets is membranous and much less rigid than the sides of the pro legs. It is known as tbe planta. The retracror muscles are inserted in the planta, usually in its centre. They consist of a variable number of separate groups of fibres. In most of the families that were examined, they consist of two distinct sets, one arising well below and the other above the spiracle, as in the Cossidae. In the HepiaJidae (H epialu.r) and Yponomeutidae (Y ponomeuta), however, all of the planta retractors arise above the level of the spiracle. Contraction of the retractor muscles pulls the centre of the planta inwards, and the whole of the planta may be more or less completely invaginated within the proleg. The crochets on the periphery of the planta have their apices direered outwards and upwards. They are attached to the cuticle of the planta for about one-half to three-fourths of their lengths in such a way that when the planta is pulled inwards against turgor pressure they are so tilted that their apices, instead of projecting upwards, now project ventrally and are more or less parallel to the long axis of the proleg. Thus by retraction of the planta the crochets are disengaged. They are engaged again when the retractors are relaxed and turgor pressure evaginates the planta. The caterpillar therefore always pulls the planta inwards before shifting the position of the proleg. Ripley (1923) claims that in the Phalaenidae muscles are inserted in the proximal ends of the crochets, but this is not so either in that or related families. Species belonging to all of the principal superfamilies of the order have been examined, and in no case were muscles inserted in tbe crochets themselves. Movements of the prolegs, other than retraction or evagination, are effected by variations of turgor pressure and by the muscles inserted in the body wall near their bases. The areas of the body wall adjoining those that give rise to the prolegs have numerous transverse, oblique, and longitudinal muscles as shown, for instance, in fig. l. It is probable that the retractors of the planta are, as in the Diptera, merely slightly specialized muscles of the body wall. Among the most simple ventral prolegs are those of the Adelidae (figs. 1 and 2). Whether this simplicity is brought about by reduction from a more complex proleg or whether it is

4 4 HINTON: Larval Prolegs Vol. 6, 00s.1-3 primitive will be discussed elsewhere. The prolegs of the Adelidae differ little in structure from the transverse creeping welts of the Diptera, and it seems certain that they resemble a stage in the evolution of prolegs from structures like creeping-welts even though they may not actually be one. The ventral pro legs of the Zygaenidae, many Papilionoidea, Geometridae, Phalaenoidea, Bombycoidea, Saturnioidea, and Sphingidae are modified in a very characteristic fashion that enables them to cling to thin twigs with a force not possible to larvae of comparable size that have prolegs of the primitive type. When the larva is at rest, the twig is clasped between the prolegs in such a way that, if narrow enough, the prolegs may completely embrace it. At first sight it appears that this type of proleg is especially adapted for climbing about on plants. This impression receives support from the fact that it is the type found in all of the exclusively arboreal caterpillars that move about freely exposed and do not live in shelters of some kind as do most of the Microlepidoptera. Furthermore, the only terrestrial or semi-terrestrial larvae that have this kind of proleg are species of Phalaenoidea that can be shown in each case to be secondarily terrestrial. Each of these specialised prolegs is somewhat tilted mesally towards the other of the same pair. The crochets, which do not differ in structure from those of the primitive type of proleg, are restricted to a straight or sinuate longitudinal line on the mesal side of the planta only. This arrangement of crochets is known as a mesoseries, and each crochet has its apex curved mesally and upwards. The planta is retracted and everted as in the Microlepidoptera. The retractor muscles often also consist of two sets, one arising on the body wall directly above the proleg dorsal to the spiracle and the other ventral to the spiracle (figs. 3-4), as in the Zygaenidae, Lycaenidae, Phalaenidae, Arctiidae, and Sphingidae. The centre of the planta in which the retractor muscles are inserted is, of course, always laterad from the crochets. As in the primitive proleg, the crochets of the specialised type are disengaged when the planta is pulled inwards, but the action of the proleg as a whole is much more complex. A variable number of muscles are inserted on the base of the proleg itself (figs. 3-4), and in some groups, e. R. the Geometridae, muscles that arise near the spiracle are inserted on its middle outer face. Furthermore, there may be, as in the Geometridae, a specialised ventral musculature between the two prolegs of a pair. For instance, among the few Geometridae examined three obviously different patterns of ventral muscles were present, one of these including muscles that extended between the inner bases of the right and left prolegs. The planra of the prolegs may sometimes function as a sucker when the larva is crawling on a smooth hard surface. At such times, as noted by Snodgrass (1935), the crochets are turned upwards and the planta is pressed ~,gainst the surface so that its fleshy periphery inside the line of crochets seals a ventral hollow. In this position, a slight contraction of the planta retractors increases the volume of the central hollow without admitting air, and thus, by lowering the internal pressure, the central hollow functions as a vacuum cup. In some caterpillars, e.g. some Lycaenidae, there is a fleshy lobe between the centre of the planta and the mesal crochets. This lobe appears to function as an adhesive organ, and it is possible that a gland or glands may open on it.

5 1952 The Lepidopterists' News 5 When the prolegs of climbing forms are much reduced or modified to subserve other functions, their muscles may be reduced or lost. For instance, such Notodontidae as Stauropus fagi L., that have very long anal prolegs adapted for defence, now lack both the planta retractors and any other muscles they may once have had. The anal prolegs of these are moved entirely by muscles inserted on the sternal areas around their bases and by variations in turgor pressure. Other Notodontidae, such as Cerura vinula L., that have the anal prolegs adapted in another way for defence, still retain planta retractors. The way in which the primitive type of proleg present in most Microlepidoptera was altered to become a specialised climbing organ is clear from a study of larval development in almost any of the Papilionoidea except some Hesperiidae and Papilionidae. As I have previously shown (Hinton, 1946), the first instar of all Papilionoidea has a primitive type of proleg with a complete circle of crochets. In successive larval instars of most species the crochets on the ourer side of the planta become less numerous and smaller relative to those of the inner side at each moult until in the mature larva, they are lost altogether and only the mesal crochets are left to form a typical mesoseries. In the final instar of some Lycaenidae the process is not quite completed, and some of the outer crochets are still present. In the more specialised arboreal caterpillars, e.g. Sphingidae, Saturnioidea, Bombycoidea, and most Geomerridae and Phalaenoidea there is no ontogenetic sequence of this kind, and a mesoseries is already present in the first instar. Both the ventral and anal prolegs that are modified as climbing organs are, except for differences in their muscles other than plant a retractors, remarkably similar in structure and appearance notwithsranding the fact that they have been independently evolved many times from a proleg like that of recent Microlepidoptera. This type of proleg has been independently evolved in the Zygaenidae, Papilionoidea, Geometridae, Phalaenoidea, Sphingidae, and in the Bombycoidea-Saturnioidea. Furthermore, it seems certain that it has been independently evolved at least three times within the Papilionoidea and may have been evolved several times in the series Bombycoidea-Sarurnioidea. Tn short, it seems certain that a climbing type of proleg has been evolved no less than eight times within the Lepidoptera. That the specialised harmoncopodous proleg of the Zygaenidae is of indendent origin seems clear from the fact that other members of the superfamily have prolegs of the primitive type, e.g. the Megalopygidae and Epipyropidae. A connection between the Zygaenoidea and any other superfamily containing species with specialised prolegs must have been through forms that had the primitive type of proleg, unless it is to be supposed that the apparently primitive projegs of the Zygaenoidea are secondarily derived from the specialised type. A similar argument may be applied to the Geometroidea, in which superfamily the Geometridae have harmoncopodous prolegs and the Drepanidae stemmatoncopodous or primitive prolegs. In the Papilionoidea, the family Papilionidae contains species, e.g. the glaucus and troilus groups of Papilio, that retain a primitive type of proleg in the final instar. In other words, unless it be supposed that the primitive prolegs of these species of Papilio are derived from specialised prolegs, it must be supposed that the archetype of the family had a primitive type of proleg in all instars. A similar ugument may be applied to the Lycaenidae, in which family some species have the outer row of crochets more or less well preserved in the final ins tar. The

6 6 HINTON: Larval Prolegs Vol. 6, nos.1-3 other families and superfamilies listed in the preceeding paragraph are not related to one another but appear to be independently derived from stemmatoncopodous Ditrysia. The evidence for this is, as probably most lepidopterists will agree, especially strong in the cas ~ of the Sphingidae. The few facts already cited are quite sufficient to show that if Kiriakoff's division of the bulk of the Lepidoptera into the Stemmatoncopoda and Harmoncopoda be accepted, it would result in placing all the early instars of the Papilionoidea and the final instar of the Hesperiidae, some Lycaenidae, and the glaucus and troilus groups of the genus Papilio in the Stemmatoncopoda, whereas the last one or two instars of most Papilionoidea would be included in the Harmoncopoda. Similarly, a most unnatural division would be made of the Zygaenoidea and Geometroidea. References Borner, C, Die Grundlagen meines Lepidopterensystems. Verh. VII. Int. Kongr. Ent. 1938, vol. 2: pp , 51 figs. Bourgogne. ]., Ordre des Lepidopteres. In Grasse, P.-P., Traite de Zoologie, tome 10: pp Chapman, T. A., Micropteryx entitletl to ordinal rank. Order Zeugloptera. Trans. ent. Soc. Lond. A 1916 vol.: pp. 310-:3 14, 12 pis. G oossens, T., Les partes des chenilles. Ann. Soc. en!. France, vol. 7 ( ser. 6 ) : pp , 1 pi. Hinton, H. E., On the homology and nomenclature of the setae of lepidopterous larvae, with some notes on the phylogeny of the Lepidoptera. Trans. R. Ent. Soc. London, vol. 9 7: pp. 1-37, 24 figs., 1946a. A new classification of insect pupae. Proc. Zool. Soc. Lond., vol. 116: pp , 64 figs. Kiriakoff, S. G., A classification of the Lepidoptera and related groups with some remarks on taxonomy. Biol. Jaarb., vol. 15: pp Ripley, 1. B., The external morphology and postembryology of Noctuid larvae. Illinois bioi. Monogr., no. 8: 169 pp., 61 figs. Schultze, E. A., Beitrage zur Kenntnis der Pedes spurii der Lepidopterenlarven. Arch. Naturg., vol. 85: pp. 1-78, 8 pis. Snodgrass, R. E., Principles 0/ Insect Morphology. London. Department of Zoology, University of Bristol, Bristol, England DISCUSSION OF DR. HINTON'S PAPER Mr. KIRIAKOFF says: "Je regrette de n'avoir re<;u aucun texre ni resume de votre expose ce qui m'empeche d'y repondre sur place en detail. Je puis cependant attirer votre attention sur deux points: 1. Les noms Stemmatoncopoda et Harmoncopoda ne sont pas Ie produit de mon imagination, mais ont ere crees par KARSCH, Vous insistez sur Ie cote phylogenetique de la question alors que dans rna classification j'avais expressemem dit que cette derniere ne devait pas etre consideree com me un schema phylogenetiqne." Prof. M. HERING remarks: "Die Fragwordigkeit der Verwendung der iiusseren Bildung der Abdominalfi.isse verkleinert deren Bedeutung flir die Aufstellung def Ordnung Zeugloptera." Dr. W. HACKMAN says: "The inner anatomy of the larval ocelli of Eriocrania is very different from the same of other Lepidoptera sensu stricto, A comparison with the eye anatomy of the larva of Micropteryx (which has not yet been studied), could perhaps bring to light facts of systematic value." Dr. DIAKONOFF says that without going into a discussion of the value of prolegs for systematics, it must be borne in mind that they and especially their crochets can be strongly influenced by the biology of the larva which may render them not fit for use as systematic characters.